EDIT – Individual images: I’ve decided to keep this page for the mile stone updates, making collages. If you want to see the individual images, please see here LINK
I bought this broken down, split spine, board separated, sorry state AD&D 1e Monster Manual:
The idea is that I’m going to colour my way through it. A bit of meditative self time. At the moment I’m doing A to Z, but I may just mix it up as I fancy.
I’m going to use coloured pencils as I don’t want colour to bleed through the pages. As you will probably spot, I’m not an artist! I should probably work out how to take better photos!
Call it sacrilege, but I figure this book can have a second life, albeit under my amateur hand.
I’ll update this post as I meander along.
EDIT – I’m not necessarily following the colours mentioned in the description of the MM (I’ve only got so many brown pencils), I’m just going with whatever I feel like on the day. It’s also fun just to reimagine the monsters. D&D is a fantasy game amIright?
Collages: A to B, C to D (except Demons, Devils, Dinos and Dragons), Demons, Devils, Dragons, E, Giants, Golems, G (other), H, I to K, …, L (Lycanthrope), L(other), M, N, O, P Q, R (cont…)
Ever wanted to play a table top war game (like Warhammer Fantasy Battle) solo, that is against an army commanded by A I General? Well, this is attempt to do that, using a Hex Flower as that A I General
The idea is that you can use this Hex Flower as an analogue A I for solo (fantasy) battles – that is, you play your army and you use the A I to move the opponents units in the battle.
The idea is that the A I can surprise you, whereas if you play the other army yourself, you can never suspend disbelief and so totally surprise yourself.
The A I and especially the indicative actions are there to help guide the units’ choices.
This A I has been ‘flavoured’ for Orcs. But, believe it or not, I looked to Sun Tzu’s art of war for inspiration, and grouped actions into the four categories. So, you can repurpose this for any army (reject the flavour and embrace the essence of the AI’s instructions). I might yet make a similar product for other standard army types … Humans, dwarves, elves, ratmen, undead etc.
Historical Note – way-back-when in the G+ days, when Hex Flowers were still a thing that we were brainstorming about, I think this ‘A I General’ was first suggested by G+ user Neuzd: IMAGE. I think Neuzd had a lot of good ideas about HF, but with G+ gone it is hard to remember all of it!
Hex Flower background – what’s a Hex Flower anywho? A Hex Flower (HF) is like a random table, but with a memory.
Hex Flower Cookbook – where I discuss the theory and design of Hex Flower Game Engines and some background and possible uses. But, you can also read some more about Hex Flowers on my Blog: LINK
This Hex Flower Game Engine is a homage to the classic arcade game Pac-Man. In this case ‘Hac-man’ runs the maze eating pellets and being chased by Ghosts. Can you get past Level 3?
I got sucked into making this Hex Flower when I read that the my Hex Flower ‘edge rules’ were a bit like the ‘warp tunnels’ in the classic arcade game Pac-Man.
It’s just a bit of fun – try it out and let me know what you think?
Important: This work is not affiliated or endorsed in any way by the owners of Pac-Man.
The game is pay-what-you-want so you can download as a PDF for free (see link above), but you can make a donation to support my work more generally.
Layout:
Hex Flower background – what’s a Hex Flower anywho? A Hex Flower (HF) is like a random table, but with a memory.
Hex Flower Cookbook – where I discuss the theory and design of Hex Flower Game Engines and some background and possible uses. But, you can also read some more about Hex Flowers on my Blog: LINK
Adding a minotaur ‘hunter’ mini to the Hex Flower maze & pursuit gave the kids more buy in!
PS – they made it to the center of the maze, got the prize (the gilded mushroom; which was siting on an island surrounded by a lake of mercury) and out again, evading the minotaur using: a bag of fine chalk (checking for drafts that tend to blow out of the maze), a bag of rats (to distract the minotaur when it got close) and a magic butterfly that helps guided anyone lost.
Mechanically, these items allowed the party to change the direction of travel on the HF by one face (or in case of the rats, the minotaur’s direction of travel by one face to help distract/evade the ‘hunter’).
