Table Of Contents

INTRODUCTION: What the hell do I

know?
Step 1: Realize You Know Nothing
Step 2: Learn the Basics
- Is sports betting legal?
- So… how do Americans bet on sports?
- What about taxes?

- What’s going on in New Jersey?

- Why aren’t sports leagues on board?
- What types of bets are there?
- So why can’t I win 50% of bets and break even?
- So what’s a good winning percentage?
- Is every bet -110?
- What happens when the point spread moves?
- What other types of bets are there?
- So what do the +/- numbers mean, exactly?
- How often must I win moneyline bets to break
even?
- I still think Seattle is a good bet. What do I do?
- So how the hell can I make money?
- What about other types of bets?
Over/Unders (totals)
Parlays and Teasers
Prop Bets
Team Totals
Season Win Totals
Futures Bets
Live Betting
Halftime Bets
Middling
- So where do I go to bet?
- Should I use more than one book?

STEP 3: Learn How Smart Bettors

Think
The 17 Commandments of Recreational
Sports Betting
1) Only bet what you can afford to lose.
2) Base your bet sizes on your expertise.
3) Seriously, TRACK EVERY BET YOU MAKE
4) Only make bets that you think have positive
expected value.
5) Make sure your reason for betting isn’t bullshit.
6) “Bet numbers, not teams.”
7) Ignore mainstream statistics.
8) Ignore win-loss records.
9) Place your bets (relatively) early.
10) Play Devil’s Advocate
11) Don’t bet parlays.
12) Don’t worry about short-term results.
13) Don’t bet on big games for the sake of it.
14) Buy low, sell high.
15) The point spread is telling you a story.
Comprehend it.
16) Take your recency bias vaccine.
17) Never buy picks from anyone.
MISCELLANEOUS:
- Do “trap games” exist?
- Does “Vegas” “know?”
- How close are point spreads, and does it matter?
- What the hell is a “sharp” bet or bettor?

Step 3: Learn NFL Betting Tips And

Tricks
In the NFL, all points are not created
equally.
Advanced Stats
- DVOA (Defense-adjusted Value Over Average)
- Yards Per Play (ypp)
- Yards Per Pass and Yards Per Carry
- Pace Stats
- Offensive Line Stats
- Red Zone Offense and Defense
- Third Down Offense and Defense
- Turnovers
- Read the paper
- Watch the games
- Drive Stats
- Other Stats
- What about outliers?
- Matchups
- Why can’t I look at individual player stats?
What matters beyond stats?
How much of NFL wins are luck?
-Football Outsiders Busts Myths
Step Four: Defeat Common Biases
The Cognitive Biases That Kill
- Recency Bias
- Gambler’s Fallacy (Law of Averages)
- Bandwagon Effect
- Confirmation Bias
- Framing Effect
- Hot-Hand Fallacy
- Insensitivity to Small Sample Sizes
- Sunk Cost Fallacy
- Ostrich Effect
- Overconfidence Effect (Optimism Bias)
- Loss Aversion
- Hindsight Effect
- The Bias Blind Spot
- The Narrative Fallacy
Final Notes
- Further Reading
- Twitter List To Follow
- Online Resources
INTRODUCTION: What the hell do I
know?
My lovely grandmother didn’t expect her $100 gift to turn into $7,896.32
from two years of sports betting.
And that lucky, cocky bastard (me) didn’t expect to lose all $7,896.32 of it
in two months.
Common estimates are that 95-99% of sports bettors lose money long term.
I’m not a professional sports bettor, but I’ve learned what it takes to grind
out profits and avoid catastrophe, as a result of my past idiocy. I’ve had
success since the disaster.
I wrote this book so you don’t end up like I did. It sucks to lose $7.896.32,
especially when you’re a jobless college junior. That’s 15,792.64 Natty
Light cans.
The book gives you every step you must take to leap ahead of the 90% of
bettors who don’t have a clue. If you learn all of the simple steps provided,
you will become… decent at sports betting. It will make sports betting a
decent investment for you at best, and a cheap hobby at worst.
I can’t make you a professional, but I can put you on the right path. If you
want to make money, you’ll first need to know everything in this book. And
I’ll point you towards the next step at the end.
The average fan makes the same mistakes over-and-over . They’re not easy
to fix, but they are simple. Becoming aware of your biases will save you
thousands in losses.
If you’re already rich betting on sports, this book isn’t for you.
But if you’re not, like 99.99999999% of the world, please come inside. But
take off your shoes; my grandmother just cleaned, and she probably hates
me now.

Step 1: Realize You Know Nothing

I’ve been a die-hard sports fan my entire life, and I worked as a sportswriter
for four years. I know where every third-string running back went to college
and I know how Roger Goodell likes his eggs (with a side of FUCK YOU,
FANS).
But I only started making money betting on sports once I realized that my
lifetime of knowledge was just making me cocky, and that I didn’t know
shit.
I don’t mean to be harsh; we’re all on the same boat. But if you want to
make money, you must give your Sports Ego away to charity.
Beating a market with a lot of money in it requires a lot more than liking
and knowing sports.
When you bet on sports, you’re betting against a market. This is a
crucial insight.
Here’s how a point spread is made.
Oddsmakers release a point spread that is just a mildly accurate guess. They
allow small bets only (called low limit bets), and see how much money is
bet on each side.
When they see respected bettors and big money on one side, they adjust the
point spread to make that side less attractive.
As more money comes in, they become more confident in their line (point
spread), so they raise the limits, allowing bigger bets.
So if the opening line sucked, they were alerted by strong betting activity
and adjusted after receiving minimal exposure. The pros were only allowed
to place small bets.
Oddsmakers don’t “know” the result. They don’t even try to predict the
future.
But they do know where the big money from rich professionals is. So: If
you bet on the NFL, you’re saying that a market shaped by many
millions of dollars is wrong.
I’m good at fantasy football, too. But I’m not smarter than millions of
dollars.
When you say “PFFFFFT, how could the Patriots lose this week?!” and bet
on the Patriots -7, it’s sort of like buying Apple stock at $98.63 today,
because you “know” it’s undervalued and you can sell it the next day for a
10% profit.
You might not think you’re flipping a coin, but you’re flipping a coin.
In the long run, ~50% of favorites and ~50% of underdogs cover the spread.
This means the market is generally efficient long term, and that the average
bet has a 50% chance of winning. But if you win 50% of your bets, you lose
money.
Say you bet four games every week in NFL, and six playoff games. 70
games in one season.
If you win 50% of your bets (at the standard -110 price, which I’ll soon
explain), you lose $700. Expensive hobby.
And many people lose more than 50% on average, because they make some
common mistakes that I’ll address.
For most games, you can’t beat the market.
So: Always assume you’re wrong, unless you have a really good reason
not to.
NFL betting is the hardest sport to profit on , because it’s the largest and
most efficient sports betting market.
Read this week’s NFL point spreads to a friend (if it’s during the season),
and listen to their reactions.
I can guarantee, for at least a couple of games, they’ll freak out and yell:
“WHAT, how is that the spread? Team X (Favorite or Recent Success) will
totally beat Team Y (Underdog or Recent Failure).”
So why isn’t your friend rich betting on sports? (Tell him to try it out; he’ll
be sad and poor like I was.)
You’ll hear the same thing on ESPN, but it’s even worse because these
people are often branded as “experts.”
My favorite life rule is: People who call themselves experts are almost
never experts.
And they’re usually trying to sell you something.
In the media’s case, “experts” are somewhat harmless, because they’re not
selling you directly; you’re just paying with your time and brain cells. I
tracked a couple of “experts” last year, and although one year doesn’t mean
much, the results were hilarious.

(Bill Simmons doesn’t call himself an expert and admits he’s bad at picking
games. But I included him because other people call him one anyway.)
If someone ever tries to sell you sports picks as an “expert,” there is a 99%
chance he or she is full of shit. I’ll discuss this later (NOTE: I will NEVER
sell sports picks).
You’re likely smarter than people who call themselves “experts,” because
you’re reading this book.
Before you move on, internalize: The market, guided by millions of
dollars, is smarter than me.
I’m sure you’re smart. But you’re not smarter than the wisdom of lots of
money.

The smartest opinion you can have on a game is: “I don’t know.” (“The
point spread seems fair so I’ll lay off.”)
Until you learn this, you will fail.
Wayne Gretzky famously said, “You miss 100% of the shots you don’t
take.” Michael Jordan had a longer quote along the same lines.
Jordan was a compulsive, losing gambler and Gretzky was involved in an
illegal gambling ring. Do not listen to these sports legends/gambling fools.
But now you get it. So let’s learn the basics.
This book is organized in order of difficulty. You might know some or much
of the information. If the beginning is too easy, move on.
Step 2: Learn the Basics
Is sports betting legal?
In the United States, it’s illegal to bet on sports offline, except in Nevada,
Delaware, Oregon and Montana.
But the options in Oregon and Montana are a joke, and Delaware isn’t much
better. It’s basically Las Vegas or bust.

So… how do Americans bet on sports?

It’s not 100% clear if online sports betting is illegal. It is 100% illegal to
take bets online (or offline), and you shouldn’t do it. You will get in trouble.
But it’s questionable if placing bets online is legal.
I’m not advocating testing an unclear law, but I will give you some facts.
- No American has ever been arrested for betting on sports
online.
- There is a reasonable argument that betting on sports online is
legal.

What about taxes?

If you make money betting on sports online, regardless of the questionable
legality of it, you do have to pay taxes on it, like any other income. Al
Capone went to jail for not paying taxes on illegal income.

What’s going on in New Jersey?

As of this writing (October 14, 2014), New Jersey is close to legalizing
sports betting.

Why aren’t sports leagues on board?

They can claim willful ignorance for the time being. All sports benefit
greatly from increased interest due to betting, fantasy sports included.
When leagues decide to profit directly from sports betting, they will likely
find a way. Adam Silver, with unparalleled transparency for a sports
commissioner, says sports betting is “inevitable.” (as reported by the
excellent sports betting journalist David Purdum ).
"If you have a gentleman's bet or a small wager on any kind of sports
contest, it makes you that much more engaged in it," Silver said. "That's
where we're going to see it pay dividends. If people are watching a game
and clicking to bet on their smartphones, which is what people are doing in
the United Kingdom right now, then it's much more likely you're going to
stay tuned for a long time."
Adam Silver: not an idiot. What a novel idea for a sports commissioner.

What types of bets are there?

A straight bet is the simplest and most common type of bet. It has nothing
to do with sexuality.
You make a bet on one thing. If the bet wins, you win. If it loses, you lose.
If it pushes (ties), you get your money back.
In the major American sports, there are a few types of straight bets.
Most common: Betting on a team against the point spread.
I’m writing this on Monday morning. There’s an NFL game tonight.
Seattle Seahawks (-7) at Washington Potato Skins
If you bet on Seattle, with the point spread, you subtract seven points from
their final score. You make a straight bet on Seattle -7.
- You win your bet if Seattle wins by eight or more points.
- You get your money back if Seattle wins by exactly seven points.
- You lose your money if Seattle wins by 1-6 points, or loses.
If you bet on Washington +7 instead, you add seven points to their final
score. You make a straight bet on Washington +7.
- You win your bet if Washington loses by 1-6 points.
- You get your money back if Washington loses by exactly seven points.
- You lose your money if Washington loses by eight or more points.
Simple.

So why can’t I win 50% of bets and break

even?
This would mean, theoretically, you could flip a coin on every game and
over your lifetime, you’d break even. Free entertainment!
Sportsbooks would not be in business if this were true. They are in business
and they are quite profitable, because:
They charge vigorish on every bet. It’s also called “juice,” and it’s the price
the sportsbook charges you to place a bet.
On a standard bet against the spread, the “vig” or “juice” is -110.
-110 means that you must risk $110 to win $100.
Let’s say you bet $110 on Seattle -7. There are three possible outcomes.
Win: You receive $210 back ($100 in profit).
Lose: You lose your $110.
Push: You get your $110 back ($0 profit, $0 loss).
Obviously, you don’t have to bet $110 – that’s just the easiest way to
explain it. You can bet any amount, at those same odds. If you bet $220,
you can win (profit) $200. If you bet $10, you can win $9.09.
To calculate potential profit: Multiply the amount you bet, times 11/10 ,
or, 9090909090909090… (.90 repeated).
But don’t worry: Type your bet amount into any betting website and it will
show you your potential winnings. No math needed here. Speaking of
math…
On standard -110 bets, you need to win 52.38% of the time to break
even.
That’s much harder than it sounds. But we’ll get there.

So what’s a good winning percentage?

