Kelly vs Flat Staking 2026 — Which Wins After 1,000 Bets?
Two camps in the betting world. Kelly disciples insist size scales with edge — bigger advantage means bigger bet. Flat-staking purists insist consistency is everything — same percentage of bankroll on every bet, regardless of edge size. Who wins after 1,000 bets? We ran the Monte Carlo. The answer is more nuanced than either side will tell you.
The two strategies in plain terms
Flat staking
You bet a fixed percentage of your current bankroll on every bet — typically 1% or 2%. Edge size doesn't change the bet. A 5% edge and a 30% edge both get the same stake. Simple, low variance, easy to execute.
Kelly Criterion (fractional)
You compute the textbook Kelly fraction f* = (p − m)/(1 − m) for each bet, multiply by a safety factor (typically 0.25), and bet that. Edge size scales the bet directly. Big edges get big bets. Tiny edges get tiny bets. Negative edges get zero bet.
The simulation setup
To make the comparison fair, both strategies face the same sequence of 1,000 bets with realistic mixed edges:
- 40% of opportunities have a real 4% edge
- 40% have a real 8% edge
- 15% have a real 15% edge
- 5% have a 25%+ edge (rare juicy spots)
- Win/loss is a Bernoulli draw at the true probability
Flat staker bets 2% per bet (a strong choice, on the upper end of what conservative bettors recommend). Kelly bettor uses quarter Kelly with 75% credible bound on the edge. Both start with $10,000.
10,000 trial paths each. Median outcomes, drawdowns, and tail behavior compared.
The headline result
Median final bankroll after 1,000 bets:
- Flat 2%: ~$58,000 (5.8× starting bankroll)
- Quarter Kelly: ~$220,000 (22× starting bankroll)
Kelly wins by roughly 4× on the median path. The reason: when there's a 25% edge, Kelly bets meaningful size; flat staking bets the same 2% it would on a 4% edge. The big spots are where Kelly compounds.
The drawdown story
Median maximum drawdown over the 1,000-bet sequence:
- Flat 2%: ~12% drawdown — manageable, almost boring
- Quarter Kelly: ~22% drawdown — meaningful, but recoverable
Kelly's drawdowns are nearly twice as large in the median case. In the worst 5% of paths, Kelly drawdowns hit 40%+; flat staking stays under 25%. Volatility is the price you pay for the higher growth rate.
Why fractional Kelly still wins overall
Two reasons:
- Kelly captures the convexity. A 25% edge is roughly 6× more profitable per dollar than a 4% edge if you size proportionally. Flat staking ignores that.
- Quarter Kelly's drawdowns aren't terminal. A 22% drawdown is recoverable. The growth rate is high enough that you climb back fast. The full Kelly version (which we did not use here) would routinely hit 60% drawdowns — psychologically crippling and sometimes fatal to a strategy.
When flat staking actually wins
Flat staking is the right choice in three specific situations:
- You can't measure your edge. If you don't know whether a bet is a 4% or 12% edge, Kelly sizing is guessing-augmented-by-math. Flat is honest.
- You're psychologically fragile to drawdowns. If a 25% drawdown will make you abandon the strategy, fractional Kelly is the wrong tool. Flat at 1-2% gives you the calmest possible ride.
- Your win rate isn't calibrated. Kelly assumes your probability estimates are roughly right. If you're systematically optimistic, fractional Kelly amplifies the cost. Flat staking caps the damage.
The hybrid play
You don't have to pick one. A practical compromise:
- Compute fractional Kelly
- Apply a hard cap — never bet more than 5% of bankroll regardless of what Kelly says
- Apply a soft floor — never bet less than 0.5% on anything you're willing to bet at all
The cap protects you from a bad estimate driving Kelly to recommend an absurd size. The floor keeps your transaction costs reasonable on small bets. The middle is where Kelly does its work.
Pros and cons summary
Kelly Criterion (fractional)
- Pros: Higher long-run growth, sized for edge, captures convexity, mathematically optimal
- Cons: Bigger drawdowns, sensitive to estimate noise, requires you to actually compute edge per bet
Flat staking
- Pros: Simple to execute, low variance, robust to estimate error, calm drawdowns
- Cons: Leaves money on the table on big edges, doesn't compound as fast, can over-bet on weak edges
Frequently asked questions
Is Kelly always better than flat staking?
On the math, yes — fractional Kelly has higher expected geometric growth. In practice, only if your edges are calibrated. With sloppy probability estimates, flat staking can outperform.
What flat percentage matches quarter Kelly's risk?
Roughly speaking, flat 1-1.5% has comparable drawdown variance to quarter Kelly on a typical mixed-edge book. But you give up most of the growth rate to get there.
Can I switch from flat to Kelly mid-strategy?
Yes — most operators start flat to learn the market, then transition to fractional Kelly once they have 50-100 tracked bets and can confirm their hit rate is calibrated.
Bottom line
Quarter Kelly beats flat 2% staking by ~4× on the median bankroll path after 1,000 bets — but it pays for that with roughly 2× the drawdown variance. If your edges are calibrated and you can stomach 20-25% drawdowns, Kelly compounds harder. If your edges are uncertain or your psychology is fragile, flat staking is the honest choice. Either way, our Kelly Criterion calculator shows you the trade-off in a Monte Carlo so you can see exactly what you're signing up for.
Related reading
- Beta Priors & Robust Kelly Sizing 2026 — Don't Trust Your Hunch
- Walk-the-Book Bet Sizing 2026 — Order Book Aware Kelly
- Top Polymarket Strategy 2026: How to Make Money
Size Your Polymarket Bets Like a Pro
Try our free Kelly Criterion calculator — Bayesian shrinkage, walk-the-book sizing, 10k-trial Monte Carlo. Built for serious Polymarket traders. Auto-fills from live ProbBrain signals.
🎯 Open the Kelly Calculator →Follow us on 𝕏 for daily crypto tips: @ProbBrain
18+ only. Gamble responsibly.
Comments
Post a Comment