People use the word ‘terse‘ a lot when describing RPG content/adventures with concise content. I think they mean this in a positive way, but to me ‘terse’ has a negative quality – abrupt to a fault. I prefer the term ‘pithy‘.
I’m hardly a RPG publishing powerhouse – I’ve released 2 adventures and neither has exactly broken any records. Perhaps having Hex Flowers as a central mechanic limits them to a niche within a niche (at least that’s what I’m telling myself).
However, dusting off any self-doubt, I have something to say about words. Too many words dilutes a message.
“The more you say, the more you put on a page, the less important all those pieces become“ Between two Cairns, Warped Beyond Recognition episode, release date 6 November 2025; 23:28-23:34
May I also suggest you listen to Tod from As If Productions, who had something to say about overwriting:
* = I have no financial connection with Tod, although before Patreon screwed up all my subscriptions, I did Patreonize Tod.
My personal learning about overwriting came from when I submitted some monsters for Ford’s Faeries as coordinated by Eric Nieudan of The Merry Mushmen fame. The idea was to prepare a faerie type monster inspired by the art of Henry Justice Ford, but using no more than 300 words. My first try ended up being about 1500 words. I cut and cut until I got down to 300 words. I was fairly shocked, the result was 5 times shorter but I think about 5 times better.
Sure, I accept that there comes a point where you cut down to the bone and are left with only a bone and no juice. I don’t think most people get that far. Editing is hard, self editing is even harder.
My suggestion is cut your first completed draft in half and see if it is improved. Then again, what do I know!
Now, to practice what I preach …
“If I had more time, I would have written a shorter letter” – Attributed to Blaise Pascal
The OSR/NSR (and splinter movements thereof) fly high the flag and associated benefits of random tables and procedurally led outcomes.
I’m fully behind this idea and the benefits.
That said, when listening to one of Fear of a Black Dragon or Between Two Cairns podcasts, one of the commentators said something that resonated with me (if I were a betting henchperson, I would say it was Brad Kerr).
They said something along the lines of: “Yeah, these random tables are neat and all, but sometimes I just want an answer“.
This podcast episode must be before or about 2023, because I tried to incorporate this idea into my procedural adventure Carapace:
(check it out – the video is from a now ended Kickstarter project)
My suggestion to anyone interested in procedural adventure design, is that when preparing procedural tables that have at least 3 outputs/parameters that stack/build a result – include some pre-generated outcomes.
Sometimes a GM will be under some time pressure and so just want a simple fast pre-generated answer. Why not give them that as an option.
Think of this as the adventure/dungeon equivalent of a pre-generated character.
I’m sure this idea is probably out there, but I just want to draw some (extra) emphasis to it.
You might not know, but I have a YouTube channel. It’s nothing to shout about, but it is there. Here are some actual plays if you are interested in that kind of thing:
Kids Play Mausritter (4 sessions, two adventures)– On hold
Kids Play AD&D 1e (2 sessions) – Ongoing
Beast Hunt Play Test AD&D 1e (6 Sessions) – Ongoing
The game is pay-what-you-want so you can download it for free (see link above), or you can make a donation to support the idea.
If the idea gets much traction, I might fancy it up a bit more, but for now here it is:
Intro:
There is a 1 Player Mode and a 2 Player Mode.
In the 1 Player Mode you are a ‘Maze-Runner’ trying to get to the Hex-Chasers base to destroy it. The Hex-Chasers are controlled by an AI and are trying to stop you.
To move around the Hex Maze, you as the Maze-Runner need to figure out the passcode – a times table answer. In your side of the Hex Maze (your territory) you need to give the correct answer taken from the 1 to 6 times tables. In the Hex-Chasers territory it gets harder and you need an answer from the 7 to 12 times table.
The 2 Player mode, is the same as the 1 Player Mode, but in this case Player 2 is in control of the Hex-Chasers and is trying to capture Player 1. Like Player 1, the Hex-Chaser needs to give the correct answer to a times table question before they can move.
I saw this interesting post by Joshy McCroo (a sort of psychic card guessing resolution mechanic) and this cool looking crowdfunding project by Aaron A. Reed (Downcrawl 2e).