Tiny differences in percentages make massive differences in profits and
losses over the long term.
Let’s say you bet $110 (to profit $100) per game. You bet five games a
week for the 17-week NFL season and nine playoff games.
If your winning percentage is 45% for one year, you’ll lose $1,520
If your winning percentage is 50% for one year, you’ll lose $1,070.
If your winning percentage is 55% for one year, you’ll win $580
Those numbers obviously increase if you bet other sports, and as you travel
across your lifetime.
Winning 50% of your bets can make you poor.
But: Professional sports bettors are happy with a long-term winning
percentage of something like 54-57%.
Professionals generally have complicated math models that they combine
with subjective analysis. They can estimate with a strong degree of
confidence that they will profit long term, by betting when they disagree
with the market.
That’s the difference between a pro and a non-pro: You’re guessing; they’re
relying on statistics and probability.
I’m a decent NFL bettor. I think I can win something like 55% of my NFL
bets, after years of study. But I can’t assert that with a high degree of
confidence, so I keep my bets relatively small and I won’t be surprised if I
lose long-term.
Is every bet
-110?
Nope.
And most importantly, there are online sports books that offer “reduced
juice.” This is arguably the best perk a sportsbook can offer you. You’re
getting a discounted fee on every bet you make.
For example, 5dimes.eu typically offers “10-cent lines,” meaning both sides
are typically -105. (Five cents on each side, as opposed to the standard 10
cents on each.)
This means you only have to win 51.22% to break even.
Also, the sides don’t necessarily have to have the same juice.
Take the Seattle vs. Washington example above.
Say the market starts betting Seattle heavily at some point on Monday. The
oddsmakers will typically shift the price to make Washington a more
attractive bet and even things out.
So, Seattle will go from -7 (-110) to a more-juiced -7.
They’ll move to -7 (-115), which means you need to win 53.49% of the
time to break even. Then maybe -7 (-120), with a breakeven point of
54.55%. Then potentially -7 (-125), with a breakeven point of 55.56%.
By moving from -7 (-110) to -7 (-125), betting on Seattle lost a massive
3.18% in value. That’s close to the difference between a coin-flipper and a
professional.
Prices are massively important.
If Seattle keeps getting bet, eventually the point spread itself will change. It
will move to Seattle -7.5 (-110).
The sports betting market is similar to the stock market. Money moves the
price, until the value is gone and it settles.
What happens when the point spread
moves?
Let’s say you took Seattle early Sunday, when the line was Seattle -7 (-110).
Then people started betting heavily on Seattle, so by game time, the closing
line (the last available line before the game started), moved to Seattle -7.5
(-115). The market really liked Seattle between when you bet and game
time.
Does this change your bet?
Nope. You get whatever price and spread you bet on. Line moves DO NOT
impact a bet you already made.
This means: You can get good numbers and prices, or bad ones.
Sometimes the market moves in your favor. Sometimes it doesn’t.
That’s the opposite of horse racing. When you bet on a horse, your odds are
not fixed. They change until the race starts and you get whatever the closing
number is.
Later on, we’ll talk about how to get the best number, and why it’s
supremely important.

What other types of bets are there?

Moneyline (ML) bets are even simpler than point spread bets.
If you bet a team on the moneyline, you are betting on that team to win the
game, straight up. That’s it. No point spread.
So why wouldn’t you bet the favorite every time?
You have to pay a premium to bet the favorite. If favorites win 52.38% of
the time, you do not make money. You will be very, very poor.
Here were the moneylines for the previous example:
Seattle (-350) at Washington (+290).
Say you bet $35 on Seattle ML (-350).
- If Seattle wins, you get $45 back on your $35 bet. So, you make a $10
profit.
- If Seattle loses, you lose your $35.
- If Seattle ties (unlikely), you generally get the $35 back (but you should
check your sportsbook’s rules just in case… it’s possible they could grade
this as a loss).
Or, you bet $35 on Washington ML (+290)
- If Washington wins, you get $136.50 back. So, you make a $101.50 profit.
- If Washington loses, you lose your $35.
- If Washington ties, you generally get the $35 back, but again, check.

So what do the +/- numbers mean,

exactly?
If you bet on the favorite, the odds will be negative (-110, -350, -475 etc).
This means your potential profits will be less than the amount you bet,
because your bet’s implied winning percentage is higher than 50%.
The exact meaning of negative odds are: You have to risk that amount to
profit $100. So, if the line is -350, you have to risk $350 to profit $100.
Obviously, the odds can be extrapolated to any bet size.
If you bet $35 on Seattle ML (-350), your potential winnings are $10. If you
bet $5,863.31, your potential winnings are $1,675.23
If you bet X dollars at (-Y) odds, your potential winnings are X
multiplied by 100/Y.
If you bet on the underdog, the odds will be positive (+100, +290, +1750,
etc.). This means your potential profits will be more than the amount you
bet, because your bet’s implied winning percentage is lower than 50%.
The exact meaning of positive odds are: The exact profit you’d make on a
$100 bet.
So, multiply your bet size by the odds/100.
For Washington ML (+290), multiply your bet size by 2.9.
If you bet $25, your potential winnings are $25 x 2.9… or $72.50.
If you lose, you lose $25. If you win, you get $97.50 back, $72.50 of which
is profit.
(Don’t worry; online sportsbooks calculate this for you before placing a
bet.)

How often must I win moneyline bets to

break even?
The reason (obviously) that you make much less betting on Seattle is that
they have a much higher chance of winning than Washington does. They’re
the better team. They won the Super Bowl and stuff.
But how do you know exactly how much more of a chance they have?
You need to know the breakeven percentage of a bet at the odds you’re
taking. That’s the percentage of bets you must win at those to make money.
First of all, there’s an online calculator that does the work for you.
But if you want the math, here it is:
For negative odds, here’s how you find the breakeven percentage:
[-(minus moneyline odds)]
__________________________________________
[(-(minus moneyline odds))+100]
So, for Seattle (-350):
[-(-350)]
____________________
[(-(-350)+100))]
350
_____
450

= 0.777777777…. or 77.78%.
Now, for positive odds:
100
_________________________________
plus moneyline odds +100

For Washington (+290):

100
___________
290+100
= 25651025641… or 25.64%.
For Seattle ML (-350) to be a smart bet, you must (correctly) think that the
Seahawks have greater than a 77.78% chance of winning. For Washington
ML (+290) to be a smart bet, you must (correctly) think that the Potato
Skins have greater than a 25.64% chance of winning.
Notice that 77.78 + 25.64 = 103.42. The two percentages don’t add up to
100%, so they’re slightly higher than each team’s implied win probability .
So, the market thinks that BOTH teams have lower win percentages than
required to be smart bets. So, neither team is a smart bet if the market is
right.
This is why sports betting is so hard.
To calculate implied winning percentage (which is mostly important just to
impress your friends or sound smarter than idiots who say “Seattle should
roll” on TV):
Divide the breakeven percentage by the sum of the two breakeven
percentages.
So, for Seattle: 77.78 / 103.42 = 75.2%
And for Washington: 25.64 / 103.42 = 24.8%.
75.2% + 24.8% = 100%
A lot of the time, you’re gonna agree with these numbers. Sports betting is
hard.
But you’re now better at analyzing matchups than 99% of media members,
so congratulations!
NOTE: Most people bet against the point spread in NFL ,NBA and
NCAAF. But don’t ignore moneylines if you think there is value. And in
sports like NHL and MLB, it’s much more common to bet on the
moneyline, because the only other option is usually betting on teams -1.5
runs or goals, which is hard to predict.

I still think Seattle is a good bet. What do I

do?
Again, the betting market is similar to the stock market. This means that:
- It incorporates everything that is public knowledge.
- It incorporates the thoughts of all professional bettors, because
their money has flown into the market and given oddsmakers
information on who they like.
- It incorporates the thoughts of many ridiculously complicated
algorithms that professional bettors have backed up with lots of
money.
This doesn’t mean you can’t ever make money.
The way you make money is by identifying the few times when you
actually are smarter than the market. But for the vast majority of games, this
won’t be the case.
But it means that if you think a point spread or moneyline is ridiculous,
you’re almost certainly missing something.
The market is almost never wildly off. It might be a bit off. The Seahawks
might have an 82% chance of winning. If that’s true, Seattle is a fantastic
bet.
In betting: You lose no money on 100% of the bets you pass on. Wisdom!
You don’t want to pass up a profitable bet. But if you don’t have a clear
reason to think your bet is smart, it probably isn’t.
For about 80% of NFL games, I deliberate back and forth, scour stats,
trends, whatever… and then I say “I don’t fucking know.”
It’s the smartest thing I ever say.

So how the hell can I make money?

People make money in the stock market. People make money betting on
sports. I heard a dude might have gone on the moon once. Difficult things
are achievable.
The topic of “efficiency” is complicated, but I will give my brief take.
The NFL market is hard to beat, but it isn’t 100% “efficient.” You can win
money, if you’re super-selective and smart.
When I cited that ~50% of favorites and ~50% of underdogs cover the
spread long-term, I’m using the closing number . That’s the point spread
available right before kickoff, when the market has the most information.
The closing line market includes all information from injury reports, stats
and the opinions of geniuses. You can certainly beat it in isolated instances,
but again, you’re usually wrong.
But point spreads and prices do move throughout the week, sometimes
significantly. And if you get a better number than the closing number, you
actually have an expected winning percentage of higher than 50%.
Which means you have a shot at making money.
Closing lines are very close to efficient and incredibly hard to beat. But
earlier lines are beatable. Beating the closing line is the best way to make
money.

What about other types of bets?

