Here's an uncomfortable fact about sports betting: the thing most likely to end your season isn't a bad read, a bad beat, or a bad week. It's bet sizing.
Two bettors can make the exact same bets, hit the exact same cold streak, and finish in completely different places — one annoyed, one broke — because of a single decision each of them made before the season started.
Most bettors never make that decision at all. Their bet size is whatever felt right in the moment, which means it's biggest exactly when their judgment is worst.
Bankroll management is the fix, and it's the least glamorous skill in betting: no picks, no reads, no highlights. It's also the only part of betting that's entirely in your control. This post covers the whole system — what a bankroll is, what a unit is, how big it should be, and the math of surviving the losing streak that is absolutely coming for you.
What a Bankroll Actually Is
Your bankroll is money set aside for betting that you can lose — all of it — without it touching your rent, your groceries, your savings goals, or your sleep.
That's the definition, and every word is load-bearing. It's not "money in my sportsbook account." It's not "whatever's in checking." It's a fixed amount, decided in advance, sitting in its own mental (or literal) account.
If losing the whole number would be a real problem in your life, the number is too big. There is no bankroll management technique that fixes a bankroll you couldn't afford to lose — and if no number feels affordable right now, betting isn't the hobby for this season of your life. That's not a footnote; it's the entry requirement for everything below.
What a Unit Is (and Why Everyone Uses Them)
A unit is your standard bet size, expressed as a percentage of your bankroll. When bettors say "I put two units on it," they're describing how much they risked relative to their own roll.
That's why units are the universal language of betting: $50 might be a huge bet for one person and a rounding error for another, but "one unit" means the same level of conviction for both.
The widely used convention, and the one we teach, is one unit = 1% to 2% of your bankroll. On a $1,000 bankroll, that's $10 to $20 per standard bet.
If your first reaction is "that's tiny" — hold that thought. The chart at the top of this post is the answer to it.
The Cold Streak Is Coming. Here's the Math.
Losing streaks aren't a sign your process broke. They're a mathematical certainty of betting in volume.
We ran the numbers: over a 500-bet season, a bettor winning half their bets has a better-than-60% chance of hitting eight losses in a row at some point, and roughly a 1-in-5 chance of hitting ten straight. Not might. Will, more often than not. The streak isn't the emergency — it's the weather.
So the real question of bet sizing is: when the streak arrives, what does it do to you? That's what the chart above shows — a 15-loss streak, deliberately drawn as the nightmare scenario, hitting three bettors with the same $1,000 bankroll and flat stakes:
- $10 bets (1% of bankroll): after 15 straight losses, $850 left. Annoying. Survivable. Still betting.
- $50 bets (5%): $250 left — three-quarters of the roll gone on a streak that involved zero bad decisions.
- $100 bets (10%): broke at bet ten. Season over, and the streak wasn't even finished.
Same bets. Same streak. The only difference is size.
And notice what the 1% bettor still has that the others don't: the ability to be there when the streak ends. Every edge in betting — real or imagined — requires you to still be solvent when the variance swings back. Sizing isn't about winning more. It's about making sure the math gets the chance to play out.
Flat Stakes: The Boring System That Works
Flat staking means every standard bet is the same size — one unit — regardless of how confident you feel, how last night went, or how loud the game is.
Its entire power is what it removes. The bettor who bets 3x "to get it back" after a loss, or doubles up because they're "locked in" after three wins, has handed their sizing decisions to the exact emotional states that produce the worst bets. Flat stakes make your worst moment cost the same as your best one.
A reasonable refinement is a small conviction range — say, one unit standard, two units for your strongest spots — as long as the range is defined before the season and the top of it stays rare. If your "max conviction" bet is showing up three times a week, it's not conviction; it's mood.
You'll also run into the Kelly criterion — a formula that sizes bets to your estimated edge. It's real math with a real problem: it requires knowing your edge precisely, and overestimating it (which nearly everyone does) makes Kelly aggressively oversize your bets. For almost every bettor, flat 1–2% does what Kelly promises with none of the blowup risk.
Resize on a Schedule, Not on a Feeling
Your unit shouldn't be frozen forever — it should be recalculated on a schedule. Pick a checkpoint (monthly, or a natural break like the bye weeks) and re-set your unit to 1–2% of your current bankroll.
Roll grew? Unit grows a little. Roll shrank? Unit shrinks — which quietly builds in the survival property from the chart above, because your bets get smaller as your bankroll does.
The rule that makes this work: never resize mid-streak, in either direction. Resizing after a bad Sunday is chasing with extra steps. Resizing after a hot week is the same emotion wearing a nicer outfit. The whole point of a schedule is that the calendar decides, not the scoreboard.
What Sizing Can't Do
One honest limit, because this post would be dishonest without it: bankroll management doesn't create an edge.
If your bets lose money on average, perfect sizing just means you lose it slowly — the vig math we laid out here doesn't care how disciplined your stakes are. Sizing is the part of the system that keeps you alive; finding genuinely good numbers is a separate skill, and it's the one everything else in DimesVault is built around. You need both. But you need this one first, because without it you don't survive long enough for the other one to matter.
And if what's actually driving your bet sizes is the need to win money back — if the number keeps growing because it has to — that's not a sizing problem, and no calculator fixes it. We wrote an honest guide on when to walk away, and the National Council on Problem Gambling's helpline — 1-800-GAMBLER — is free, confidential, and answers 24/7.
Frequently Asked Questions
How much of my bankroll should I bet per bet?
The standard convention is 1–2% per bet. On a $1,000 bankroll, that's $10–$20. It will feel small — that feeling is exactly why most bettors go broke on streaks that a 1% bettor shrugs off.
What is a unit in sports betting?
A unit is your standard bet size expressed as a percentage of your bankroll — the universal way bettors describe risk relative to their own roll. One unit typically equals 1–2% of bankroll.
How big does my bankroll need to be?
Whatever amount you could lose entirely without it affecting your life. There's no minimum — a disciplined $200 bankroll with $2–4 units is a better setup than an oversized roll you can't actually afford. For the full starting numbers — and what a $200 first month actually looks like — see how much money you need to start sports betting.
Should I increase my bet size when I'm winning?
Not mid-streak. Recalculate your unit on a fixed schedule (monthly or at season checkpoints) from your current bankroll. Scheduled resizing captures growth; emotional resizing — in either direction — is how discipline leaks.
Do losing streaks mean my strategy is broken?
Not by themselves. Over a 500-bet season, a bettor winning half their bets will more often than not hit an 8-game losing streak, and has about a 1-in-5 chance of hitting ten straight. Streaks are variance; sizing determines whether you survive them.