This got me wondering if you could make a card type oracle with a “memory”. That is, the last card effects the result of the next card.
I’ve never thought of this before but my Hex Flower Game Engines are game oracles of sort, just often with specific applications in mind.
Set up (let’s engineer a probability profile):
Draw an initial card (just to get things going, or use the 7 of Hearts).
Draw the next card and compare this card with the current card.
In this order, determine if there is a match:
Number match (6% or about 1 in 20), if not is there a
Suit match (17% or about 3 in 20), if not is there a
Colour match (31% or about 6 in 20, if not
No match (46% or about 9 in 20)
Draw a new card and compare this with the card drawn last time. Rinse and repeat.
A. Let’s use this to make a simple YES/NO oracle:
Rules:
If the new card is EVEN = is a YES result, if the new card is ODD = is a NO result
Number match = Extreme Result
Suit match = Result + AND
Colour match = Result + BUT
No match = Bare Result
So, if the current card is 6 of Clubs and the next card is (this represents 7 possible options):
1 6 of Diamonds (EVEN* card with a NUMBER match), then that would be an Extreme YES
2 4 of Clubs (EVEN card with a SUIT match), then that would be a YES, AND
3 7 of Clubs (ODD card with a SUIT match), then that would be a NO, AND
4 5 of Spades (ODD card with COLOUR match), then that would be a NO, BUT
5 8 of Spades (EVEN card with COLOUR match), then that would be a YES, BUT
6 2 of Hearts (EVEN card no match), then that would be a (bare or very marginal) YES
7 9 of Diamonds (ODD card no match), then that would be a (bare or very marginal) NO
* = if you don’t like the extreme event being YES or NO linked directly from the start card, flip the drawn card, if it lands face up = Extreme YES, if face down = Extreme NO.
B. Let’s make another simple YES/NO oracle that has more of a “memory”
So far it could be argued that there is no true “memory” going on here, the memory as such is just using the last result to engender a probability structure into 52 cards split into 4 suits of two colours etc.
Also, if you think that life should not be an even 50/50 split of good/bad (and that things should tend to regress to the mean), then instead of simply considering whether the new card is ODD or EVEN, you could consider whether the new card is HIGHER or LOWER than the current card.
In this way the next card (and so the result) is influenced by the last card.
New Rules:
If the new card is HIGHER (or equal*) = is a YES result, if the new card is LOWER = is a NO result
Number match = Extreme Result
Colour match = Result + BUT
No match = Bare Result
So, if the current card is 6 of Clubs and the next card is (this represents 7 possible options):
1 6 of Diamonds (SAME* card number with a NUMBER match), then that would be an Extreme YES
2 4 of Clubs (LOWER card with a SUIT match), then that would be a NO, AND
3 7 of Clubs (HIGHER card with a SUIT match), then that would be a YES, AND
4 5 of Spades (LOWER card with COLOUR match), then that would be a NO, BUT
5 8 of Spades (HIGHER card with COLOUR match), then that would be a YES, BUT
6 2 of Hearts (LOWER card no match), then that would be a (bare or very marginal) NO
7 9 of Diamonds (HIGHER card no match), then that would be a (bare or very marginal) YES
* = if you don’t like the extreme event always being YES or NO linked directly from the start card, flip the card, if it lands face up = Extreme YES, if face down = Extreme NO.
In this system it is hard to have a lucky/unlucky streak (like in real life) as you have to continually be above (or below) the previous card. On the down side, you can guess when your luck is likely to change … holding an ace or a 2 is a sign that things are going to change! I suppose if you are holding the top or bottom 3 cards (Ace, King, Queen, 2, 3 or 4) you could get a re-shuffle and re-draw to set the base card again?
