Bankroll Calculator: How to Set Your Starting Bankroll and Unit Size
Bankroll calculator guide with 3 methods for setting your starting bankroll. Includes risk tolerance math, unit sizing formulas, and 4 worked examples.
What a Bankroll Calculator Actually Does
A bankroll calculator answers two questions: how much money should you allocate to betting, and how large should each bet be? These are not opinion questions. They are math problems with specific inputs and outputs.
The inputs: your edge, your risk tolerance, your bet frequency, and the odds you typically play. The output: a starting bankroll that gives you a defined probability of survival over a defined number of bets. Get this wrong and even a profitable strategy goes broke before the edge compounds. Get it right and you build a system that survives variance and grows.
This guide walks through three methods for calculating your starting bankroll, from simple rules of thumb to simulation-based approaches. Each method links to tools that run the math for you. Start with the Kelly Criterion calculator to size individual bets, then use the framework here to determine how large your total bankroll needs to be.
Method 1: The 100-Unit Rule
The simplest bankroll calculation works backward from your bet size. If you want to bet $50 per play, you need $5,000. If you want to bet $20, you need $2,000. Your bankroll equals 100 times your standard unit.
Why 100 units? At typical sportsbook odds (-110), a bettor with a 2% edge has roughly a 5% chance of going broke before the edge kicks in with a 100-unit bankroll. Drop to 50 units and that risk jumps to 18%. At 200 units, it falls below 1%.
Worked example. You have $3,000 to allocate. Your unit size is $3,000 / 100 = $30 per bet. At -110 odds with a 2% edge, your probability of surviving 1,000 bets is approximately 95%.
The 100-unit rule works for most recreational bettors because it requires zero edge estimation. The weakness: it ignores your actual edge and the odds you play. A bettor consistently playing +300 underdogs faces more variance than someone betting -150 favorites, even with the same edge. The same 100-unit bankroll gives very different survival probabilities for these two bettors.
For a deeper dive into why unit sizing matters and how to adjust as your bankroll changes, read the full bankroll management guide.
Method 2: Risk-of-Ruin Based Calculation
Risk of ruin (RoR) is the probability that your bankroll hits zero before your edge generates enough profit to sustain you. It depends on four variables:
- Edge (e): Your average advantage per bet as a decimal
- Win probability (p): How often you win
- Odds (d): Average decimal odds
- Bankroll in units (B): Total bankroll divided by bet size
The simplified RoR formula for fixed-stake betting:
RoR = ((1 - e) / (1 + e))^B
Where e is your edge as a proportion of the amount wagered.
Worked example. You estimate a 3% edge betting at -110 (decimal 1.909). Your win probability is 55.3% (the break-even rate at -110 is 52.4%, plus your 3% edge).
e = 0.03, so RoR = (0.97 / 1.03)^B = (0.9417)^B
For a 5% risk of ruin: 0.05 = (0.9417)^B. Take the natural log: ln(0.05) = B × ln(0.9417). Solve: B = -2.996 / -0.0601 = 49.8 units.
So 50 units gives you a 5% chance of going broke. If you want a 1% risk of ruin: B = ln(0.01) / ln(0.9417) = 76.6 units, round up to 80.
At $50 units with a 1% RoR tolerance, your minimum bankroll is $4,000. At $25 units, it is $2,000. The Kelly Criterion calculator handles the sizing math once you have your bankroll set.
The critical insight: RoR drops exponentially as you add units. Going from 50 to 100 units does not halve your risk. It makes it nearly negligible. But going from 100 to 200 units barely changes anything. The sweet spot for most bettors is 75-100 units at a 1-3% RoR.
Method 3: Kelly-Derived Bankroll Sizing
The Kelly Criterion tells you what fraction of your bankroll to wager on each bet. You can reverse-engineer the starting bankroll from your desired bet size and Kelly fraction.
Kelly fraction = (bp - q) / b
Where b = net odds (profit per dollar wagered), p = win probability, q = 1 - p.
