Baseball Betting Bankroll Management: Unit Sizing for a 162-Game Grind
Available in US
I blew my first baseball bankroll in seventeen days. Seventeen days — out of a six-month season. The bets were not bad; the sizing was catastrophic. I was staking 8-10% of my bankroll on games I liked, which meant a five-game losing streak — completely normal in baseball — wiped out nearly half my funds. By the time the approach started to show positive results, there was nothing left to bet with. The edge was real. The bankroll management was not.
Baseball is a 162-game marathon per team, producing over 2,430 games across a regular season. Even a profitable strategy will endure double-digit losing streaks multiple times. Without a bankroll system designed for that volume and variance, the best analysis in the world is worthless.
The Unit System: How to Define and Size a Betting Unit
A unit is simply a fixed percentage of your total bankroll assigned to a single bet. The concept exists to separate the emotional question of “how much do I want to bet?” from the mathematical question of “how much should I bet?”
The standard recommendation in baseball is 1-3% of your bankroll per unit. If your bankroll is 1,000 pounds, one unit is 10 to 30 pounds. A standard play gets one unit. A strong play — where your edge is larger than usual — might get two units. Your absolute maximum should never exceed three units on a single game, regardless of conviction.
Why 1-3%? Because the math protects you from ruin. At 2% per unit, you could lose 25 consecutive bets before your bankroll halves. A 25-game losing streak on well-selected bets is virtually impossible — the probability is astronomically low. Even a 15-game streak, while painful, leaves you with 74% of your bankroll intact and fully capable of recovering. At 5% per unit, a 15-game streak drops you to 46%. At 10%, you are effectively finished after eight or nine losses in a row — a run that happens to sharp bettors multiple times per season.
I set my unit at 2% for standard plays and allow up to 3% for what I call “A-grade” spots — games where my projected probability diverges from the market by more than 5%. Below that threshold, every play gets the standard unit. This removes ego from the equation. I do not need to feel brilliant about a bet to size it correctly; I need to measure the edge and let the system decide.
One practical point: your unit size is based on your current bankroll, not your starting bankroll. If you begin the season with 1,000 and your balance grows to 1,300 by June, your unit increases proportionally. If your balance drops to 800, your unit decreases. This dynamic recalibration — sometimes called the percentage-of-bankroll method — prevents you from over-betting during drawdowns and allows you to capitalise on growth.
Flat Staking vs Variable Staking: Pros, Cons, and When to Use Each
The debate between flat staking (every bet gets the same unit size) and variable staking (sizing up on stronger plays) has occupied betting forums for years. After eleven seasons, I have used both and settled on a hybrid.
Flat staking is simpler, more disciplined, and harder to abuse. Every bet gets one unit. No exceptions. The advantage is that it eliminates the risk of over-sizing on a “lock” that turns out to be a loss. I have seen bettors use variable staking as an excuse to chase — tagging a recovery bet as a “5-unit max play” to win back yesterday’s losses. Flat staking makes that impossible.
Variable staking, when executed properly, captures more value from your best spots. If your projected edge on Game A is +3% and your edge on Game B is +8%, it makes mathematical sense to bet more on Game B. The projected return per pound wagered is higher. But “executed properly” is the key phrase. You need an honest, quantified assessment of your edge — not a gut feeling. If your edge measurement is even slightly miscalibrated, variable staking amplifies the error.
My hybrid approach: I use flat staking as the default (1 unit per bet) and allow a step-up to 2 units only when my model’s projected probability exceeds the market’s implied probability by at least 5 percentage points. That threshold is high enough to filter out marginal edges and low enough to capture genuine value spots two or three times per week. I never go above 2 units. The marginal gain from a 3-unit bet is not worth the marginal risk of being wrong.
Kelly Criterion in Baseball Betting: Theory vs Practical Application
Every serious bettor encounters the Kelly Criterion eventually. The formula calculates the mathematically optimal bet size based on your edge and the odds offered. In theory, Kelly maximises long-term bankroll growth. In practice, it is dangerous in its pure form.
The Kelly formula is: stake = (bp — q) / b, where b is the decimal odds minus 1, p is your estimated true probability of winning, and q is 1 minus p. If you estimate a 55% chance of winning a bet priced at 2.00 (even money), Kelly recommends staking 10% of your bankroll. That is aggressive — far more than the 1-3% range I recommend.
The problem is that Kelly assumes your probability estimate is perfectly accurate. It is not. No one’s is. If your true edge is 3% but you believe it is 8%, Kelly will tell you to overbet dramatically. The consequences of overestimating your edge compound across hundreds of bets over a season. One study I read estimated that using half-Kelly — staking half of what the formula recommends — captures roughly 75% of the theoretical growth while dramatically reducing drawdown risk.
I use Kelly as a sanity check, not as a staking system. If Kelly suggests 6% of bankroll on a bet and my standard approach says 2%, I do not split the difference — I stick with 2%. But if Kelly says 0.5% on a bet I was planning to stake at 2%, that discrepancy tells me my perceived edge might be smaller than I think, and I reconsider whether the bet is worth placing at all.
Planning for Drawdowns: Stop-Loss Rules and Season-Long Pacing
Every profitable baseball bettor I know has experienced a drawdown of 15-20% at some point during a season. The ones who survive it are the ones who planned for it. The ones who do not plan assume the drawdown means their approach is broken, panic, increase their stakes to recover, and accelerate the losses.
My drawdown plan has three tiers. Tier one: if my bankroll drops 10% from its peak, I reduce my unit size to 1.5% (from 2%) and limit myself to one-unit bets only — no step-ups. Tier two: if the drawdown reaches 20%, I drop to 1% per unit and reduce my daily maximum to three bets. Tier three: if the drawdown hits 30%, I stop betting for one week, review my recent picks, and assess whether the losses reflect bad variance or a genuine flaw in my process.
The pacing element is equally important. A 162-game season per team means there are games every single day from April to October. The temptation to bet daily is strong — the action is always available. But not every day offers value. Some days the pitching matchups are murky, the weather is unpredictable, or the lines are too sharp to find an edge. On those days, the best bet is no bet at all. I aim for 3-5 bets per day during the season, which translates to roughly 500-800 bets across six months. That is more than enough sample size to let a positive edge compound without forcing action on marginal spots.
Bankroll management is not exciting. It does not make for good content on social media, and nobody brags about their unit sizing at the pub. But in my experience — and I have the spreadsheets to prove it — the difference between profitable and unprofitable baseball bettors is more often discipline than analysis. The sharpest model in the world loses money without a bankroll system built for the grind.
How much money do I need to start a baseball betting bankroll?
There is no fixed minimum, but your bankroll should be large enough that a 2% unit size represents a meaningful but not painful amount. If 2% of your bankroll is too small to place at your bookmaker’s minimum stake, the bankroll is too small. As a practical benchmark, a bankroll of 500 to 1,000 pounds allows 10-20 pound units, which is enough to bet comfortably at most UK operators while maintaining proper discipline.
Should I adjust my unit size mid-season if my bankroll grows or shrinks?
Yes. Recalculating your unit size based on your current bankroll — not your starting bankroll — is essential. If your bankroll grows, increasing your unit proportionally captures the benefit of compounding. If your bankroll shrinks, decreasing your unit protects against further drawdown. Most serious bettors recalculate weekly or after every 20-30 bets, whichever comes first.
Written by the editors at Betting on Baseball Games.