Over/Unders (totals)
For every game, there’s a point spread, and a total . The total is also called
the over/under , because you can bet that the total points scored in the game
will be over or under that total.
More on this later, but non-advanced bettors usually struggle with totals.
It’s tempting to think that over/unders are all about matchups, but most
bettors don’t realize that gameplans, game situations (blowouts, etc.) and
pace (do the teams play slow or fast), are incredibly important.
When starting out, monitor totals… but generally stay away until you’re
comfortable finding value. Track your opinions; you’ll be surprised at what
you find.
Parlays and Teasers
Parlays and teasers are the cocaine of sports betting. They’re attractive, and,
sure, people have had fun and maybe even some sex because of them… but
you know you’re being stupid putting them up your nose.
A parlay is when you make two or more bets, but you must win all of the
bets in the parlay or you lose. (But if one bet pushes, it’s simply removed
from the parlay and you can still win.) Obviously, you get a high payout if
you win a parlay.
In general, they are attractive, but stupid bets. I’ll go more in depth later in
the book, but internalize that they’re almost always stupid .
Teasers are also attractive and often dumb. The most common types are, 2-
team, 6-point teasers, and 3-team, 10-point teasers.
For a standard 2-team, 6-point teaser, you make two bets (teams against the
spread or over/unders), and you must win both of them. But the payout isn’t
higher like for parlays. The benefit is that you add six points to the bet you
made .
So if you teased Seattle -7.5 with Denver -8.5, your bet would turn into:
Seattle -1.5 AND Denver -2.5 (-110).
If both bets win, you win a standard -110 bet. If one wins and the other
loses, you lose your money. If one of them pushes (ties), you win with a
really small payout, generally.
I’ll go more in depth soon, but here’s the smart rule: Stay away from
teasers… EXCEPT if you can tease both teams through BOTH 7 and 3
, like in the above example.
So if a team is -7.5 to -8.5, or +1.5 to +2.5… they’re a potentially profitable
teaser leg.
Prop Bets
Prop bets are generally on things like: Who will score the first touchdown?
Will X player have over/under X yards receiving/passing/rushing? Will
there be a safety? Which quarter will have the most points?
If you research these deeply, money can be made. The markets are smaller
(as are the limits), and thus they are less efficient. If you research matchups
and opportunity, you can do well here.
Prop bets are like fantasy football… but be really careful. As with
everything, don’t make assumptions.
“Oh, he’s got a good matchup… OVER on rushing yards!” (Yeah, but will
his team be up early and thus not give him a lot of carries? Has the coach
been pimping up his backup? Will they go down early and be forced to
throw?)
Always consider every angle. Always play Devil’s Advocate. Unless you’re
really sure, you’re wrong. And even if you’re really sure, keep trying to
prove yourself wrong. There’s a good chance you still are.
Team Totals
Team totals can be yummy. They’re over/unders… but just for one team.
They’re a smaller market, so there are small inefficiencies at times.
For example, since acquiring Peyton Manning, the Broncos play really fast
and have an insane red-zone offense. They’ve generally been money on
team totals.
In Peyton’s first year, Denver scored 31+ points in six of eight home games
(30 and 25 in the two others). I don’t have the exact data, but I believe the
TT (team total) went over in seven of eight (and then their playoff game). I
believe they were 11-5 in team totals overall.
In his second year, they scored 35+ in six of eight home games, and I
believe they went 13-3 on team totals overall.
My numbers may be slightly off, but I believe they entered 2014 on a 24-8
run on team totals. They’ve started off 2014 at 3-2, for whatever that’s
worth.
Still, never bet blindly. Every matchup is a new animal.
Season Win Totals
Every preseason, there is a market on win over/unders for every team.
I don’t recommend these bets for non-advanced bettors, because they tie up
your bankroll for the entire year. If you love a bet, sure, go for it, but I
usually stay away from these.
To identify value, use everything you would on a game-to-game basis. Was
a team lucky last year? Are people overreacting to a few new additions?
(They usually don’t underreact, but it’s possible.)
The problem here is you basically need to come up with a win probability
for each game, which means you need to not only evaluate the team you’re
betting on, but all of their opponents.
This is VERY reliant on strength of schedule. Don’t just assume teams that
you think are good will go over. And leave leeway for things like injuries
and bad luck.
One good rule of thumb is to lean towards unders, because bettors are
generally biased towards betting overs on these.
Futures Bets
Futures bets are bets before or during a season on a team to win a
championship, conference, division, etc.
There are also player futures for awards like MVP and rookie of the year.
Again, don’t go nuts on these, because you’re tying up your bankroll for a
long time. But like everything, there are exceptions where there is value.
Like any good bet, a good futures bet is a value bet.
An obvious key is: Try to buy a team at its lowest stock.
This is obviously easier said than done, like everything in this book.
For NFL, buy a team after a bad loss. Or after the toughest part of its
schedule. Or before 3 straight division games that you think may catapult
them from last-to-first.
It’s all about schedule and projecting how a team can make the playoffs.
When you buy futures and your team doesn’t make the playoffs, obviously
you are screwed. But it’s important to realize: You don’t have to think that
team will win the Super Bowl. You just have to think they have a better
chance than the odds suggest.
And you can win money even if they don’t win it all, by hedging your bet.
Example: I bet on the Giants to win the 2014 Super Bowl at +6700 (67/1
odds). If you asked me for a Super Bowl pick – hell no, I wouldn’t have
picked the Giants. But I thought they were undervalued. +6700 gives the bet
a 1.47% breakeven mark. So if they have better than a 1.47% chance at a
Super Bowl, it’s a value bet.
To figure this out, estimate their playoff chances. This is harder than it
sounds.
It means you should know the win probabilities of not just the Giants, but
their division-mates and Wild Card contenders. I estimated that they were
nearly even with Philly and Dallas and had a decent Wild Card shot.
The truth is, it probably wasn’t a good bet, but I like to take a longshot
every year for fun. There’s no way for me to prove it was a good bet
without getting heavy into the math. For example, even if they made the
playoffs, there’s no guarantee they’d have higher than a 1.47% chance.
I could do the math out for you, but here’s a rule of thumb:
A futures bet is only worth it when the team is ridiculously
undervalued. Not a little bit undervalued. Ridiculously.
Take the Giants example. Unless they shockingly became an elite team,
their best bet would be winning their tough decision, likely at 8-10 wins, or
winning a Wild Card spot. They would almost certainly need to go through
a team with crazy home-field advantage like Green Bay or Seattle or New
Orleans, where they’ll assuredly be given something like a 20-30% chance
of victory.
And that’s if they even make the playoffs… and then probably win a couple
of games as underdogs.
Without home field advantage, the Giants would probably be lucky to have
a 5% chance of winning the Super Bowl… even after making the playoffs.
And even in my optimistic estimate, I realized there was a good chance they
would miss the playoffs.
Futures markets are good for taking fliers on longshots for fun. You might
win occasionally, and you will be thrilled. But they’re rarely smart bets.
But if you do make one, and your team makes the playoffs, you can hedge
if you want to guarantee profits. I generally don’t hedge until right at the
end of my bet, because the only reason I make futures is for huge returns. I
don’t want to cut into potential winnings.
But hedging is easy. Simply bet on the opponent of the team you have a
futures ticket for. If your team wins, all you did was cut into potential
profits. And if the other team wins, you made some money (or broke even,
or reduced your losses).
Determine your risk tolerance and handicap each game, and hedge
accordingly.
Maybe you loved the Giants preseason, but hate them in a particular
postseason matchup. So, hedge a lot. It all depends.
One more thing about futures: People like to bet them before the playoffs,
because it’s fun to predict who’s gonna win the Super Bowl once we know
who has a shot.
But this is usually stupid.
If you want to bet on an underdog to win the Super Bowl, you can do what
is called a moneyline rollover . What that means is: Bet on that team on the
moneyline. If they win, take your winnings and bet them on that team
again, on the moneyline. So now your bet is bigger. Keep doing this until
the Super Bowl.
It’s not always better to do this (especially if a lot of upsets happen), but
most of the time it is.
Also: If you think a team is undervalued going into a season, you don’t
have to bet them on futures. You can just bet them at the beginning of the
year until the market catches up!
As for my Giants bet, it turns out the Giants are 3-3 while the Eagles and
Cowboys are 5-1 and all of a sudden the Giants are 90/1.
I suck and betting is hard.
(Here’s a good example on the NCAA tournament and why moneyline
rollovers are usually smarter than futures .)
Live Betting
Live betting is awesome and dangerous.
The point spread updates in real time, and you can often get a better number
on a team you like, just minutes in the game.
Say you bet on NYJ +10 vs. Denver. Denver gets an opening-drive TD. The
line might move to NYJ +13.5, still in the first quarter.
Of course, you’re already down seven points. The Jets are now really +6.5
for the rest of the game. But they also haven’t even gotten the ball yet!
If you think the update is an overreaction, then bet. But, say you see
something you didn’t expect in a gameplan or mentality and change your
mind, it might not be.
Live betting can be lucrative if you’re smart, catastrophic if not.
It’s usually better to bet against what you see . Usually, don’t bet on the
team that looks good. Buy low.
Basically, bet mostly based on your pregame thoughts. Don’t let a small
portion of the game affect your mindset. Look for overreactions to small
events. Realize that momentum is not real .
Both teams can get hot and cold within one game. Exploit overreactions.
That said, you also need to adjust for game situation. Late in games, teams
play prevent defense if they’re winning by a lot. The line will reflect this,
but can you find a situation that isn’t accounted for?
Halftime Bets
At halftime, a point spread and over/under are posted for the second half .
Those numbers only apply to the second half score.
The same exact strategies apply here as for live betting. The only difference
is that you have all of halftime to react, rather than just an instant. Still,
these lines usually move within a minute or so of being posted. So be ready
to pounce.
Middling
Live betting and halftime betting can give you opportunities for middles .
That’s when you bet on both teams, but at different times with different
spreads and you win both bets .
This can be done pregame, too, if you get two different lines on a game. Say
you bet on Steelers -2.5 vs. Baltimore early in the week. The line reaches
-3.5 later on, so you take Baltimore +3.5, because you think the fair line is
-3.
The Steelers win by 3. You win both bets.
Or, say you bet on Patriots -3.5 at Buffalo.
It turns out to be a great bet: The Pats dominate the first half and are up 27-
10. The halftime line (for the 2nd half score) is Buffalo -3.5. That means the
updated game line is Patriots -13.5. You bet on the Bills 2H.
Buffalo wins the second half, 14-10. The final score is Patriots 37, Bills 24.
You win both bets. You just middled!
Some people recommend always taking middle opportunities, because if
your bet tanks in the second half, at least you make most of your money
back and just lose the vig. But if your team continues to dominate (as you
correctly predicted by betting on them pregame), you lose all of your
winnings.
Anyone that tells you to do something blindly is dumb.
Evaluate every single bet individually.
If a bet doesn’t have value by itself, don’t make it. Don’t let separate bets
influence each other. That would be like quitting your job because your
girlfriend cheated on you.
You should only make a 2H bet if you have reason to believe that things
will change. If not, stick with your full-game bet. It looks good. Sit back
and relax.

So where do I go to bet?
You can bet with a local bookie, in-person, but as I wrote earlier: This is
illegal. Although nobody has been arrested for sports betting, you have to
be careful. And if you decide to do it, make sure your bookie will pay you if
you win, and won’t do mean things like break your limbs and then stab you
with the shards.
It’s sort of like buying weed: Tons of people do it and have zero legal or
safety issues. But don’t be an idiot.
If you’re betting with a “local,” ALWAYS compare their odds and prices to
a major online sportsbook (you can go to a place like Oddsshark.com or
ScoresAndOdds.com to check odds across books).
A lot of bookies will “shade” the line towards the more popular team,
giving you more favorable odds if you bet the underdog (or out-of-town
team or something like that). This can be a benefit if you notice, or a killer
if you don’t.
This is why smart people often say: You bet on numbers and prices, not
teams.
That’s an exaggeration – you are betting on a team… but you’re betting on
them at a specific number and price . They mean nothing without each
other.
You’re not picking winners; you’re hunting for value bets that will
result in profits long term.
There might be value on the NY Jets +10 at New England from your local
Boston bookie… and there might also be value on New England -7.5
online.
For online betting – the best sportsbooks for recreational American bettors
are: 5dimes.eu and betonline.ag.
5dimes’ advantage is that they offer “reduced juice.” Lines are usually -105
on each side (or -108/-102, etc.). This is huge, but make sure you
specifically sign up for it when you first deposit, or you’ll miss out.
BetOnline’s advantage is that they release lines very early (the following
week’s games are usually posted near the end of the current weeks’). The
limits (highest allowed bets) are low for pros, but they aren’t low for
recreational bettors. If you can spot a bad line, you can do really, really well
on these. (Easier said than done.)
Another popular book is Bovada, but they’re crappy for a few reasons.
Bovada’s advantage is that they cater to uninformed bettors. That means
they shade their lines toward popular teams (usually favorites), so you can
get some extra value on underdogs. But they are quick to shut you out if
exploit them, and I have heard some bad stories about problems with
payouts. Test them out with a small amount of money for the good lines on
underdogs, but stick to and 5dimes and BetOnline for the long haul.
NOTE: I do use and genuinely recommend 5dimes and BetOnline, so I
looked into the legality and logistics of becoming an “affiliate” (getting a
commission from referring people with a link here).
It turns out: I would get paid based on how much money you lose.
Fuck those assholes.
I still think they’re one of the better places to bet, but I’m sure as hell not
making money off of your losses.
To deposit money at these places, you generally need to use a Visa Credit
Card or Western Union. Your credit card may try to block your transaction,
because they can get in trouble if they get caught letting you bet on sports.
An unconventional option that I use is betting with Bitcoin. NitrogenSports
(affiliate link) is a really cool sportsbook that only deals in Bitcoin. It’s
painless, but make sure you understand how bitcoin works before
depositing. (Email me at [email protected] if you have Bitcoin
questions.)
If you’re based outside of the US, Pinnacle is arguably the most respected
sportsbook in the world. They don’t take US customers, but you should
refer to their prices as the market standard. They take big bets proudly.
But if you check PinnacleSports.com for odds, you’ll realize the odds look
funny. That’s because they’re European, and Europeans look funny. (I’m
allowed to say that; I’m writing this from Prague.)
You can use this tool to enter the odds and convert them to American, like a
good freedom child.
Should I use more than one book?
Sportsbooks are the opposite of assholes: The more the merrier.
You should have as many “ outs ” (places to bet) as possible. Prices are
usually pretty similar across online books, but not always. You should
always be “line shopping” to ensure you get the best price on a bet.
I recommend starting small and just depositing in BetOnline and 5dimes,
and possibly a bit in Bovada. As your bankroll grows, you should consider
branching out.
Now that we know the basics: Let’s learn how to find value bets.
STEP 3: Learn How Smart Bettors
Think
The 17 Commandments of Recreational
Sports Betting
These commandments were passed down from Moses to Floyd
Mayweather. But Floyd Mayweather ripped them up because he is cocky
and probably horrible at sports betting .
I have since pieced the scroll together to give you 15 rules that will put you
ahead of the ~80% of people who lose before they even start.

1) Only bet what you can afford to lose.

Obvious, but crucial. Don’t see sports betting as an investment, until you
have proven you’re a winner. See it as a hobby. Even professionals run the
risk of ruin. The market is ever-changing and getting smarter, and betting
on sports is HARD.
There’s a chance you will lose all of your money. Especially if you don’t
follow the nine remaining rules.

2) Base your bet sizes on your expertise.

The “optimal” way to decide how much to bet on a game is determined by
the “Kelly Criterion.” It’s a rule that gives you an exact percentage of your
bankroll to bet, based on the exact “edge” you have.
But for anyone who isn’t advanced: You don’t have a clue what your exact
edge is. Don’t worry; I don’t either.
Most of the time you have no edge, and there’s no way you can determine if
you have a 1.3% or 3.7% or 2.7893587% edge on the market unless you
have your own proven algorithm and a lot of experience and data on your
successes and failures.
What’s gonna happen is: You think you have a small edge on a side. You
may or may not be right. Occasionally, you’ll feel a lot more strongly about
a bet.
If you really are confident you have a bigger edge on a particular play, I
recommend doubling your average bet. But keep track of how you do on
your big bets. Over time, you might realize you perform the same (or
worse) on these bets. If so, stop doing that.
Start with an average bet size of 1-3% of your bankroll. Go a little
higher if you don’t have much in there, but realize that’ll mean your
chances of bankruptcy are higher. This probably seems conservative, but
it’s the smart thing to do.
You WILL go on hot and cold streaks where you’re bankroll swells and
shrinks. You can’t escape probability; IT WILL FIND YOU.
Trust me: You don’t want to lose everything because you had a bad
weekend. I’ve been doing quite well recently… but then I went 2-9 in NFL
bets last weekend. And I got out of control and made two bets than were 3
and 4x my normal bet size.
I lost 5 weeks of impressive profits in one weekend. Don’t be stupid like
me.
That said, I think you should stagger your bet sizes, based on how much
you pay attention to a particular sport or conference.
I pay closest attention to the NFL, NHL and Big Ten basketball. I have
tracked my performance for a while in them, and I do best in NHL. So I
have a separate bet size for hockey, that’s higher than NFL and B1G
basketball. It’s justified by my degrees of success.
If I ever get an itch to do something stupid like bet on the NBA (which I
don’t pay close attention to), I make tiny bets relative to the sports I actually
know.
At the beginning, though, you won’t know how good you are at everything.
So bet a bit more on what you think you know better, but track your success
and adjust accordingly.
Just don’t freak out over a few games; you need a large sample to
understand your true success, and even then, things can change.
Start out ultra-specific and ultra-cautious. And realize you might surprise
yourself. You might not be good at betting on your favorite teams’ games.
You might be overly optimistic or pessimistic. Track everything.

3) Seriously, TRACK EVERY BET YOU MAKE

This is absolutely non-negotiable. If you don’t track your bets, you’ll have
no idea if you’re winning at 49 or 53.2 or 59%, if you’re getting good or
bad numbers, if you’re struggling with certain types of bets, etc.
It will also stop you from making impulse bets or going beyond your means
if you know you can’t avoid looking at the results.
You can make your own spreadsheet to track your bets, but that’s annoying.
I recommend one of these two options:
1) Use a premade template, where all you have to do is type in your bets
manually and then enter if they won or lost.
I found one a few years ago that I use and love. If you’d like it, send an
email with the subject line: “template” to [email protected], and I’ll
send the file to you for free.
2) Use Winafy , an incredible software that does all of this automatically if
you use one of their supported online books (5dimes and BetOnline are
supported). You can also enter offline bets manually.
They’re awesome, but you do have to pay a small fee monthly. It might not
be worth it if you’re not betting a lot of money.
You can try it out for free HERE (that’s an affiliate link, I get a few bucks
for referring you, full transparency).
Both will track your records for each sport, for favorites, underdogs,
moneyline bets, etc... including ROI (return on investment). You can track
big bets and small bets and you will learn a ton about any biases or
deficiencies you have. Everyone has them.
For example, I’ve found:
- I used never bet against my alma mater, Michigan. I did very
well when I bet on Michigan, but there were lots of missed
opportunities betting against them, because I hated doing it. I
just convinced myself to stay away and lost out on potential
winnings.
- I’m best at betting NHL. This isn’t surprising, because it’s a
much smaller market than NFL.
- I make stupid bets when I’m on losing streaks. I know not to
“chase” when I’m losing, but I do it anyway. I’m human. And
I’m stupid.
You won’t be able to change these behaviors until you spot them in
objective data.

