SELECT COUNT(tmp.[GroupID])
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement)
HAVING COUNT(tmp.[GroupID]) > 3;
-- Returns one row containing: 4


-- Remove HAVING and nothing changes:
SELECT COUNT(tmp.[GroupID])
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement);
-- Returns one row containing: 4


-- Change HAVING predicate and it still sees a single, aggregated group:
SELECT COUNT(tmp.[GroupID])
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement)
HAVING SUM(tmp.[Measurement]) = 14.3;
-- Returns one row containing: 4


-- Add in an empty GROUP BY and nothing changes:
SELECT COUNT(tmp.[GroupID])
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement)
GROUP BY ()
HAVING COUNT(tmp.[GroupID]) > 3;
-- 4

SELECT COUNT(tmp.[GroupID])
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement)
HAVING tmp.[GroupID] > 3;
/*
Msg 8121, Level 16, State 1, Line 36
Column 'tmp.GroupID' is invalid in the HAVING clause because it is not contained in either an aggregate function or the GROUP BY clause.
*/


SELECT tmp.[GroupID]
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement)
HAVING COUNT(tmp.[GroupID]) > 3;
/*
Msg 8120, Level 16, State 1, Line 44
Column 'tmp.GroupID' is invalid in the select list because it is not contained in either an aggregate function or the GROUP BY clause.
*/


SELECT tmp.[GroupID]
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement)
HAVING tmp.[GroupID] > 3;
/*
Msg 8121, Level 16, State 1, Line 55
Column 'tmp.GroupID' is invalid in the HAVING clause because it is not contained in either an aggregate function or the GROUP BY clause.
*/

SELECT COUNT(tmp.[GroupID])
FROM   (VALUES (1, 2.2), (1, 3.5), (2, 5.1), (7, 3.5)) tmp(GroupID, Measurement)
HAVING 999 > 998;
-- 4