C. Let’s try for a “Change in Circumstance” engine (rather than a YES/NO oracle)
New Rules:
If the new card is HIGHER (or equal*) = is a POSITIVE Change in Circumstances, if the new card is LOWER = is a NEGATIVE Change in Circumstances (CiC)
Number match = EXTREME CiC
Suit match = LARGE CiC
Colour match = MODERATE CiC
No match = NO CiC
So, if the current card is 6 of Clubs and the next card is (this represents 7 possible options):
1 6 of Diamonds (SAME* card number with a NUMBER match), then that would be an Extreme POSITIVE change in circumstances
2 4 of Clubs (LOWER card with a SUIT match), then that would be a LARGE NEGATIVE CiC
3 7 of Clubs (HIGHER card with a SUIT match), then that would be a LARGE POSITIVE CiC
4 5 of Spades (LOWER card with COLOUR match), then that would be a MODERATE NEGATIVE CiC
5 8 of Spades (HIGHER card with COLOUR match), then that would be a MODERATE POSITIVE CiC
6 2 of Hearts (LOWER card no match), then that would be a NO CiC
7 9 of Diamonds (HIGHER card no match), then that would be a NO CiC
* = if you don’t like the extreme CiC always being YES or NO linked directly from the start card, flip the card, if it lands face up = Extreme YES, if face down = Extreme NO.
D. Let’s try for a “Chaos Factor” engine (rather than a YES/NO oracle)
New Rules:
If the new card is HIGHER (or equal*) = is an INCREASE in Chaos Factor, if the new card is LOWER = is an DECREASE in Chaos Factor (CF)
Number match = 2 step change in CF
Suit match = 1 step change in CF
Colour match = 1 step change in CF (this system has a broader middle)
No match = NO Change in CF
So, if the current card is 6 of Clubs and the next card is (this represents 7 possible options):
1 6 of Diamonds (SAME* card number with a NUMBER match), then that would be an +2 CF
2 4 of Clubs (LOWER card with a SUIT match), then that would be a -1 CF
3 7 of Clubs (HIGHER card with a SUIT match), then that would be a +1 CF
4 5 of Spades (LOWER card with COLOUR match), then that would be a -1 CF
5 8 of Spades (HIGHER card with COLOUR match), then that would be +1 CF
6 2 of Hearts (LOWER card no match), then that would be a NO change in CF
7 9 of Diamonds (HIGHER card no match), then that would be a NO change in CF
* = if you don’t like the extreme event always being an INCREASE or DECREASE linked directly from the start card, flip the card, if it lands face up = Extreme POSITIVE change in CF e.g. +2CF, if face down = Extreme NEGATIVE change, e.g. -2 CF.
Other options
These are my first impressions, I feel more could be done. Cards have so many options – two packs, mixed packs, what if numbers are adjacent, face cards are special, reshuffle on a joker … and bespoke cards.
I really want to think about bespoke/custom cards like the cards in Downcrawl 2e, which like geomorphic terrain cards can have set edges that are used to align events. I think a custom pack with rules would have the best chance of having more of a true memory effect.
Summary
Once you understand the probability structure and the potential “memory” effect with this card system you can build up other Oracles. Better still make your own probability structure and rules
What have I missed? Did I get this wrong? Has this been done before (probably)?
For ‘that Reddit user‘: Sure, please keep using the method you like, I’m not the D&D police
It’s been a little while since I made a Hex Flower, so I thought I’d make a new one.
I’ve seen quite a lot of webposts looking for a good mechanic for mass combat, this is a Hex Flower version which I hope might be of interest to some.
Mass Combat Victory Tracker Hex Flower – this Hex Flower is intended to be a reasonably light weight minigame to track a victory condition in a mass combat.
The idea is that after the PCs conclude their personal combat round within the wider mass combat, an army moral determination is taken, but where the PCs actions can tip the scale for victory (or otherwise) and where there is also a weighting for the relative strength of the respective armies (i.e. the army power modifier):
There are some rules and an example page too (see the PWYW PDF to read this at better resolution):
This is particularly pleasing because it’s Pay What You Want (PWYW), so despite being essentially free, at least 1001 people wanted to give me at least 20cents to say thanks.
I was recently asked about interesting ways to generate D&D type stats (more below) and that got me thinking about the possibility to generate these using a Hex Flower. Tricky to get this to work with a 19 HEX flower to replicate (or at least ape) the probability profile of standard dice methods. Also, which probability profile to follow 3D6 or 4D6 drop the lowest (4D6DtL)?