4) Only make bets that you think have positive

expected value (+EV).
The real rule is: Only make bets that actually have positive expected value.
But as we’ve discussed, a recreational bettor doesn’t know his exact edge.
So at least make sure you’re aware of the implied winning percentage of the
bet. For a standard -110 bet, you don’t have an edge if you’re not more than
52.38% confident in it.
Expected value is expected profit or loss you think you’ll achieve, on
average, by making a bet.
To figure that out, you need to know:
- The market-implied probability of a bet winning (for the Seattle
example… what do the odds say their chances are?)
 - Your expected probability (what do you think Seattle’s chances
are?)
- Why do you think you’re right and the market is wrong?
This sounds really simple, because it is. If you disagree with the odds, you
should theoretically make a bet. But as I keep repeating, there’s one more
piece.
You’re usually wrong.
So the Golden Rule for recreational bettors is: Only make a bet if you
disagree with the market, AND you have a significant reason to believe
you are correct.
If you assume the market is generally right (or efficient), the expected value
of your bets against the closing line is almost ALWAYS negative. You
should assume you made a bad bet if you don’t have a better number
than the close.
(Realize: You WILL win bets where you get a bad number, and you WILL
lose bets where you get a good number. But the point is, if the market is
efficient, your good and bad luck will even out long term and you will lose.)

5) Make sure your reason for betting isn’t bullshit.

Most people make their bets based on complete bullshit. They disagree with
a large market based on cherry-picking stats or trends, or by believing their
lying eyes.
For example: Seattle (-350) at Washington (+290) , again.
Here are three common reasons people would bet on the Seahawks:
A) Cocky Idiocy
“They’re the best team in the NFL! Their odds should be like 90%!”
You’re smart enough to realize how stupid this reasoning is. You are
completely guessing based on a common opinion that shows no proof you
are smarter than the market.
Cocky idiocy is basically 90% of the “expert analysis” you see on ESPN
and in much of mainstream (and even non-mainstream) media.
Saying the better team “should” win and/or cover because you think
they’re better is the single dumbest reason you can have for making a
bet. You WILL lose if you do this. Do not do this.
B) Useless Trends
Here were some trends on this game.
- “(Pete) Carroll-coached teams are just 3-6-1 against the spread in their
first game off the bye.”
- Seattle Is 3-5-1 ATS In Its Last 9 Games When Playing Washington
- Seahawks are 17-35-1 ATS in their last 53 games in October.
- Redskins are 4-1 ATS in their last 5 games in Week 5.
Wow, Washington sure seems like a LOCK! (NOTE: Anyone who uses
the word “lock” non-ironically is a fool.)
But wait!
- Washington Is 1-3-1 ATS In Its Last 5 Games When Playing Seattle
- Washington Is 4-8 ATS In Its Last 12 Games
- “ Since 1997, Washington is just 2-12 straight-up and against the
spread at FedEx Field.”
- Seahawks are 10-0-1 ATS in their last 11 games after accumulating more
than 250 yards passing in their previous game.
Ah, crap, now I like Seattle!
You can find a trend to say almost anything.
That doesn’t mean all trends are useless. For example, a simple, obvious
trend like this does have meaning:
- Seahawks are 37-16-1 ATS in their last 54 games overall.
You can’t argue that trend is meaningless. But it also is not in any way,
shape or form a reason to bet on Seattle by itself.
With trends and statistics, you need to determine exactly what the trend or
statistic is telling you, and then use that as a part of your analysis, realizing
that most insights you’ll have are already incorporated into the line by the
market .
This trend tells us one thing: The Seahawks have outperformed market
expectations in the past 54 games (3+ seasons).
This is Russell Wilson’s third year in the league. The Seahawks have had an
above average amount of continuity in the past three years. It’s a fairly
significant timeframe.
But what does this mean ? Why the hell do we care?
All it means is that the Seahawks have been undervalued the past three
years.
Right now, everyone knows they’re good. This is 100% publicly available,
obvious information, and while valuable for analysis, it gives us no edge by
itself.
It says that there was an edge for the past three years. But now that
everyone knows that and is aware of how good they are, there’s a good
chance that edge is gone.
If everyone knows something, it doesn’t give you an edge. In fact, it means
you should consider the other side: Are people overrating them now
because of recent success? Always dig beyond the surface. Question
everything.

6) “Bet numbers, not teams.”

Most smart bettors repeat this ad nauseam, which is why I’m repeating it.
Yes, the team you’re betting on does matter. But I constantly get questions
like:
“Do you like (X team) this week?”
I understand the question, and I used to ask it. But you need to stop doing
this if you want to make money.
Evaluate every team AT THE PARTICULAR PRICE YOU CAN GET. It
means nothing to say, “I think the Seahawks are a good bet.”
Last week, I saw the Seahawks at -325 on the moneyline hosting Dallas at
the beginning of the week. They moved to -500 before game time. That’s a
MASSIVE difference.
The breakeven percentage at -325 is 76.47%. At -500, it’s 83.33%.
If you missed the early line, you lost an insane amount of value.
Dallas pulled the upset. If you got the pricier line, you lost 154% more
money than someone who got the better line. Long term, that compounds.
I can’t repeat this enough: You’re not picking winners. You’re making
constant value bets that will win at a higher percentage than the
market expects.
The best bettor in the world isn’t going to win more than 60% of his -110
point spread bets, long term. He’d be thrilled with 55%.
You’re gonna lose ALL THE TIME. So make sure you’re getting value and
being a stingy price shopper.

7) Ignore mainstream statistics.

99% of stats ESPN, standard football broadcasts and most media members
cite are complete horseshit.
For example: Last year, the Buffalo Bills were “second in the NFL in
rushing.”
You’re all smart enough to know the Bills weren’t close to the second-best
rushing offense in the NFL. So why can announcers say crap like that quote
above?
Simple: They ran the ball a lot, mainly because their passing game sucked.
They were first in the NFL in rushing attempts by a large margin.
If you used a simple stat like yards per attempt, you’d see Buffalo as 14th in
the NFL. Much more reasonable. But still not optimal.
Go one step further, to DVOA, a much better stat that adjusts for game
situations and difficulty of opponents. Buffalo was 17th in rushing offense.
Never take a stat as gospel, even an advanced one like DVOA. But
estimating that Buffalo had the 17th best rushing offense, as opposed to 2nd ,
will get you MUCH further along on the money-making curve.
Keep reading for a full section on advanced stats and DVOA, but the most
important takeaway is: Stats don’t “lie,” but they do try to trick you.
Understand exactly what they are saying. When some writer writes about
the Bills’ “second-ranked rushing offense,” he means that the Bills rushed
for the second-most yards of NFL teams. That’s it. Nothing more. Don’t put
words in stats’ mouths.
This does not account for the fact that they ran the ball more than any team,
and other factors that skewed the data. It says ABSOLUTELY NOTHING
about how good they were at running the ball.
“Counting stats” like this are virtually useless.
Remove mainstream media from your diet and you will feel better and have
more time.

8) Ignore win-loss records.

“You are what your record says you are." - Bill Parcells
Bill Parcells is lucky he went into coaching and became a legend. Because
if he bet on sports, he’d be homeless.
A lot of the time, I have no idea what a team’s record is until I decide to bet
on them. That sounds stupid, but it’s a competitive advantage.
The only time a team’s record matters is if it affects their motivation.
And that’s only relevant in rare cases, like a team on the brink of
elimination, a team that’s coasting, or a team that has given up.
You should generally be aware of records. But I wouldn’t look at them
before handicapping a game. All they do is lead you to misevaluate teams.
A good record obviously has something to do with talent. But it’s also
shaped by strength of schedule and luck. Advanced metrics are infinitely
better predictors of future wins and losses, than past wins and losses are.
Wins and losses are just like bad stats. They don’t necessarily lie , but they
do try to trick you.
Use advanced stats, in combination with highly skeptical trust in your own
eyes, to evaluate how good a team is.
Oftentimes, the best value is in betting on a team with a bad record that isn’t
actually so bad, or betting against a team with a good record that isn’t
actually so good.
This weekend in college football, No. 10 undefeated Arizona is hosting
unranked, two-loss USC.
And USC is favored by 2.5 on the road against the undefeated, top-10 team
.
Point spreads try to predict the future. Rankings and records evaluate the
past. The two have nothing in common.
Wins and losses are important in real life, but all you care about as a
bettor are things that predict future performance.

9) Place your bets (relatively) early.

Look at lines as soon as they open (for BetOnline and the NFL, it’s usually
near the end of Sunday’s games). If a line seems suspect and you’re
experienced enough to trust your judgment, don’t wait for the market to
move it.
But when you’re just starting, I would write down the opening lines and
watch them move throughout the week. Then try to reverse-engineer why
they moved.
Did the oddsmakers overreact to what was happening during those games?
Did they overvalue popular teams? Did they undervalue an unpopular team
that had great advanced metrics? Have they not caught up to a much-
improved team that the market likes?
Early in the week, you’re not competing with the big money. The limits are
usually something like $500 early on. Pro bettors laugh, but smart
recreational bettors should salivate. The lines are not at all efficient.
Give it some practice, and you’ll spot the bad lines.
This is your absolute best strategy for making money betting the NFL.
Ignore it and your wallet will hate you.
(NOTE: If you don’t see value in the openers, wait and monitor the lines
during the week. Lines move around, sometimes due to professionals
manipulating the line as a smokescreen. You can get value during the week,
too. The worst thing to do, usually, is to bet right before game time.
Because as I’ve discussed, the closing line is the most efficient.)

10) Play Devil’s Advocate

EXAMPLE : A few weeks ago (or many weeks ago, depending on when
you read this) many people freaked out when they heard the Bucs were
“only” 7-point underdogs in Pittsburgh.
This was because the Bucs had just lost 56-14 in Atlanta the week before,
and Pittsburgh had just beaten Carolina on the road, 37-19.
These results were 90% of the reason people freaked out.
But these results aren’t just public knowledge, they’re literally the first
things that appear in everyone’s mind when thinking about this matchup.
So why would that be a reason to bet? The results are clearly, 100%
factored into the line. So are other, less obvious things.
If you looked a bit beyond just the most recent results, you’d realize that the
Bucs were getting abysmal QB play and couldn’t possibly be downgrading
with Mike Glennon, that they were getting Doug Martin back against a
leaky run defense, that losing in Atlanta should be expected… and that the
Steelers beat the Panthers without DeAngelo Williams and with a version of
Cam Newton that couldn’t do anything but throw with his upper body.
All of a sudden, a 7-point spread made more sense.
The point isn’t that it was obvious the Bucs would win or cover, but the
point was it certainly wasn’t ridiculous that the Steelers were favored by
“only” 7.
A point spread is almost never ridiculous in the NFL. Maybe a tiny bit off,
at best.
If anything, the market was too high on Pittsburgh. Always play Devil’s
Advocate.

11) Don’t bet parlays.

Parlays are like babies: Everyone likes them and gives them money even
though they make you poor.
A tweet from WagerMinds gives you the numbers:
“For every $100 bet on parlays in 2013, bettors lost $34.63”

You can prove parlays are (almost always) dumb by simple math.
A redditor did that with Pinnacle’s numbers, very similar to other online
sportsbooks’.
The parlay odds at Pinnaclesports.com are:
2 teams (+260)
3 teams (+600)
4 teams (+1200)
5 teams (+2400)
For everything but 3-team parlays, your win percentage has to be a bit
higher than 52.4% to break even. So: You have to be a bit better to profit on
parlays than you do to profit on straight bets, and making money on straight
bets is already pretty damn difficult.
But the real problem is that you will go a long time without winning
parlays. So even if it evened out long term, you’d be subject to long losing
periods, where your bankroll will be compromised. You might have to
decrease your bet size, making things not even out long term.
If there is less value and a smaller chance of winning, why bet them?
The most obvious exception is a correlated parlay , which is
mathematically smart. That’s when your first bet winning makes your
second bet more likely to win.
For example: You parlay the first-half OVER with the full-game OVER.
There’s a good chance they’ll either both lose… or both win. If the game
goes over the first half, there’s a great chance it will go over for the whole
game.
But sportsbooks aren’t dumb, and they don’t allow you to make the above
bet.
There certainly are exceptions (say, you think the underdog only has a
chance to win in a shootout, or something), but again, there’s a good chance
you’re wrong if you think a parlay is correlated.
It would make sense to bet, say Jets +6 and Bengals -1 if the Jets winning at
4:30pm had an impact on the Bengals winning at 8:30pm. They would be
correlated!
But as your trusty brain will tell you, there is no correlation between one
team winning one game and another team winning another completely
different game.
So, bet them separately.
NON-MATHY TAKEAWAY: Parlays are (almost always) bad bets.
It takes four lines to prove that you should never bet parlays as a
recreational bettor.
But you’re probably gonna make them anyway. They’re fun.
You know alcohol is bad for you, but you probably still drink. Like
drinking, I indulge in parlays with specific rules in mind:
 - I allow myself one unit (one average bet size) in parlays a
month.
- I usually split that into four ¼ bets, but one bigger one is fine.
- I usually make them ridiculous long-shot bets, like six-team
parlays with an underdog or two or three on the moneyline.
They never win and are stupid, but they’re fun and not very harmful since I
limit them. Controlled stupidity is one of life’s underrated pleasures.
NOTE: Payouts in Vegas are usually worse than online, and I’d assume the
same will apply for your local bookie. If this is the case, parlays can be
significantly dumber.
There are exceptions to every rule, and as you get more advanced, you may
find some situations that justify parlays. But you can almost never go wrong
staying away.
One specific instance where they may be smart is if you see value in a
massive favorite, but don’t have the bankroll to risk on that favorite. Parlay
it with another +EV bet and you have yourself a +EV parlay, assuming
you’re betting at a book like BetOnline or 5dimes that gives you decent
parlay odds.
Most people have a desire to parlay two big favorites together, as if that
creates value out of thin air. If a side doesn’t have value, don’t bet it.
Parlaying it does not help.

