A 3% house edge sounds like a rounding error next to the 5.26% on double-zero roulette. Run it across 500 bets at $10 a hand, though, and you're looking at an expected loss of roughly $150 — more than the price of a decent dinner, gone not in one bad beat but in 500 small, boring increments. The edge doesn't need to be dramatic to do damage. It just needs repetition.

Why 500 bets is the number that matters

Most players fixate on a single session. A hundred bucks in, a hundred bucks out, call it a night. But the math that actually governs your bankroll is cumulative, and 500 bets is roughly one solid month of casual play for someone logging 15–20 hands a night, four or five nights a week.

At $10 flat, you're putting $5,000 into action. A 3% edge means the casino expects to keep $150 of it. That's the theoretical number, not a guarantee — variance can swing you $400 up or $600 down over that stretch depending on the game. But the drift points one direction, and it points there every time.

Where 3% shows up

  • Blackjack with basic strategy: about 0.5% — you'd lose $25 over 500 hands
  • Baccarat banker bet: 1.06% — about $53
  • Three-card poker (ante only): 3.37% — about $168
  • Slot machines at 96% RTP: 4% — about $200
  • American roulette: 5.26% — about $263

Notice the pattern. The gap between 1% and 3% triples your expected loss over the same 500 bets. The gap between 3% and 5% nearly doubles it again.

The compounding illusion

Here's where players get tripped up. Over 50 bets, a 3% edge produces an expected loss of just $15. That feels negligible. Over 500 bets, it's $150. Over 5,000 bets — a year of steady play — it's $1,500.

The edge doesn't grow. Your exposure does. This is why "I'm up $80 on the night" stories are real and also meaningless. Any single session can beat the house edge. Five hundred sessions cannot.

A concrete anchor: the American Gaming Association reported commercial casino gaming revenue of $66.5 billion in 2023. That number isn't magic. It's the aggregate of millions of players each running their own 500-bet experiment and losing 1–6% of handle on average, with a few lucky outliers subsidized by the rest.

What this means for your bankroll

If you're playing a 3% game and buying in for $300, you have roughly 100 bets at $3 flat before expected loss eats your stack — assuming average variance. If you're betting $10, you're mathematically underwater after about 30 bets.

That doesn't mean don't play. It means size your bets against the edge, not against your mood. A 3% game at $5 a hand over 500 bets costs $75 expected. The same game at $25 a hand costs $375. Same edge, same 500 bets, five times the bleed.

Wagering requirements on bonuses work this exact way. A $100 bonus with a 30x rollover means $3,000 in handle. At 3% house edge, you're expected to lose $90 working through it — which is why so many bonus offers net out close to zero for the average player.

So the real question isn't whether 3% is beatable in a session. It obviously is. The question is whether you know which game you're actually playing before you sit down for the 500th hand.