Anyway, this is what I cooked up:
Here are some examples of the three suggested methods:
How well does it work? I think well enough, but use at your peril:
So, it’s mainly for fun, but again, results will not be as consistent as a dice method, but it is more like a fun little minigame you can try. Overall, I think you will get more high/low rolls as compared to a dice method (i.e. more min-maxish), but these STATs shooooould even out-ish (well on average).
The PDF ( here ) is self-contained, goes into the data a bit more, and can be used as a template.
Other fun (non-conventional) ways to generate stats :: Rubik’s Cube Method – LINK :: Caterpillar Method (pseudo self-balancing method) – LINK :: Bar/Grid Flip (pseudo self-balancing method; not mine) – LINK
Hex Flower background – what’s a Hex Flower anywho? A Hex Flower (HF) is like a random table, but with a memory.
Hex Flower Cookbook – where I discuss Hex Flower Game Engines and some background and possible uses
I love the idea of mazes. They captured my imagination when I first encountered the story of the minotaur
The problem is, mazes are not very interesting in RPG / D&D like games.
Well, that is not quite true – mechanically mazes are not interesting to explore because of the language barrier; that is DM & players needing to communicate the precise spatial arrangements of the walls would be excruciating (especially if the walls are not all right angles).
Here’s a map from Against the Cult of the Reptile God – it’s not even a very big maze, but I wouldn’t want to run it in a hurry:
Really, someone needs to make Doom-like 3rd person shooter like geomorph cards:
Then players could explore more intuitively without the need for language – but setting up and indexing the geomorph cards to make a maze would be a bind for the DM (maybe it has been done?).
“Maze concept module”
So, anyway, back in 2017 I first wrote a ‘concept module’ about exploring labyrinths, first called “Into the Hive” and released on G+; it got redrafted in 2018, and revamped and Kickstarted in 2023 as “Carapace” (now with fancy art – linked at the bottom!) .
Essentially, the ‘concept module’ gave 3 x ways to explore a maze:
I won’t dwell on the exact methods used, except to say the ‘Labyrinth Move’ by Jason Cordova was the genesis of my ‘concept module’.
“Big Idea”
At the heart of these 3 x methods was MAPLESS exploration that was EPISODAL.
The ‘big idea’ was that in a finite period of exploration (whatever that time was in your game – 10 minutes, an hour etc.) the party would progress in the maze (or not), and they might have an encounter (or not) in that ‘episode’ of play.
Rinse + repeat until the party was dead or escaped the maze – simples.
Other examples
(i) Gardens of Ynn
This ‘big idea’ of mapless episodal exploration was also encapsulated in Garden of Ynn (GoY); the idea here is you go “deeper” into the maze-like garden or you retreat out.
So again, there is no map, and each move in the maze is episodal.
I don’t own a copy of Labyrinth by Ben Milton (perhaps I should), but his video review appears to show that exploration in his Labyrinth also follows this ‘big idea’:
So, again, exploration is mapless and episodal. It also has an interesting Fighting Fantasy type vibe (and that also reminds me of this Mindstorm Ladder mechanic). The mechanic in this case presumes you are trying to get to the centre of the maze, so has a one-way ‘going deeper’ vibe. Also, like Carapace it has defined zones to track progress in the maze and to give each zone its own flavour.
(iii) Flux Space
I was recently listening to Ben Laurence’s podcast about Mega Dungeons (episode 5), and you should listen to this podcast too, and became aware of Nick LS Whelan’s “Flux space” idea.
Again, this method neatly obviates the use of a map to explore a maze and again embraces episodal exploration. Take a look.
Summary
Carapace, GoY, Labyrinth and Flux Space all ditch the physical maze layout as uninteresting, in all cases there is abstracted progress through a maze which is episodal.
–> Go make your own Abstracted Maze Exploration Tool today!!
Please tell me more other neat ways of exploring mazes (someone must have done a Jenga Block or deck of cards based version ?!)
Oh, get your copy of Carapace today, it comes it two stripes FREE and FANCY (cover below):
Recently I been thinking about functional entries in random tables.
In my ‘In the Heart of Oz’ (a system neutral) sandbox, based in the land of Oz, I wanted to add some random tables but wanted these to “work hard” for the limited space available.
