Bankroll Management: Why Most Punters Go Broke (And You Will Not)

Imagine two punters. Both find +3% expected value on every bet they place. Both bet on the same matches at the same prices. After 12 months, one has doubled their starting bankroll. The other is broke and explaining to their partner that the betting was meant to be a hobby.

The difference is not skill. It is stake sizing. Variance, the natural swing of luck around your expected outcome, will bury anyone who bets too big regardless of how good their edge is. This guide is about the boring, unsexy maths that keeps you in the game long enough for your edge to actually pay out.

What a bankroll actually is

Your bankroll is the pool of money you have set aside specifically for betting. It is not the money in your everyday account. It is not money you need for rent. It is a ringfenced amount, ideally in a separate bank account or bookmaker balance, that you have mentally written off the moment you funded it.

The reason for the mental writeoff is psychological. If your bankroll is also your rent money, you will make terrible decisions when it dips. You will skip +EV bets because the variance scares you, then chase losses with oversized stakes when you fall behind. A ringfenced bankroll lets you bet the maths instead of betting your emotions.

A reasonable starting bankroll for an Australian punter trying to actually beat the books is between $500 and $2,000. Below $500, the unit sizes become so small that the maths still works but the dollar amounts feel meaningless and you stop caring. Above $2,000, you are just starting bigger, which is fine if you can afford it.

The unit: your basic bet size

The simplest staking system is flat unit betting. Pick a percentage of your bankroll (typically 1% to 3%) and that becomes your "unit". Every bet you place is one unit, regardless of how confident you feel.

For a $1,000 bankroll at 2% unit size:

• 1 unit = $20
• A normal bet = $20
• A high-confidence bet = $20
• A "lock of the century" = $20

The reason every bet is the same size is that you are not actually that good at distinguishing your 55% confidence bets from your 60% confidence bets. Almost no punter is. By forcing equal stakes, you remove the worst source of variance in your results: your own emotional sizing.

If you are running pure arbitrage (no probability estimation, just locked-in profit), you can bet larger units, up to 5% of bankroll, because the variance is essentially zero. If you are betting +EV bets where your probabilities are guesses, stay at 1 to 2%.

Kelly criterion: when sizing actually matters

For bettors with a defined edge and a reliable probability estimate, there is a mathematical formula for the optimal stake. It is called the Kelly criterion, and it answers the question "how much of my bankroll should I bet to maximise long-term growth without going broke".

The formula:

kelly stake = (probability × decimal odds - 1) / (decimal odds - 1)

Worked example. Suppose you believe Sydney Roosters have a 55% chance of winning, and your bookie has them at $2.10:

• (0.55 × 2.10 - 1) / (2.10 - 1)
• (1.155 - 1) / 1.10
• 0.155 / 1.10
• 0.141, or 14.1% of bankroll

Full Kelly says bet 14.1% of your bankroll on this match. On a $1,000 bankroll, that is $141. Compared to your 2% flat unit ($20), Kelly is telling you this bet is worth seven times your normal stake.

Here is the problem. Kelly assumes your probability is exactly correct. If your 55% estimate is actually 50% in reality, Kelly will systematically over-stake and you will go broke faster than a flat bettor. Real punters are almost never as good at probability estimation as they think they are.

For a +EV bet with a 5% edge, quarter Kelly will typically suggest a stake of 1.5% to 2.5% of bankroll, which is almost identical to flat 2% unit betting. The two systems converge for typical edge sizes. The difference only matters when you find rare large-edge bets, where Kelly correctly tells you to lean in harder.

Variance: what your bankroll has to survive

Even with a genuine +3% edge, your bankroll will go through brutal drawdowns. The maths is unavoidable. Some sample statistics for a bettor placing one bet per day at average odds of $2.00 with a 3% edge:

The lesson is not that you will lose money. The lesson is that you will frequently feel like you are losing money even when the long-term trajectory is up and to the right. Your bankroll has to be large enough, and your stake small enough, that a 20% drawdown is survivable both financially and emotionally.

Practical bankroll rules

A simple ruleset that works for the overwhelming majority of Australian punters:

1. Set a starting bankroll. Write it down. Never quietly top up.
2. Bet 1 to 2% of starting bankroll per +EV bet, up to 5% for pure arbs.
3. Recompute your unit size every month, based on the new bankroll. If you have grown to $1,500, your 2% unit is now $30, not $20. If you have dropped to $700, your 2% unit is $14.
4. Never bet more than 5% of bankroll on a single match, no matter how confident you feel or what Kelly says.
5. Keep a simple log: date, match, stake, odds, result, profit/loss. A spreadsheet is enough. Without a log you cannot tell whether you are actually winning or just remembering the wins.
6. Review monthly, not daily. Variance dominates daily results. Monthly is the shortest window where edge starts to show through.

What to do next

Bankroll management is the foundation that lets every other strategy on this site actually work. Without it, even a 5% edge will eventually run you broke through variance. With it, even a modest 2% edge compounds into real money over a few thousand bets.

Once your bankroll is set and your unit size is defined, the next move is to make sure every bet you place is at the best available price. That is the easiest piece of edge in betting, and it is what the line shopping guide is about. After that, the +EV guide covers how to find bets that pay above fair value.