12) Don’t worry about short-term results.

Don’t worry if you lose one week, or even five weeks in a row.
But don’t get cocky and change things if you win big one week, or even
five weeks in a row.
Variance is a thing and you cannot escape it.
In layman’s terms: No matter how good you are, you will go on bad streaks.
You will lose two and five and seven in a row if you bet long enough. You
will lose for a month or three. It also works in reverse. No matter how bad
 you are, you will win five in a row or even 20 out of 25 or whatever. It’s
statistics.
If you made a significant living as a professional sports bettor for 50 years,
you’d probably have a losing year or multiple, just due to variance.
One game means NOTHING. One bet means NOTHING. Even one season
can mean very little. I made good money betting on NFL last year, but there
is zero chance I would ever promise someone my picks will make money
this year.
Your luck will even out over the long term, so all that matters is your skill
and discipline.
Along these lines: Don’t get caught up in emotions.
You’ll be tempted to double your winnings after a good day, or chase to
recover losses on a bad day. DON’T EVER DO THIS. Remember the
golden rule: Every bet is independent, and only worthwhile if it is +EV.

13) Don’t bet on big games for the sake of it.

Everyone wants to bet on big games, even though they’re often the hardest
ones to handicap, due to large amounts of action and the difficult nature of
predicting a competitive game.
You can make the same money betting on Iranian dodgeball as you can
betting on the Super Bowl. I hope my point is clear.
But I understand. You want to bet big games, because it’s fun.
If you absolutely must bet, make your bet tiny and realize it’s stupid and
just for fun. I had no opinion on the Super Bowl last year, but I wanted to
bet it and I had a good year, so I let myself make a small bet. But I knew it
wasn’t smart… and it lost.
Try to limit these stupid bets, like parlays, as much as possible. Consider
them an “idiot tax,” like lottery tickets.

14) Buy low, sell high.

Just like the stock market, it’s ideal to buy low and sell high.
But unlike the stock market, you can’t hold something long term, so buying
a team’s stock when it’s high just because you think they’re a good bet for
10 years doesn’t work.
It’s more like day trading.
The way to win is to understand the market’s opinion, understand why they
think that, and then have a clear, intelligent reason why that’s wrong.
The other way to win is to realize winning is hard as poop and do
something else with your time. But we all love sports, so we’re here.
The best people do this with complicated algorithms, but you’re probably
not reading this if that’s you.
Without serious math, you aren’t going to be right often. But if you
specialize and bet early lines, you absolutely have a chance.
For example: I’m a big Jets fan. I know the Jets very well.
Before Week 1 in 2013, everything fan and media member was talking
about how the Jets would be the worst team in the NFL, and many were
literally calling them the worst team in NFL history.
Whenever everyone is echoing each other, they’re usually wrong.
Especially in football.
They entered Week 1 as 3.5-point home underdogs to the Tampa Bay Bucs,
who were being weirdly praised due to signing Greg Schiano, who quickly
turned into a punchline. People also seemed to ignore the fact that Josh
Freeman had been as bad as Mark Sanchez the previous year, and that no
matter how bad Geno Smith could be, he couldn’t possibly perform worse
than Sanchez had the previous year.
2012 was the perfect storm of bad luck for the Jets, and while they weren’t
a good team by any means, literally EVERYTHING went wrong. I had seen
nothing like it.
They had a talented defense, a talented defensive coach, and like I said,
there was no way their offense could be worse. Almost as bad, maybe, but
not equally as bad.
So I saw value on them. And I bet on them, despite 40 of 40 “experts”
picking them to lose .
I say this not to toot my own horn. My hindsight, like yours, is 20/0.
I’m cherrypicking an example to prove a point, but that point is important:
If you understand exactly what the market is saying and you can find a
hole in its logic, you’ve found value.

15) The point spread is telling you a story.

Comprehend it.
In the Jets’ instance, the point spread was saying that the Bucs were a
mediocre team, and they were 3.5-point favorites on the road in New York.
That means New York was perceived as close to the worst team in football.
If you understand the story and have reason to dispute it, you may have a
+EV bet.
I disagreed here, because the Jets weren’t the worst team in football the year
before, and I saw significant reason they’d be better. I understood why
people thought they were falling apart, but from my seat in New York, it
just seemed like a media-created shitstorm that flooded into the market,
short term.
Another important point: Judge how much stock changes week-to-week.
The Jets won on a fluky penalty that week (although they were covering the
whole game), and people still doubted them. They remained undervalued
for another couple of weeks.
Sometimes a team’s stock doesn’t adjust enough because it was wrong to
begin with (like the Jets). That’s called “anchoring.” It’s unintuitive to stray
far from the anchor of knowledge you started with. If the Jets started out as
a joke, it’s hard for your mind to grasp them being decent.
But the market isn’t dumb, and caught up after just a few weeks.
16) Take your recency bias vaccine.
FACT: Recency bias has killed more Americans(‘ bankrolls) than natural
causes.
One game means nothing. Repeat this over and over and over and you’re
smarter than 80% of bettors.
You know how experts are always “shocked” when a team who played
badly one game won the next week? Yet it happens EVERY DAMN WEEK
IN EVERY DAMN SPORT.
Sports are fluid.
Recency bias is a natural human reaction, but natural stuff can hurt you,
man.
Recency bias is our tendency to remember a team’s most recent
performance (especially the result).
One game can tell you something . But if a team’s stock goes way up or
way down based on one game, it’s likely that’s an overreaction.
For example, entering Week 5 of the 2014 NFL season, Tom Brady and the
Patriots were playing like absolute crap. There was talk of Brady being
benched and the dynasty being over… and this is why you should never
listen to the media.
Remember that time we were all gonna die from swine flu?
New England closed as 2.5-point home underdogs to Cincinnati, whom
everyone was anointing the best team in the NFL.
The Pats dominated.
But then their stock shot up, everyone said they were “back,” and they
jumped up to 3.5-point favorites in Buffalo, when the line assuredly
would’ve been something like a PK or -1 if it had taken place a week
earlier.
What changed? Well, Brady proved he wasn’t “done.” The overreactions
were definitely overreactions. But acting like everything is fine… that’s
also pretty clearly an overreaction. One impressive performance at home
against a bad road team in a good spot with ultimate motivation doesn’t
disprove four weeks (not to mention the previous year of declining Brady
performance).
It’s pretty clear that both reactions were overreactions.
As with most things in life, the true answer lies somewhere in the
middle of the two extremes.
How do you make money? Find the overreactions and exploit them.

17) Never buy picks from anyone.

99% of people who sell picks are full of shit.
I’ve written about this at length (calling out ESPN, Grantland and the New
York Times and CNBC , but here’s the short version:
1) A very small fraction of people win money betting on sports.
2) These people generally keep a low profile.
3) Selling picks makes it much more difficult for the seller to win, and
there are many logistical problems with selling picks. It’s a lot of
trouble.
4) If someone has picks worth selling, it is much, much easier for him
to become rich betting on his own.
5) If someone is selling valuable picks, they will be very expensive,
and also very difficult to take advantage of.
I know of two examples (among hundreds of scammers) of honest pick
sellers that seem to be potentially valuable investments. I won’t mention
them, though, because I don’t think they’re wise for the average person.
The three main problems are:
1) Even they can’t guarantee they will continue winning. Things get
harder every day.
2) They’re expensive.
3) You have to place the bets immediately when they are sent to you,
because the market moves just based on their reputation.
Don’t buy picks. And if you ever see ESPN or some mainstream media
outlet link to some guy that sells picks, I can assure you that person is a just
a successful, sleazy marketer.
(NOTE: I write this in 2014. Maybe this will change. But for the time
being, we have a long way to go.)

MISCELLANEOUS:
Do “trap games” exist?

There are two types of trap games. Let’s start with the non-gambling
version.

A good team takes a bad opponent for granted, and the bad opponent is
fired up to take out the good opponent.

A legit psychological edge for the bad team, generally. Coaches and
players’ makeups affect this, but the situation is real. Again, you’ll have to
determine its value yourself.

The bookie “knows” the result.

This does not exist. As you now know, it’s not bookies that “know,” it’s the
market that sort of “knows.” If a line seems strange, you’re probably
missing something. It’s probably one of the common biases.

Does “Vegas” “know?”

As we’ve discussed, “Vegas” plays second fiddle to online sportsbooks, and

if anyone “knows,” it’s the market.

Via the best sports betting journalist out there, David Purdum:

“Bettors are more responsible for the closing lines than oddsmakers. In
fact, after glancing at their power ratings, checking injury reports and
spitting out a number designed, in most cases, to attract even action on both
sides, oddsmakers’ jobs are pretty much done. It then becomes the bettors’
job to decide whether a team will beat another by a certain amount of
points, thus shaping the line into its closing number.”

How close are point spreads, and does it matter?

More from Purdum:
“ Nearly 60% of NFL games have been decided by more than seven
points of the spread , according to Spreadapedia. More than 28% of NFL
games have been decided by more than 14 points over the past 10 seasons .
And only 5.5% of games have landed exactly on the closing point spread.
”
So, no, “Vegas,” doesn’t “know.” And neither does the market. Nobody
“knows.”
But that doesn’t mean betting is easy.
Check all this out (numbers updated prior to the 2013 season):
93 – The number of games since 2003 that have moved more than three
points from the opening point spread at PinnacleSports, according to Bet
Labs .
341 – The number of games since 2002 that have finished with a margin of
victory of three points. ( Complete chart of margin of victory )
22 – The number of times a 3.5-point favorite has won by three points since
2002.
8 – The number of times a 6.5-point favorite has won by seven points since
2002.
61.3 – The percentage of games Green Bay Packers coach Mike McCarthy
has covered the spread in, the best ATS mark among active coaches with
more than two seasons under their belt.
24.5 – The largest point spread of any NFL game since 1975. The New
England Patriots were favored by 24.5 over the Philadelphia Eagles in
Week 12 of the 2007 season.
1,463 – The number of underdogs to cover the spread since 2002.
1,415 – The number of favorites to cover the spread since 2002.
2,886 – The number of games to go OVER the total since 2002.
2,878 – The number of games to stay UNDER the total since 2002.
See? There are some opportunities (~9 games per year have massive line
movement) and betting on Mike McCarthy (Aaron Rodgers?) was
profitable… but still, an almost even amount of favorites and underdogs
cover, as is true with over/unders.
What does this mean?
Point spreads aren’t “correct guesses.” But they are the best predictive tools
available, they are decently close, and they do accurately predict that 50%
of favorites and 50% of underdogs will cover, long term. Just not
necessarily by that exact number.
Bookies don’t need to or want to “know.” They just want to not fuck up
royally, so they can take their 10% cut and spend it on nice things.
I’m not saying their job is easy, but it’s much easier than yours.
What the hell is a “sharp” bet or bettor?
If you go on Twitter on NFL game day, the only gambling information you
will see is:
1) Point spreads moving (as bigger bets come in under higher limits)
2) Mention of mostly-idiotic trends,
3) Mention of “sharps” and “squares.”
What the hell is a “sharp” or “square?”
A “sharp” bet is one that is +EV. So, beating the closing number is “sharp.”
A “square” bet is one that is worse than the closing number.
So if you feel the need to call people “sharp” or “square,” “sharp” bettors
are ones that don’t fall prey to the biases I’ve laid out, beat the closing line
and make money.
So, the Denver Broncos open -9 at the New York Jets. The number gets to
-10.5 at one point in the week, and closes at -10.
Anyone who bet Denver -9 or NY +10.5 made a “sharp” bet. Those are the
“sharp” sides. They have the highest expected value.
A lot of people will say things like the Jets are the “sharp” side, because
they’re less popular. That’s bullshit. They’re only the sharp side if you got a
sharp price.
There’s nothing “square” about betting a popular team at a good price.
The reason the average bettor is “square” is because he doesn’t realize this
and bets Broncos -10 before game time when -9 was available earlier in the
week.
Step 4: Learn NFL Betting Tips And Tricks
In the NFL, all points are not created equally.
Points are not people, so we can and should discriminate against them.
In the NFL, there are what we call “key numbers.” They’re the most common margins of victory,
and they are KEY.
SportsInsights was kind enough to compile all of the margins of victory from 2003-2013.

MARGIN OF VICTORY # OF GAMES % OF TOTAL GAMES

3 411 15.39%

7 255 9.55%

10 164 6.14%

6 149 5.58%

4 142 5.32%

14 128 4.79%

17 94 3.52%

1 94 3.52%

2 91 3.41%

21 84 3.15%

8 83 3.11%

13 81 3.03%

5 79 2.96%

11 68 2.55%

20 62 2.32%

18 58 2.17%
 MARGIN OF VICTORY # OF GAMES % OF TOTAL GAMES

15 42 1.57%

12 42 1.57%

16 40 1.50%

9 39 1.46%

19 31 1.16%

0 2 0.07%

3 is in a class of its own, and 7 is also massively important.