So, I wondered about replacing single entry random tables with functional entry random tables. For example instead of ‘sword’ you use ‘bladed weapon’ and let the user decide if it is a sword or an axe or a ‘bec de corbin’.
This is a pretty trite example but of course you can take this further (see below).
I’m sure this idea/concept has happened in the past, but when I have done this (at least), it’s been more an unconscious design choice rather than a conscious design choice. I’d like to hold this idea up as a conscious design choice for the DM’s toolbox.
Benefit – gives a much bigger spectrum of outcomes
Downside – require the user to bring more of themselves to the table (which might be the very thing they are trying to avoid by using random tables!).
Some might like the idea, but might displease others … I’m not the RPG police so please feel free to ignore this idea.
The idea is to do this on the fly (or pre-prepare some), but also source the table – throw this at players – Tell me what item you find based on these 3 constraints … etc
Below is an text only extract from the above linked PDF:
Random Creature Builder
Roll once on each of the two tables below:
Base Creature
1 Apex predator
2 Large herbivore or domestic animal
3 Monkey, ape or lemur
4 Rodent or marsupial
5 Amphibian or reptile
6 Bird or dinosaur
7 Insect or arachnid
8 Crustacean or echinoderm
9 Gastropod or cephalopod
10 Human(oid) – or you decide
Creatures are normally at least man-sized
Twist
1 Roll on table above – hybrid of both e.g. a Human(oid)-crustacean
2 Unusual locomotion: Fly, hover, burrow, crawl, creep, jump etc.
3 Oversized/mutantous body part or extra parts – limbs, mouth, eyes etc.
4 Unusual sized e.g. giant, long etc.
5 Exotic colouring or display
6 Made of artificial material
7 Mushroom, vegetable creature, or made of natural material
8 Roll on ‘Terrain descriptor’ table
9 Roll on this table twice – has both
10 Roll again on this and on the below table
Special
1 Offensive: looks, smell, charisma etc
2 Insubstantial: ghostly, ethereal etc.
3 Resembles man-made object, can mimic things, or is a doppelganger
6 Contradictory property of a typical creature of the ‘Base’ creature class
7 Teleport locally, or is transparent
8 Ability to petrify – gaze, touch etc.
9 Ability to cast magic or illusions
10 Roll on this table again – has both
– – –
More info:
In the Heart of (the Land of the Wizard of) Oz is a system-neutral procedural sandbox setting which draws inspiration from the books of L. Frank Baum and with artwork by the wonderful Nate Treme (Highland Paranormal Society)
In the Heart of Oz contains procedural Encounter, Terrain and Weather Engines for Wilderness Hex Crawling across the fairy-tale like land of Oz. There is also a procedural ‘dungeon’ to take on the Wicked Witch of the West in her Yellow Castle.
Preview image
Hex Flowers … what? A Hex Flower is a bit like a random table, but with an inbuilt ‘memory’ (because the last turn affects the next outcome). Each time you enter a new HEX, you procedurally determine what is going on.
Printing This document can be conveniently printed using the ‘booklet’ option on printers that can print on both sides of the paper. Folding gives an A5 booklet.
I’m extremely grateful to Tito B.A. for translating my ‘Hex Flower Cookbook’ (Manual do “Hex Flower”) into Portuguese and so making it more accessible to more gamers!
I will donate all money to help Ukrainian resist Putin. I’ll will also personally match the amount donated.
– – –
More info:
In the Heart of (the Land of the Wizard of) Oz is a system-neutral procedural sandbox setting which draws inspiration from the books of L. Frank Baum and with artwork by the wonderful Nate Treme (Highland Paranormal Society)
In the Heart of Oz contains procedural Encounter, Terrain and Weather Engines for Wilderness Hex Crawling across the fairy-tale like land of Oz. There is also a procedural ‘dungeon’ to take on the Wicked Witch of the West in her Yellow Castle.
Preview image
Hex Flowers … what? A Hex Flower is a bit like a random table, but with an inbuilt ‘memory’ (because the last turn affects the next outcome). Each time you enter a new HEX, you procedurally determine what is going on.
