Sample size determines whether your statistics are meaningful or noise. Knowing when a stat is reliable prevents costly mistakes.
What Is Sample Size?
Sample size is the number of hands or situations you’ve observed. The more data, the more reliable the stat.
When Stats Become Reliable
VPIP/PFR: ~1,000 hands (basic tendencies emerge)
3-Bet %: ~3,000-5,000 hands (less frequent action needs more data)
Fold to C-Bet: ~500-1,000 hands (fairly frequent situation)
Win rate: ~50,000+ hands (high variance requires massive samples)
The Danger of Small Samples
With 200 hands on an opponent, a 35% VPIP could easily be a 22% VPIP player running hot. Don’t adjust your strategy based on unreliable data.
Practical Guidelines
Against unknown opponents: use population tendencies, not individual stats
After 500 hands: start using VPIP/PFR to categorize (fish vs reg)
After 2,000+ hands: trust most stats and make exploitative adjustments
Sample Size and Your Own Stats
Your own win rate needs at least 50,000 hands to be meaningful. Don’t celebrate or despair based on 5,000 hands.