If you bet on a team at -3.5 when -3 or even -2.5 was available before or after you bet, you missed
out big-time.
Your half or full-point miss seems like no big deal at first glance. But 15.39% of games land on 3.
And in an efficient market, you’d expect that games projected to be close will land on 3 even more
often than that.
If you get bad numbers consistently, you’re absolutely screwed long term.
But as always, you need to dig beyond the surface.
Key numbers are public information and commonly discussed among bettors. A lot of people think
they’re being smart by “buying” points (paying a premium to go from, say, -3.5 to -2.5), but books
will generally only let you do this if it’s profitable for them .
Never trust a sportsbook. They’re aware of the chart above.
Fortunately, our friends at Winafy have looked deeper into this, and found some insights for you .
“There is one glaring example, however, where purely betting on numbers can be profitable:
teasing home teams six points either way through 3 and 7 has been successful 74.4% (equivalent
to a 55.3% breakeven percentage) of the time since 2004. Smart sportsbooks prevent their
customers from betting on this +EV wager either by raising the vig or adjusting the line, but there
remain a handful of books that will still allow such 2-team teasers at -110.”
Like I said earlier: Generally only bet teasers if you can tease through 3 and 7. This works for
favorites and underdogs. But still make sure you like both teams; don’t do it blindly.
Also:
“Since 2004, 2.5 and 6.5-point dogs are a combined 154-134 (.535) ATS. Meanwhile, 3.5 and 7.5-
point dogs are 141-162 (.465) ATS over that same period…
Blindly betting based on this evidence is probably not wise, but gamblers should at least pause
before they bet against these trends.”
The takeaway here: People are so aware of the importance of key numbers that their value might
be gone. In fact, the obsession with being smart may have opened an opportunity for betting
against key numbers because the market now overvalues them.
The market is constantly evolving. Always be on the lookout, but never assume anything is easy.
Inefficiencies may appear from time to time, but they’re often tiny and hidden and quick to
disappear.
Winafy also answered: What price is fair to buy points?
“Below are tables summarizing the historical success rates of buying a half or full point on sides
in the four point spread scenarios we discussed. For example, if you can move a 2.5-point dog to
+3.5 at a price better than -155, that wager is historically +EV.”
From the tables, you’ll see the different historically fair prices associated with buying points for
common dilemmas.
Advanced stats are key… if you
understand them.
Advanced stats are not everything. You should watch the games and look
for discrepancies. But your eyes lie more than stats do.
If you understand what advanced stats are saying and don’t put words into
their mouth (like many cocky analytics gurus do), they can tell you valuable
information.
For NFL, these are my favorite advanced stats. If you know of any other
good ones, please email me at [email protected] .
The two sites I use most often for NFL stats are:
TeamRankings.com and Football Outsiders.com .
ProFootballFocus.com is quite cool if you don’t mind paying for access.
Here are my favorite advanced football stats.
An important caveat for all NFL stats, though: The season is short, and
susceptible to big changes based on one weird week. Always be
skeptical, especially early in the season when the sample size is
miniscule.
DVOA (Defense-adjusted Value Over Average)
DVOA is my favorite stat to use while betting. It’s not close.
There’s a reason media constantly cites rushing and passing ranks for
offenses and defenses: It’s really valuable to know how good a team is on
each side of the ball. It also helps determine good and bad matchups for
teams.
But standard offensive and defensive ranks are useless. All they do is tell
you the yards that unit has allowed or produced. Those numbers are
distorted by: Strength of schedule, game situations (garbage time, for
example), pace (slower teams get fewer drives and thus fewer yards).
DVOA adjusts for: Game situation, strength-of-schedule, down, distance,
fumble luck and special teams DVOA even adjusts for type of stadium
(warm, cold, dome, Denver). Teams are compared to league average.
Example: In 2013, the Super Bowl champion Seattle Seahawks finished
first in overall DVOA, at 40.0%. That means they were 40% better than the
league average, overall. They were just 9.4% better on offense, but 25.9%
better on defense. (Most teams aren’t above average on both sides of the
ball, hence the higher total number.)

For example: A nine-yard draw play on 3rd -and-14 is not considered a

success. But in yards per carry (or, worse, total rushing yards), it’s
considered a massive success.
In the past, DVOA has been used effectively to bet games. But it was a big
failure in 2013, and has been awful so far in 2014. I just use it as one tool in
my arsenal.
Remember: Past success doesn’t guarantee future success.
DVOA is good at exactly what it aims to do: Evaluating past performance
on both sides of the ball (they also break stats down into rushing and
passing for offense and defense).
DVOA is the best publicly available stat to evaluate how good a team
has performed so far on offense, defense, special teams and overall.
But realize: It doesn’t account for injuries, teams get better and worse, and
no stat is perfect. Use it as a major tool, but understand it’s just one piece of
the puzzle.
And as I wrote earlier, the stat gets more useful as the season progresses.
Late-season numbers are more helpful than early-season ones, because the
sample size is larger.
NFL seasons are very short, but by the end of the year, the sample size is
decent, because the amount of plays is fairly large (Drew Brees had 687
passing plays and 35 rushing plays in 2013). But after Drew Brees has
thrown 50 passes, you can’t put much stock in advanced metrics.
For the 90th time: Betting is hard. Even when stats tell the truth, they don’t
tell you that much. It’s like the Seinfeld “yada yada” episode. The Jets were
great on defense in the first three weeks of 2013… yada yada yada… you
lost money on them and they missed the playoffs because of their mediocre
pass defense.
Yards Per Play (ypp):
Yards per play is a pretty good indicator of an offense or defense’s
performance.
But game situations and strength of schedule affect its numbers, and
especially early in the year, it doesn’t tell you a ton.
Ypp also rewards passing teams, because a good pass gets more yards than
a good rush. That’s often fair, but it also means that a really good team that
runs often in garbage time might have deflated numbers, and a bad team
that’s always playing catch-up might have inflated numbers.
Always hunt for an exception. That’s often where you find value.
But yards per play is an infinitely better than, say, total yards.
Always break things down to one standardized unit. Not everyone gets the
same amount of plays in a game. So you can’t compare teams based on
yards in one game . But you can compare them based on yards in one play .
It’s a very good stat.
Use offensive yards per play, and defensive yards per play allowed in
conjunction with DVOA. If there are big discrepancies, try to figure out
why. It’s all a puzzle.
Yards Per Pass and Yards Per Carry
Breaking down yards per play further, you get yards per pass on offense,
and yards per pass allowed on defense.
They’re very good measures of pass offense and defense.
And of course, you can do the same with yards per carry.
But again, they’re skewed by context. Use them in conjunction with
DVOA.
Pace Stats
A major revelation for advanced basketball stats was making everything
pace-adjusted . If you watch and bet college basketball and don’t know
KenPom , slap yourself and go there immediately.
Who the hell cares if a team scores 120 points a game if they give up 125?
Isn’t an offense that scores 86 points a game actually better than one that
scores 96 per game… if the first team plays 25% slower and thus gets fewer
possessions and shots? (Yes, duh.)
Time of possession seems helpful for NFL, but it doesn’t mean much. As
always, we can take things one-step further to eliminate noise that clouds
insights.
Most casual fans see an over/under and either think: “Two good offenses,
bet over!” Or, “two good defenses, bet under!”
As always: You’re on the right track, but you have many other steps to
take. Most people stop after the first, most-obvious piece of analysis. That’s
how you lose money.
For example, the Chargers had the No. 2 offense in the NFL by DVOA in
2013. They were fifth in yards per play.
And their defense was 29th in yards per play allowed and 32nd in DVOA.
Great offense, putrid defense.
You’re probably thinking they went over in almost every game, or at least
that their over/unders (totals) were sky-high, right?
Nope. Just 5 of their 19 games went over the total. The average points
scored in their games was 43.26 points. Not especially high.
They were 14th in the NFL in points scored, despite having a top-5 offense,
by all metrics. And they were ninth in points allowed, despite having
arguably the worst defense in the league.
Situations like this are common. Don’t shrug your shoulders and move on.
If you want to win at sports betting, ask questions here, and find the
answers.
Why the hell were San Diego games relatively low-scoring?
It’s actually a simple answer.
They played really slow.
To use a conventional stat, they were second in the league in average time
of possession.
That tells you something , but it’s really just that they had the ball a lot. If
you watched them, you’d know they played slow… but how can we
quantify exactly how slow they played?
Fortunately, we can break time of possession down with Football Outsiders’
awesome “pace ” stat.
They were second-to-last in pace. The second-slowest offense in the league.
Time of possession didn’t lie in this instance.
When they played the high-scoring Broncos last year, my initial reaction,
similar to 99% of the world, was that it would be a shootout.
And if you looked at pace, you’d see that Denver was the fourth-quickest
team in the league. Hm. Clash of styles.
The two teams played three times last year. The scores: 47, 48 and 44.
Besides a weird playoff game against New England, those were the three
lowest scores in Denver games the ENTIRE year.
Denver played its THREE lowest-scoring regular-season games against the
worst defense and best offense it played. Crazy, right?
Well, coach Mike McCoy knew his defense wasn’t good enough to stop
Denver, so he took his already-slow offense and made it snail-slow. Peyton
Manning, good as he is, can’t score from the bench.
The Chargers held the ball for 35:27, 38:03 and 38:49, significantly higher
than their already-high average of 32:19 (2nd in the NFL) for the entire
season.
San Diego’s offense was good enough to dictate the pace when they were
on the field, so it limited the total points scored in all three games.
The totals in the three games were set at: 56, 56.5 and 55. Each game went
way under.
Betting on sports is hard. There aren’t glaring inefficiencies. But if you can
spot the small ones like this beforehand, well, that’s how you win.
Offensive Line Stats
“Offensive lines win championships.” – Gandhi’s uncle, Steve.
If you don’t think offensive lines are massively important, you don’t know
football.
But how the hell do you know if a team has a good offensive line?
The best way is probably to have elite scouting skills and to watch every
minute of game tape, taking notes. But you probably have a life, so:
I look at: running back success vs. natural ability, quarterback pressures vs.
time of release/WR ability, and Football Outsiders’ OL metrics .
For the running game, FO separates plays into: losses, 0-4 yards, 5-10 yards
and 11+ yards. According to FO, “the offensive line is 20% more
responsible for lost yardage than it is for yards gained up to four yards, but
50% less responsible for yardage gained from 5-10 yards, and not
responsible for yardage past that.”
They don’t count QB runs, unfortunately.
Also: They track “open field yards,” second level yards, the percentage of
“stuffed” runs at the line, and “power success,” which measure third and
fourth down running success, and do count QB runs.
These are all decent stats, but they’re definitely flawed. It’s really hard to
evaluate offensive lines with stats.
The creators admit the ratings “only somewhat separate the offensive line
from the running backs,” for run blocking. And for pass blocking, they
adjust sacks allowed for number of pass attempts, situations and opponents
faced… but they don’t include hits and hurries.
So I don’t put a ton of stock into either. But they’re valuable tools in the
arsenal.
Another good tool is ProFootballFocus’s offensive line grades, but you
must pay for access.
Red Zone Offense and Defense
TeamRankings has offensive and defensive Red Zone scoring percentages
(TD only), as well as appearances on offense and allowed on defense (not
pace-adjusted).
Third Down Offense and Defense
Same as above. Percentages and total attempts and successes (not pace-
adjusted).
Turnovers
Turnovers are amazing predictors of wins and losses – straight up and
against the spread.
Via Winafy :
“According to Spreadapedia (going back 10 seasons), the team that wins
the turnover battle is 79.9% SU and 79.0% ATS – marks a bit higher than
records of 78.9% SU and 77.6% over the database dating back 36+
seasons.”
The problem is: Turnovers are really random and hard to predict.
You can predict fumbles and interceptions to a degree, but the better way to
use turnover stats is to find unsustainable numbers that inflate or deflate a
team’s value.
But realize: People know this, so it may be factored into the line. Just
monitor it.
For example: The 2014 Giants started out -6 in turnovers the first two
games, going 0-2. Eli Manning threw four picks on 72 attempts, for a horrid
INT% of 5.55. He throws a lot of picks… but his career rate is 3.4%.
Add in the fact that their good defense forced ZERO turnovers, and it was
clear their results would improve as their luck evened out.
Still, the market soured on them, especially because they were considered
mediocre before the season.
The next game, they were +2 in turnovers. Then, +5! Then +1. They
covered the spread in all three.
Clearly, the first two games predicted nothing about the next three.
But equally important: This could mean they were over valued after that.
They likely won’t be +8 over many three-game stretches. Eli will always be
Eli.
Don’t be afraid to sour on a team that has been nice to you. Always be
skeptical.
Then again, their fumble recovery rate was still just 27.27%, well below the
expected ~50% number. (Defenses recover ~51% of fumbles, so numbers
like these even out over the long term, though single teams can certainly be
unlucky for a game, a few weeks or even an entire year or two.)
See? Question everything. You never know what you’ll find.
Monitor the market, and see if the lines make sense, all things considered .
If it seems like the market thinks the Giants suck, but the important,
sustainable pieces are in place, well, you might have an edge.
But as usual: You’re still gonna be wrong most of the time.
Turnovers are probably most useful to look at in the offseason to project the
following year. Via Pinnacle :
“Over the last 10 seasons, the top ranked teams for turnover margin were
ranked an average of 13 th in the same metric the following season (and had
2.5 fewer wins) and the bottom ranked teams averaged 14 th place (and had
3.5 more wins).”
Extremes usually even out.
Read the paper
I am counting this as a stat, because it’s my book.
Very simple, very underutilized.
Follow local beat writers. See what coaches and players are saying – sure,
most of it is canned bullshit – but occasionally you’ll hear about a
gameplan, an injury, how a team is reacting, if they’re out for revenge,
when an impressive backup is about to get his chance, etc.
Gems can be found if you look. Just don’t overreact to news.
Watch the games
Again, this counts. Shhh.
We all have cognitive biases. Our eyes lie to us.
If your eyes have one opinion and the stats say something else, generally
trust the stats. But do dig deeper – there’s a chance the stat is right, it’s just
including some noise or bad information that is skewing the insight it’s
suggesting.
You should question what your eyes see, but you shouldn’t ignore them.
You might realize that your favorite team struggles against a certain type of
opponent, for example. But don’t assume you’re right: Look for the stats to
back it up.
Drive Stats
Drive Stats are cool, too.
They aren’t adjusted for strength of schedule or situation, but they give
“total number of drives,” and average yards, point, touchdowns, punts,
turnovers, interceptions and fumbles lost, all on a per drive unit .
They also give starting field position (marked as LOS/Drive). That’s
incredibly useful, because it shows if an offense is being hindered or
boosted by its defense and special teams’ impacts on field position.
Example: Right now, before Week 6 of the 2014 NFL season, the Colts and
Chargers are middle-of-the-road in offensive DVOA despite putting up a
ton of points and being called elite offenses by the media.
Unsurprisingly, the Chargers are 1st in LOS/Drive and the Colts are 2nd .
Their offenses have received exceptional field position.
But their defenses are mediocre too. So are they just getting lucky? Does
this mean they’re due to start playing worse as luck evens out?
Don’t assume this is the case. Maybe it is. It’s a start. But always go one
level deeper.
Check special teams. Indy’s is pretty good (7th in DVOA). San Diego’s isn’t
great (17th .)

