The difference between a bettor who lasts a season and one who does not is almost never picking. It is bet size.
The difference between a bettor who lasts a season and one who does not is almost never picking. It is bet size. Someone with a genuine edge betting too large goes broke; someone with no edge betting small stays entertained for a year.
A bankroll is money set aside specifically for betting, separate from everything else you own. It is not your account balance, and it is not what you could bet if you moved some things around.
Every other idea in this lesson is a percentage of that number. If the number is wrong, the percentages do not help.
A unit is one standard bet, expressed as a share of the bankroll. One per cent is the common choice: a $2,000 bankroll makes a unit $20.
Two reasons to think in units rather than dollars. It keeps bet size proportional as the bankroll moves, and it makes a record comparable — "+12 units" means the same thing whatever the money behind it, which "+$240" does not.
| Unit | Bets to go broke at a 50% rate | Suits |
|---|---|---|
| 1% | Very unlikely within a season | Almost everyone |
| 2% | Survivable, with real swings | A well-tested edge |
| 5% | A 20-bet losing run ends it | Nobody, honestly |
| 10%+ | A bad weekend ends it | Not betting — gambling |
Flat betting means staking the same amount on every bet regardless of how you feel about it. It is the simplest discipline there is and the baseline any other staking plan has to beat.
It is also more defensible than it looks. Varying stake by confidence only helps if your confidence is calibrated — if the bets you feel best about really do win more often. Most people's confidence is not calibrated, and they find that out expensively.
There is a principled way to vary stake by edge, and it has a lesson of its own next. The honest framing is that Kelly is flat betting's better-informed cousin, not its replacement.
A drawdown is the distance from a bankroll's peak to its lowest point after it. Everyone has them. Betting at a 2% edge with 1% units, a 15-unit drawdown across a season is completely ordinary.
Chasing — raising stakes to win back what you have lost — does not improve the odds of the next bet by a single percentage point. It converts a bad month into a serious problem faster than any other behaviour in betting. If you find yourself sizing bets by how far down you are, stop for the day. If that is hard, the resources linked at the foot of every page on this site are there for exactly this.
The rule that prevents most of it: set the unit size before the session, and do not change it during one.
Log bets in units and watch the bankroll curve, not just the win-loss record. Open the Bet Tracker.