Sample size
Sample size is the number of settled results a performance figure is based on.
In plain English
Betting outcomes are noisy enough that small samples routinely produce impressive-looking records by luck alone. Hundreds of settled bets are generally needed before a modest edge can be distinguished from variance.
Example
A 20 percent ROI over 15 bets is roughly meaningless. The same figure over 800 bets would be remarkable.
Why it matters
Sample size is what determines whether a record is evidence or an anecdote, and it is the first thing worth asking about any published number.
How V8 uses this term
V8 shows its settled-pick count next to its performance figures and states plainly when the sample is still early rather than implying significance it has not earned.