Most Bettors Focus on Finding Edges. The Best Ones Focus on Sizing Them.
You've identified an edge. EdgeScouts is showing a 12% advantage on a weather market — the consensus forecast says 68°F, but Polymarket has the YES contract priced like there's only a 38% chance it hits. Your probability model says 51%. That's real edge.
Now what? How much do you bet?
This is where most people get it wrong — not on the edge itself, but on how much to put on it. Bet too little and you're leaving money on the table over a lifetime of edges. Bet too much and a losing streak wipes out months of profits, or worse, your entire bankroll.
Bankroll management isn't the exciting part of edge betting. But it's the part that separates bettors who last from bettors who don't.
The Kelly Criterion: The Mathematically Optimal Solution
In 1956, physicist John L. Kelly Jr. published a formula for maximizing the long-run growth rate of a bankroll. It's been the gold standard for serious bettors ever since.
The Kelly Criterion says to bet this fraction of your bankroll:
f = (bp - q) / b
Where:
- f = fraction of bankroll to bet
- b = net odds received (what you win per unit wagered)
- p = your estimated probability of winning
- q = probability of losing (1 - p)
Let's use a real example. Say you're betting on a Polymarket contract priced at 40 cents (40% implied probability), but your model says the true probability is 52%. The contract pays $1 if it resolves YES, so your net odds are $0.60 profit per $0.40 risked — or 1.5 to 1.
- b = 1.5
- p = 0.52
- q = 0.48
f = (1.5 × 0.52 - 0.48) / 1.5 = (0.78 - 0.48) / 1.5 = 0.30 / 1.5 = 0.20
Kelly says bet 20% of your bankroll. On a $1,000 bankroll, that's $200.
That might feel aggressive. That's because it is — and that's actually the problem most people run into with full Kelly.
Why Most Edge Bettors Use Half Kelly (Or Less)
Full Kelly is mathematically optimal only if your edge estimate is perfect. In the real world, your probability model is wrong sometimes. Maybe not by much, but enough that full Kelly produces terrifying drawdowns.
Run a simulation: a bettor using full Kelly with a 55% win rate on even-money bets will see their bankroll cut in half roughly every 40-50 bets during a bad run. That's not a theory — it's a property of the math. Kelly maximizes geometric growth, but it also maximizes volatility.
The solution most professional bettors use: Half Kelly.
In the example above, instead of betting 20%, you bet 10%. You give up some expected value, but you cut your variance dramatically. Specifically, half Kelly reduces variance by 75% while only reducing expected growth rate by about 25%. That's an extremely favorable trade.
Some sharp bettors go even further — Quarter Kelly, or flat percentages between 1-5% per bet — especially when they're uncertain about their edge estimates or when dealing with markets that can move against them quickly.
The Edge Size Matters as Much as the Kelly Fraction
Not all edges deserve the same sizing. A 15% edge is not the same as a 3% edge, even if Kelly would technically tell you to bet both.
Here's a practical framework based on edge size:
- Edge under 3%: Pass or minimal bet (0.5-1% of bankroll). Transaction costs and model uncertainty eat into this.
- Edge 3-7%: Standard sizing (1-3% of bankroll). Solid edge, worth taking, but don't overcommit.
- Edge 7-15%: Full allocation (3-5% of bankroll). This is what you've been waiting for. Bet confidently.
- Edge over 15%: Maximum bet, but pause and double-check your model. Markets this mispriced sometimes have a reason.
EdgeScouts displays edge percentage for every signal. Use these tiers as your starting framework, then adjust based on your own bankroll size and risk tolerance.
The Most Dangerous Mistake: Bet Sizing Without Tracking
Here's a scenario that plays out constantly: a bettor hits a 7-game winning streak, gains confidence, and starts bumping their bet sizes. Then variance regresses. They're now betting 3x their original size into a losing run. The profits from the winning streak evaporate in a handful of bets.
This is not bad luck. It's bad math.
Proper Kelly betting means your bet size scales with your bankroll, not with your recent results. Win $200 on a $1,000 bankroll, your bankroll is now $1,200. Your 2% bets go from $20 to $24. That's correct. What's not correct is deciding your confidence is up and jumping to 8% bets.
Three rules that protect against this:
- Recalculate your bet size from current bankroll, not peak bankroll. If you're down 30%, you bet 30% less than at peak. That's how you survive drawdowns.
- Never bet more than 5% of bankroll on a single position, regardless of edge size, unless you're running a very small bankroll (under $200).
- Track every bet. If you can't produce your P&L on demand, you're guessing at whether your system is working.
Sizing Across Multiple Simultaneous Edges
EdgeScouts often surfaces multiple edges at the same time — a weather market, a crypto strike, and a sports total all flashing opportunity simultaneously. How do you size when you have more than one bet open?
The core principle: your total exposure should reflect your conviction in your models, not just the sum of individual Kelly recommendations.
If Kelly says 5% on three simultaneous bets, you're not necessarily fine being 15% exposed at once. The bets may be correlated (all three resolve the same day, all three are weather-adjacent), and you need to consider that.
A practical cap: limit total simultaneous exposure to 15-20% of your bankroll when bets are uncorrelated, and tighter (8-12%) when there's any correlation between outcomes.
More edges on the board doesn't mean you should deploy more capital proportionally. It means you have more opportunities — some of which you can skip because you're already at your exposure limit.
The Long Game
Bankroll management is how edge bettors survive long enough to let their advantage compound. A 5% edge sounds modest, but over 500 bets with good sizing, the math is extraordinary. The problem is most people don't make it to 500 bets because they over-bet early and blew up their bankroll on variance.
The best edge bettors in the world aren't the ones with the highest win rates. They're the ones who've been at it the longest, survived the inevitable bad runs, and let compounding do its work.
Full Kelly says bet big. Experience says bet smaller and stay in the game.
Putting It Together
When you spot an edge in EdgeScouts, your sizing decision should take 10 seconds:
- What's the edge percentage? (Use the tier framework above)
- What's my current bankroll? (Not what it was — what it is)
- How many other bets do I have open? (Check total exposure)
- Apply Half Kelly or your flat percentage. Bet.
Edges are the foundation, but bankroll management is the architecture. You need both.
EdgeScouts surfaces the edges — we track win rates, expected value, and historical performance across every market category so you know exactly what you're working with. The sizing is up to you, but now you have the framework.
Want to see live edges across sports, weather, crypto, and economic markets? Start your free EdgeScouts trial and put this framework to work.