Check turnover margin. San Diego is 3rd . Indy is 12th .

It seems like those two stats explain both of their good fortune.
The next step would be… do we think this will persist? For special teams,
I’m honestly not sure. But for turnovers, it looks like San Diego has been
pretty damn lucky. They’ve recovered 83.33% of their fumbles(!) and
Philip Rivers has thrown an interception on just 1.21% of his throws,
compared to a 2.5% career rate.
This doesn’t mean that the Chargers are screwed, but it means they’re
probably going to stop having such good luck, and their real skill will be
revealed. It doesn’t mean they’re bad. But it means they’re probably worse
than they’ve looked.
The only trick in evaluating stats: Figure out exactly what they’re telling
you.
Don’t put words in their mouth, and realize they might just be yelling
meaningless gibberish.
Other Stats
Search around TeamRankings . They have gems. The ones above are the
main ones I use, but there are other great stats that help you determine
teams’ skills, styles, strengths and weaknesses.
What about outliers?
Sometimes you’ll come across weird teams.
We already discussed how San Diego had relatively low-scoring games
(and thus profitable unders) despite a great offense and horrible defense.
You had to keep digging beyond the surface, as always.
Always look for weird teams like that.
Like the 2013 Denver Broncos, who converted 72.15% of their red zone
opportunities into TDs. Early on, you’d probably hypothesize that was
unsustainable. But once you realized they had insane skill players, and
Peyton Fucking Manning , well, it meant they were nearly impossible to
stop in the red zone.
They got more out of every yard, and coupled with their quick pace, it was
clear that their margins of victory were going to be really high.
Their 13 regular season wins came by an average margin of 17.15. They
went 11-5 against the spread. And as I discussed earlier, they went 13-3 on
team totals.
This is easy to say in hindsight, but it’s something you may have seen
beforehand if you were paying close attention.
NHL examples: Ottawa Senators (ridiculous amount of first period OVERs
last year), and the New Jersey Devils, who play an odd style that keeps
scoring massively low, margins of victories small and opens some
interesting opportunities for and against.
Matchups
Mainstream analysis always discusses how teams “match up” against each
other.
Typically, the media puts too much emphasis on matchups.
But if you don’t think matchups matter, you’re clueless. The key is
determining which matchups matter, how much they matter, and if they
offer any value not seen by the market.
For example: I tend to downplay matchups except in extreme situations.
Please, test and research these things yourself, though. Ask questions and
look for evidence.
How does the best run defense do against elite run offenses? Do they cancel
each other out? Does the defense usually win? How does a team do when
forced to be one dimensional?
Try to script out the matchup, generally, make some assumptions, test them,
and hope for the best.
Betting is hard.
The topic of matchups probably needs its own book. I’ll get started.
Why can’t I look at individual player stats?
You should give a bit of weight to individual stats, but in a team game like
football, this usually leads to incorrect assumptions.
Quarterbacks are ridiculously important and a good one can overcome a lot
of other weaknesses. Peyton Manning would still be good on the Jets. But at
the same time, Geno Smith would probably look decent on the Broncos.
The point is: It’s almost impossible to evaluate players individually, since
they’re influenced by so many outside factors. This is what advanced stats
aim to do, but they’re much better at a team level in football. It’s a lot
different than baseball.
As FO puts it :
“Unfortunately, when it comes to individual player ratings, we are still far
from the point at which we can determine the value of a player independent
from the performance of his teammates.
That means when we say, ‘In 2013, Marshawn Lynch had a DVOA of
5.8%,’ what we are really saying is ‘In 2013, Marshawn Lynch, playing in
Darrell Revell’s offensive system with the Seattle offensive line blocking for
him and Russell Wilson selling the fake when necessary (and opening things
up with the pass), had a DVOA of 5.9%’ ”
A good rule of thumb is to assume that no player besides a very good
quarterback or an absolute superstar will make much of a difference in a
point spread. AJ Green was out last week, and the line moved about half a
point. QB or GTFO.
What matters beyond stats?
A team’s skill, obviously, isn’t the only thing that determines wins and
losses. Their motivation, health, mentality, travel, coaching, the weather…
all of it can affect matchups.
Is a team flying cross country? Are they on a short week? Are they coming
off a heartbreaking loss? What is a coach’s record against the spread off of a
bye? Does that seem to be factored into the line?
As I discussed earlier, a lot of trends are bullshit. But plenty aren’t.
Here’s an example of great analysis on weather . That’s how you should be
thinking.
Still… if a trend is telling you something, that doesn’t mean you should
blindly bet on it. It means you should consider it as one thing going in that
team’s favor, and see if it offers value against the current line.
The worst thing about trends: If people spot a good trend, by definition the
value will be sapped out of it.
If I said that teams off a bye play at a 25% better level than expected output
and proved it (I am making this up to make a point), then everyone would
know and the line would be shaded towards the other team.
Spreadapedia has an incredible database that allows you to research most of
these things. It costs money, but it’s a great investment if you’re serious
about betting.
Schedule Spots

Letdown games, “sandwich games” (in between two big games), rivalry
games, “must-win” games… all of these things exist.

But every team responds differently to them, and sometimes they’re not as
big of a deal as you think. And sometimes they’re a bigger deal than you
think.
These things are less important in the NFL than sports with longer seasons.
In NBA, NCAAB and NHL, these things are crucial, as are things like
back-to-back games.

It’s up to you to figure out how much it matters. But ignore these things at
your peril.

A tool like Spreadapedia for NFL is invaluable. You can come up with
things like this :

“Home dogs getting 7+, off a road loss in which they lost by 20+: 71-47-
2 ATS $ DENNYJ (verified via @ Spreadapedia )”
(If you were wondering, betting on the Jets, as suggested above, lost. But it
was a horrible beat. You can make a good bet and lose. It happens all the
time.)
What is home-field advantage worth?

Most people say home field advantage is worth three points. That’s a good
average estimate overall, but that’s all it is. Every team’s home-field
advantage is different.

The average fan underrates home-field advantage. How could “Insert Good
Team” lose to “Insert Bad Team?” They were on the road. It happens all of
the time.

Teams play better at home for psychological (comfort, routine) reasons, and
physical (less travel). It exists and if you don’t acknowledge it you will fail
miserably.

Track how often you bet on road favorites. For the average person, this is
their most common bet, because they can’t fathom how a good team would
lose to a bad team, even on the road. Don’t discount home-field advantage.

How much of NFL wins are luck?

We have established that wins and losses have little predictive value.
But how much are records impacted by luck, exactly?
Papa Sabermetrics Bill James created something called Pythagorean wins .
It sounds complicated, but it’s just a team’s predicted win-loss record based
on runs scored and allowed. It’s not perfect, but it’s way better than wins
and losses. It takes much of luck out of the equation.
The football version is more flawed, but again, it’s a big improvement on
wins and losses. It calculates a record based on points scored and allowed.
I’m not a huge fan of it in football, but it does is tell you two important
things:
1) Teams with a lot of close wins are lucky.
2) Teams with a lot of close losses are unlucky.
3) Teams that blow teams out are better than their record.
It’s like your fantasy league: You know that team that made the playoffs
despite not scoring a ton of points? Their record counts, but it doesn’t mean
their team is good.
If you want to look at wins and losses, look at Pythagorean wins and losses.
Or look at records, but account for close wins and losses. The difference
between a close win and loss is usually just some luck.
The goal of most advanced stats is to evaluate skill, by removing luck and
randomness from the equation.
Wins = skill + opponent skill + game situation + luck/randomness
The fantastic Brian Burke of Advanced NFL Stats explains luck in football
perfectly :
“When we watch a football game, we’re only seeing a tiny sample of what
could happen in a game that, in theory, could go on forever. That is, if
players didn’t get too injured or tired, we could let the teams play on and on
until we were absolutely certain one team is better than the other. But, of
course, we can’t do that. Instead, we’re treated to an arbitrary 60-minute
sample of everything that could happen between the two teams, and that’s
not nearly long enough to definitively know which one is better.”
“When we statistically compare the real N.F.L. to coin flips, only about
half of all games are determined by the relative strength of each
opponent. ”

This means a National Coin Flip League would “have almost as many
winning streaks, losing streaks and late-season drama… epic collapses and
big upsets.”

But “it’s human nature to force explanations on things that sometimes

neither have nor require an explanation.” We convince ourselves that skill
explains everything.

When are people wrong about luck?

There was a time when people could argue that the Patriots were really
lucky and “due for regression” or something like that.
Basically, they were outperforming market expectation.
Except that it has persisted for years. Bill Belichick has a career 56.2%
record against the spread, and that’s counting his pre-Pats days.
You could have blindly bet on Belichick every game his entire career, with
no analysis at all, and you would have made money.
There are exceptions to everything. They’re just really hard to find
beforehand, and when you find something in hindsight, there’s no guarantee
the trend will continue.
What I’m saying is the value on Belichick might be gone. Or it might not.
The same arguments have been made for Sir Alex Ferguson, Gregg
Popovich and even some quarterbacks (Andrew Luck so far).
They’re just one (major) factor in their teams’ success, though, so this could
be due to external factors. Is Bill Belichick great because of his coaching?
Because of Brady? Is it the combo? The offensive line? Playing the Jets
twice a year? Anal beads?
We’ll never know, but the narrative will persist.
Strength of schedule
Most people don’t understand strength of schedule.
Before every season, people cite SOS based on the previous year’s records,
ignoring player turnover, but more importantly… ignoring the fact that wins
and losses are useless and records change vastly from year-to-year.
If you want to project a team’s schedule strength preseason, use Super Bowl
odds and team win total over/unders.
As you now know, records mean nothing. Especially last year’s records.
Football Outsiders Busts Myths

Read Football Outsiders’ “basics” page . It busts a lot of common myths

about the NFL. I’ll summarize the most important takeaways below.

“Establishing the run” does nothing.