Worked example. You find a bet at +150 (decimal 2.50, b = 1.50) where you estimate 45% win probability.
Kelly = (1.50 × 0.45 - 0.55) / 1.50 = (0.675 - 0.55) / 1.50 = 8.3%
Half Kelly (the standard for real-world application): 4.2%
If you want each bet to be $100, your bankroll needs to be $100 / 0.042 = $2,381. Round up to $2,500.
If you want each bet to be $50: bankroll = $50 / 0.042 = $1,190. Round up to $1,200.
This method is the most precise because it accounts for your specific edge and odds. The downside: it requires accurate probability estimates. If you overestimate your edge by even 2 percentage points, your Kelly fraction is too high, and your bankroll is too small. That is why half Kelly or quarter Kelly is standard practice. The margin of safety absorbs estimation error.
For prediction market traders, the same math applies with one addition: platform fees reduce your effective edge. A 5% raw edge on Kalshi becomes roughly 3.5-4% after the 7% winner fee. Size your bankroll to the fee-adjusted edge, not the raw edge. The PM EV calculator gives you the net edge after fees.
How Risk Tolerance Changes the Numbers
Your risk tolerance is a personal input, not a mathematical one. But it has mathematical consequences. Here is how different RoR targets translate to bankroll requirements for the same bettor (3% edge, -110 average odds):
| Risk Tolerance | Risk of Ruin | Min Units | Bankroll ($50 unit) |
|---|---|---|---|
| Aggressive | 10% | 38 | $1,900 |
| Moderate | 5% | 50 | $2,500 |
| Conservative | 1% | 77 | $3,850 |
| Ultra-conservative | 0.1% | 115 | $5,750 |
The difference between "aggressive" and "conservative" is roughly 2x the bankroll. That is the cost of safety. Most serious bettors choose the 1-5% range. Below 1%, you are over-capitalizing relative to the edge. Above 10%, you are taking unnecessary risk of going bust.
One factor this table does not capture: emotional tolerance. If a 20-unit drawdown makes you change your strategy, abandon your models, or chase losses, you need a larger bankroll regardless of what the math says. The math assumes you bet consistently through losing streaks. If you cannot do that, add units until the maximum expected drawdown stays within your psychological comfort zone.
Track your actual results against these projections using the edge calculator to verify whether your edge estimates are calibrated.
Unit Size Adjustments: When and How
Your initial calculation is a starting point. As your bankroll grows or shrinks, your unit size should adjust. There are two schools of thought.
Fixed percentage. Recalculate your unit as a fixed percentage of current bankroll after every session or week. If your bankroll grows from $2,500 to $3,000, your unit increases from $50 to $60. This lets winners compound. The risk: drawdowns also compound. A 10-unit losing streak at increasing stakes hurts more than the same streak at fixed stakes.
Step function. Only adjust at specific bankroll milestones. For example: units stay at $50 until the bankroll hits $3,500, then increase to $65. If the bankroll drops below $2,000, units decrease to $40. This approach is less mathematically optimal but more psychologically manageable. Most bettors execute this better because the decisions are infrequent and clear.
Worked example of step function:
| Bankroll Range | Unit Size | Units Available |
|---|---|---|
| $1,500-$2,499 | $30 | 50-83 |
| $2,500-$3,999 | $50 | 50-80 |
| $4,000-$5,999 | $75 | 53-80 |
| $6,000+ | $100 | 60+ |
Notice: units available never drops below 50. That is the safety floor. When your bankroll shrinks, your units shrink to maintain the minimum buffer.
The turnover calculator shows how unit size adjustments interact with bankroll turnover to affect your compounding rate. Faster turnover with correct sizing is the real engine of bankroll growth.
Bankroll Calculation for Prediction Markets
Prediction market traders face two complications that traditional sports bettors do not.