Printing This document can be conveniently printed using the ‘booklet’ option on printers that can print on both sides of the paper. Folding gives an A5 booklet.
I was listening to Ray Otus’s podcast (which I like). He was talking about Wolfram’s Four colour theorem, which states (perhaps paraphrased as):
“Any map in a plane can be colored using four-colors in such a way that regions sharing a common boundary (other than a single point) do not share the same color”
I wondered if this could be used to map out relationships where each colour represents a certain intensity in that relationship from high to low. The idea being to design a relationship map that is random, but with some “design” to it.
Let’s try it out …
Example 1 – D&D starting characters – using Wolfram’s Four colour theorem
Maybe this design concept could be applied to D&D stats, where certain stats tend to go together … :
Rules:
Populate a 7 hex Hex Flower with each stat and one for wealth, so STR, INT, WIS, DEX, CON, CHAR & wealth
Using up to 4 pencils colour/outline the Hexes using Wolfram’s Four colour theorem, starting from the middle
Randomly assign each colour a different value from 1 to 4 using a D4 where:
Step 1 – populate the Hex Flower with the 6 stats and wealth:
Step 2 – colour according to Wolfram’s Four colour theorem (in this case using all 4 colours)
Step 3 – randomly assign each colour a different value from 1 to 4:
Red 2 (- ) e.g. 5 days’ living wage Orange 4 (++) i.e. 15-18 Green 1 (- -) i.e. 6-9 Blue 3 (+) i.e. 12-15
So the result is:
So the new PC is strong and tough (15-18), quite wise and personable (12-15), clumsy and dumb (6-9) and has modest wealth. So perhaps an fighter then from a poor background.
Let’s try it again (using the same stat placement), but with fewer colours more of a-go-for-broke min-max arrangement) and re-rolling:
Red 3 (+) e.g. 3 months’ living wage Orange 1 (- -) i.e. 6-9 Blue 4 (++) i.e. 15-19 (no green)
So the new PC is very smart, wise, personable (15-18), but very weak, fragile and clumsy (6-9). But quite wealthy. Hrmmm, perhaps a Magic-user then for an affluent family.
If this sort of min-max 3-colour-option option does not appeal, then you could get rid of the central hex (and get rid of wealth slot) and insist that all four colours are used.
Example 2 – Random NPC’s personality
Rules:
Populate a 7 hex Hex Flower with a personality trait, one from each group (you can make your own traits)
Using up to 4 pencils colour/outline the Hexes using Wolfram’s Four colour theorem, starting from the middle
Randomly assign each colour a different value from 1 to 4 using a D4 where:
Step 1 – populate the Hex Flower with the 7 traits:
Step 2 – colour according to Wolfram’s Four colour theorem (in this case using all 4 colours)
Step 3 – randomly assign each colour a different value from 1 to 4:
Red 4 (++) Orange 3 (+) Green 1 (- -) Blue 2 (-)
So the result is:
So, they really desire something from the PCs, they are quite wealthy and greedy, they are not especially smart or personable, they are of low social status but at least not very treacherous. A quest giver, perhaps a merchant, who desperately wants something to (further) enrich themselves and is happy to pay the PCs to get it.
What happens if we had coloured the hex’s differently, using just 3 colours e.g.:
Now they would have some desire to get something from the PCs, they would be smart, greedy and treacherous with no money, status and quite cold. Doesn’t sound like a nice chap. Perhaps a street thief full of animal cunning and wants the PCs’ coin.
Again, three colours gives a more polarized outcome.
Some initial conclusions – I think you can try and set this up so coloured pairs go together, perhaps wealth & status, greed & duplicity and wit & warmth.
That said, I wonder if we should scrap the middle hex and insist that 4 colours must be used to introduce less homogeneity. Perhaps.
Is this useful (?)– not sure, just brainstorming here. Perhaps if you planning a session and want some inspiration about an NPC, or have a bunch of templates ready to pull out of a hat??
More Clearly, these are not just the two ways this idea could be used.
<– (in progress)
Get the higher resolution PWYW PDF 
Goblin's Henchman 