“There is no correlation whatsoever between giving your running backs a

lot of carries early in the game and winning the game.”
“A great defense against the run is nothing without a good pass
defense.”
“With rare exceptions, teams win or lose with the passing game more than
the running game – and by stopping the passing game more than the
running game. The reason why teams need a strong run defense in the
playoffs is not to shut the run down early. It’s to keep the other team from
icing the clock if they get a lead. You can’t mount a comeback if you can’t
stop the run. Note that ‘good pass defense’ may mean ‘good pass rush’
rather than ‘good defensive backs’.”
Stats are very dependent on strength of schedule.
“In 2007, Tony Romo and Derek Anderson had similar numbers, but Romo
faced a much tougher schedule than Anderson did. Romo was better that
year, and better in the long run.”
Consistent running backs are better than boom-or-bust backs (if their
yards per carry are equal).
Getting stuffed a lot puts quarterbacks in long third-down situations. Bad.
Long gains are good, but not when mixed with lots of stuffs, which leads to
punts, which should be looked at as turnovers (but with better field
position).
Also, consistent running backs can eat more clock late in
games.
“Rushing is more dependent on the offensive line than people realize,
but pass protection is more dependent on the quarterback himself than
people realize.”
“There are certainly good running backs who suffer because their offensive
lines cannot create consistent holes. Most boom-and-bust running backs,
however, contribute to their own problems by hesitating behind the line
whenever the hole is unclear, looking for the home run instead of charging
forward for the four-yard gain that keeps the offense moving.”
For passing, moving in and out of the pocket, throwing the ball away and
getting rid of the ball quickly, due to good pass-rush instincts, is key. Peyton
Manning never gets sacked and his offensive line loves him.
“In fact, a scrambling quarterback will often take more sacks than a pocket
quarterback, because while he’s running around trying to make something
happen, a defensive lineman will catch up with him.”
(Although he of course creates positive plays running and throwing on the
run.)
“Shotgun formations are generally more efficient than formations with
the quarterback under center.”
But defenses might adjust. Everything is fluid.
Wide receivers must be judged on both complete and incomplete
passes.
Wide receiver catch rates are consistent and important… but also depend on
types of routes. Like, Wes Welker and Darren Sproles are gonna have high
catch rates.
FO has a metric called “plus/minus” that measures catches vs. expected
catches, adjusting for difficulty.
Offense and defense are worth the same amount. Special teams is worth
1/3 of one of them.
Also: “Offense is more consistent from year to year than defense, and
offensive performance is easier to project than defensive performance.
Special teams is less consistent than either.”
“Field-goal percentage is almost entirely random from season to
season, while kickoff distance is one of the most consistent statistics in
football.”
So if a team is getting shitty field goal kicking, chalk (much) of it up to
luck. Especially since they’ll probably cut that asshole.
Teams with more offensive penalties generally lose more games, but
there is no correlation between defensive penalties and losses.”
“One reason is that defensive penalties often represent good play, not bad.
Cornerbacks who play tight coverage may be just on the edge of a penalty
on most plays, only occasionally earning a flag.”
“The penalty that correlates highest with losses is the false start, and the
penalty that teams will have called most consistently from year to year is
also the false start.”
Fumble recoveries are random, as are blocked kicks and touchdowns
on turnover returns.
“These plays are not ‘lucky,’ per se, but they have no value whatsoever for
predicting future performance.”
As Mike Tanier explains :
“Here is the basic version: If your defense forces lots of fumbles, that’s a
skill. If your offense avoids fumbles, that’s a skill. But recovering the
fumbles is not a skill; rather, it's simply a matter of whether an oblong
object wobbled into the hands of “a defender, wedged itself in the belly of a
guard or bounced out of bounds.
Some people have a problem with teams being called “lucky.” If you’d like,
replace “luck” with “random” or “benefitting from events that hold no
future predictive value.” It’s all the same.
Field position is crucial.
Defensive stops are important, but so is where the defense stops a team.
“The most underrated aspect of an NFL team’s performance is the field
position gained or lost on kickoffs and punts.”
What I’m saying is Devin Hester is more valuable than Tom Brady.
(KIDDING! But Hester and his ilk are underrated.)
Red-zone performance is crucial, but inconsistent.
Generally, good teams are good in the red zone, etc.
(But as I discussed earlier, the Broncos appear to be an exception. Always
look for exceptions.)
Staying healthy is important, but it’s luck.
This one is especially interesting.
According to FO
“Teams with a high number of injuries are a good bet to improve the
following season. However, while injury totals tend to regress towards the
mean, there’s also no doubt that certain teams have a record of staying
healthier than others.”
So there may be exceptions to this rule. This goes along with my entire
betting theory. There are exceptions to every rule, and that’s where value is
found. If you assume that every team performing better than expected is
“lucky” and “due to regress,” you may just be missing something.
Some teams may have better training staffs, nutrition, or whatever.
ALWAYS HUNT FOR EXCEPTIONS.
Depth > stars and scrubs.
Depth is key. People get hurt.
Running backs decline after 28, TEs after 29, WRs after 30, QBs after
32.
Again, there are exceptions. But this is the average.
For non-skill players, FO’s preliminary research suggests the ages of
decline for are: DBs 29, LBs and OL 30 and DT 31.
College football teams are actually much more consistent from year to
year than NFL teams.
This is because recruiting is usually consistent. This isn’t surprising, as the
NFL is built for parity, and NCAAF is not.
“Championship teams are generally defined by their ability to
dominate inferior teams not their ability to win close games.”
“Football games are often decided by just one or two plays – a missed field
goal, a bouncing fumble, the subjective spot of an official on fourth-and-1.
One missed assignment by a cornerback, or one slightly askew pass that
bounces off a receiver’s hands and into those of a defensive back five yards
away and the game could be over. In a blowout, however, one lucky bounce
isn’t going to change things.”
Step 5: Defeat Common Biases
Don’t judge your bet by if it wins or loses.

If you wouldn’t judge a team by 16 wins and losses, why would you judge
your decisions based on one or two or 16 scores?

Don’t grade single bets. All that matters is if you’re consistently beating the
closing line and exploiting small inefficiencies.

You’ll have “bad beats,” where your bet “deserved” to win, but lost on
something fluky. And you’ll get lucky on bad bets. The other team will
“backdoor cover.” Or “frontdoor.” You’ll be on both sides of luck.

Watch the games, and grade your performance over long periods of time.
Look at your bet tracking and find if there are any areas where you’re
struggling.

“How often did I beat the closing line?” is a much better question than
“how often did I win over an insignificant sample size?”

Only worry about your record long-term .

The Cognitive Biases That Kill

Almost every mistake the average losing bettor makes can be explained by
a well-known cognitive bias.

Regardless of your intelligence, you WILL fall for these things if you aren’t
aware of them. I have many times, and I still do, when I don’t keep myself
in check.

These sneaky bastards steal your money without you even knowing.

Recency Bias
I discussed recency bias earlier, so I’ll keep this brief.

ONE GAME MEANS NOTHING. NEVER MAKE MORE THAN A

SMALL JUDGMENT BASED ON ONE RESULT.

Every team plays over its head for a game or two or three. Every team plays
badly for a game or two or three. In every sport.

If a team pulls a massive upset, they’d still be huge underdogs if there were
a rematch the next day.

Real-life example: “Bro, that was the hottest girl I’ve ever seen!” Bro, it’s
just the most-recent hot girl you have seen.

Gambler’s Fallacy (Law of Averages)

I constantly hear that a team is “due” for a good or bad performance.

Statistically, this is complete nonsense. It’s called the “gambler’s fallacy,”
or the “law of averages.”

In sports terms, it’s that Barry Bonds is gonna hit a home run today because
he hasn’t in 20 games, or that Eli Manning can’t possibly have another poor
performance, or Indiana can’t possibly have another poor shooting game at
home.

As sabermetrics/advanced stats/#fancystats/analytics gain mainstream

acceptance, writers are constantly falling for the gambler’s fallacy, thinking
that it’s “analytics.”

In baseball, a player has a high BABIP so he’s “due to regress” and get
fewer hits.

In hockey, a team has good possession numbers (Corsi/Fenwick), but keeps

losing, so they’re “due to start winning.”

In football, a team hasn’t been recovering fumbles and is losing, so they’re

due for better luck and to start winning.

There’s an important thing missing here, though.

Just because a team has had bad luck recently doesn’t mean they’re
“due” for good luck. What it means is that you should extract their good
fortune when evaluating their past performance.

And when predicting future performance, assume they’ll have normal luck.
But they aren’t “due” for anything. They still have the same chance of bad
luck as they did in the game before.

Over the long-run, luck tends to even out. But over the short run, nobody is
“due” for anything. There’s no way to predict when the good or bad luck
will come.

If the Giants had really bad fumble luck in Weeks 1-5, it means that they are
probably better than their record. It doesn’t mean they’ll start recovering
everything.

Again, past bad luck does not mean good luck is going to happen short
term .

Real-life example: It rained the past three days… it can’t possibly rain
today!

Bandwagon Effect

Everyone like the hot team.

Real-life example: Everyone hates Justin Bieber. Conformists!

Confirmation Bias

As I showed with the Seahawks vs. Potato Skins trends, it’s incredibly easy
to find a statistic or piece of evidence to support an opinion you already
hold.

Confirmation bias means that you instinctively look for stats and evidence
that confirm what you think. Instead, actively try to prove yourself wrong.
If you really can’t, then you might have a reasonable opinion.
Real-life example: A mother has a demon-child, and her friends tell her
this when he steals $1,000 from the family safe. But she instinctively looks
for excuses. “He probably got kidnapped!” No, stupid mother! You are
falling prey to confirmation bias!

Framing Effect

Always be aware of how statistics are framed.

With trends, for one, whoever compiled the stat usually stopped at an
arbitrary number to make his point.

The Seahawks are 4-1 in their last 5 games against the spread.

That sounds great! But they probably didn’t cover six games ago. And they
might not have covered the game before that.

So if I told you they were 4-3 in their last 7, would you really want to bet on
them?

Technically, framing uses the exact same stat and presents it in a different
way (usually the negative), but this version is much more common in
sports.

As always: Look beyond the stat.

Real-life example: “Doctor X saved two lives.” Or “Three of Doctor X’s

five patients died.”

Same information, opposite interpretations.

Hot-Hand Fallacy

Kind of like the gambler’s fallacy. “I’ve won my last 8 bets, so I’m hot!”
No, you’re lucky and should do everything exactly the same. Conversely,
you’re not “due” for a cold streak.

Real-life example: It’s most prevalent in gambling, or basketball, when a

player is hot takes more shots due to cockiness.
Another stupid example would be a lucky roulette player upping his bets
because he’s “hot.” The mistake is confusing good fortune with being “hot.”

Insensitivity to Small Sample Sizes

Billy Volek plays well for four games, and all of a sudden people think he’s
Joe Montana.

Then he goes back to being Billy Volek.

Real-life example: The stock market went way down this month. Are we
all doomed?!

Sunk Cost Fallacy

This is more of a fantasy football problem.

“I picked Doug Martin in the first round, I can’t trade him for this fifth
rounder who is clearly better!”

It’s also why Trent Richardson is still in the NFL.

Real-world example: “ I hate this shirt, but I paid good money for it, so
I’m gonna wear it.”

Ostrich Effect

Ignoring potential catastrophe. “Yeah, I can bet my whole bankroll on the

Seahawks at home,” they’ve got an 85% chance of victory! You silly,
bankrupt man.

Real-world effect: “Health insurance, schmealth insurance!”

Overconfidence Effect (Optimism Bias)

Everybody thinks they can win at sports betting. Almost nobody actually
does.
You almost certainly think your winning percentage is higher than in
actuality.

Real-world example: “I’ve got a shot at winning the lottery. Why not?”

Loss Aversion

Most people are afraid to bet underdogs on the moneyline, even if there’s
value. You know there’s greater than a 50% chance of losing, and people
hate losing money short term.

People “prefer avoiding losses to acquiring gains.”

Real-world example: People will get pissed and change cable companies if
their rate gets raised 5%, but they won’t go through the trouble of switching
if they find a company that saves them 5%.

Hindsight Effect

I totally knew they were good! (Then why didn’t you bet on them?)

Real-world example: You “knew” your girlfriend would cheat on you.

Then why did you go out with her in the first place?

The Bias Blind Spot

Nobody thinks they do these things. But you do, assuming you are a human.
(If not, I’m not sure how this book got to your planet, but leave an honest
review!)

Falling for these doesn’t mean you’re dumb. It means you’re human.

The Narrative Fallacy

“Looking backward and creating a pattern to fit events and constructing a

story that explains what happened along with what caused it to happen.”
This is where the media is at its worst.
It’s human instinct to come up with a story to explain everything. But in
sports, the answer often is: “Shit happened.”
A lot of the time, “choking,” or “revenge” or a “gutsy comeback,” or
“clutch finish” is just luck. Close games happen, and then shit happens.
But that would make a boring story week after week.
Think about it: What happens to every team, every year?

They play at their best for a few games. Their worst a few others. They get
lucky and unlucky. Sometimes they’re more lucky one year, or play better,
or get injured, blah blah blah.

A loss almost always signals panic. How much lower can they go? Will
Coach X be fired? Is this one loss to a decent team THE END OF A
DYNASTY? A win means a team might hit the next level – are the
Jets/Browns/insert bad franchise here turning things around… are the
Cowboys/Eagles/insert mediocre franchise here ready to make a Super
Bowl run… are the Seahawks/Broncos/Patriots insert successful franchise
here “the team to beat?”
Final Notes
NFL betting is the hardest. If you only cared about profits, you’d focus on a
smaller market. You’d bet NHL, specialize on a small NCAAB conference,
or even bet something like WNBA. I could probably clean up on Czech
hockey if I wanted to.
But you’re likely doing this for fun, as am I. So, realize that there’s a good
chance you’ll still lose, even with all of this knowledge.
But now you’re ahead of 90% of people, so now you’ve got a shot in hell.
Godspeed, and please email me at [email protected] if you ever have
any questions.
And please follow me on Twitter .
Further Reading: Conquering Risk, by Elihu Festel
If you want to get to the next level, read this book. It involves lots of math.
If you’re not willing to math it up, there’s not much you can do from here,
besides devouring game tape.
Twitter List to Follow
One of the biggest problems for the average bettor is: Where the hell do I
find good information?
I’ve given you the best stats sites: Football Outsiders and TeamRankings.
Also check out ProFootballFocus and ProFootball reference. Follow local
beat writers for news and quotes (but realize most analysis is narrative BS),
and search for the best national writers who have a clue.
They usually aren’t the most popular ones.
I curate all of my news through Twitter, where dumb analysis is
everywhere, but if you follow the right people, you create a feed of constant
value.
I put together a list of the best Twitter follows. Enjoy. Just click and follow
along . I’m all about limiting noise. Get the helpful information and stats
you need, and no more.
Online Resources: Winafy , TeamRankings , FootballOutsiders ,
ProFootballFocus , OddsShark , NitrogenSports , BetOnline, 5Dimes,
PinnacleSports, Spreadapedia .
I think I need some sort of disclaimer here, because gambling is bad blah
blah blah.
I don’t accept or place bets or promote any form of prohibited online
gaming. This book is for entertainment purposes only! Were you not
entertained?
 This le was downloaded from Z-Library project

Your gateway to knowledge and culture. Accessible for everyone.

z-library.se singlelogin.re go-to-zlibrary.se single-login.ru

O cial Telegram channel

Z-Access

https://wikipedia.org/wiki/Z-Library
ffi
fi