Capital lockup. When you buy a contract at $0.35 on Kalshi, that $0.35 is locked until the contract resolves or you sell. If the resolution is 6 months away, your capital is frozen. The bankroll turnover effect is dramatic: a $5,000 bankroll with $2,000 locked in long-dated contracts is really a $3,000 active bankroll for sizing purposes.
Correlated positions. Buying "Democrats win Senate" and "Democrats win House" creates correlated exposure. Your effective position size is not the sum of the two bets treated independently. It is larger because a single thesis drives both outcomes. The correlation calculator estimates this adjustment. Read the correlated positions guide for the full framework.
For prediction market sizing, start with the Kelly-derived method but apply these corrections:
- Use fee-adjusted edge (not raw edge) from the fee calculator
- Reduce available bankroll by the amount locked in positions resolving in 30+ days
- Apply a correlation discount to total exposure in related contracts
- Use quarter Kelly instead of half Kelly for contracts resolving in 90+ days (the longer the lockup, the more conservative you should be)
The prediction market bankroll management guide covers the full position-management system, including when to exit early versus hold to settlement.
Common Bankroll Calculation Mistakes
Using your whole net worth as bankroll. Your bankroll is money you allocate specifically to betting. Not your savings. Not your checking account. The math only works if you can lose the entire bankroll without changing your lifestyle. If losing it would hurt, it is too large.
Ignoring the edge requirement. No bankroll size saves a bettor with zero edge. A $100,000 bankroll with no edge simply means you lose $100,000 more slowly. Before calculating bankroll size, verify your edge with the edge calculator and at least 200 tracked bets.
Assuming past edge persists. Your edge over the last 3 months may not reflect your true edge. Markets adapt, lines sharpen, information advantages erode. Size your bankroll to a conservative estimate of your forward-looking edge. If you think your edge is 4%, size as if it is 2%. The math is more forgiving of over-capitalization than under-capitalization.
Skipping the drawdown simulation. Even a 3% edge bettor at -110 odds will experience a 25-unit drawdown roughly once every 1,000 bets. If your bankroll is 50 units, that drawdown takes you to half your starting bankroll. Decide in advance: is a 50% drawdown something you can stomach without changing your strategy? If not, increase your units. Read about risk of ruin for the simulation framework.
Not adjusting for bet frequency. A bettor placing 10 bets per day needs a different bankroll than one placing 10 per month. Higher frequency means faster convergence to your true win rate, which means fewer units are needed for the same RoR. But higher frequency also means more total variance in absolute dollars. The interaction is non-obvious. Run the numbers for your specific volume.
Frequently asked questions
- How many units should a sports betting bankroll have?
- 75 to 100 units for most bettors. At 100 units with a 2-3% edge betting at -110 odds, your risk of ruin is below 5%. Drop below 50 units and the probability of going bust rises above 10% even with a real edge.
- How do I calculate my bankroll for prediction markets?
- Use the Kelly-derived method with fee-adjusted edge. On Kalshi (7% winner fee), a 5% raw edge becomes roughly 3.5% net edge. At quarter Kelly with 3.5% edge, your bankroll should be about 67 times your average position size. A $3,350 bankroll supports $50 average positions.
- Should I increase bet size as my bankroll grows?
- Yes, but use a step function rather than continuous adjustment. Increase units at specific bankroll milestones (e.g., every 40-50% growth) rather than recalculating after every bet. This limits the compounding effect of drawdowns while still capturing growth.
- What is the minimum bankroll for sports betting?
- The minimum depends on your bet size and edge. For $10 units, 100 units means $1,000. For $25 units, $2,500. The absolute minimum that allows meaningful Kelly sizing with a real edge is roughly $500 at $5 units, but variance at that level makes the experience brutal.
- How does risk of ruin affect bankroll size?
- Risk of ruin drops exponentially as bankroll increases. Going from 50 to 75 units cuts your RoR by roughly 80% (from about 5% to 1% with a 3% edge). Going from 75 to 150 barely changes it. The mathematically efficient range is 75-100 units for most edge profiles.