The Kelly criterion is a formula for bet sizing that maximizes long-run bankroll growth: bet the fraction of your roll equal to your edge divided by the odds. It's the mathematical answer to "how much?"
The formula
Kelly fraction = (b × p − q) ÷ b
where b is the profit per $1 at the odds (decimal minus 1), p is your true win probability, and q = 1 − p.
At +120 with a 48% win probability: b = 1.2, so (1.2 × 0.48 − 0.52) ÷ 1.2 = 4.7% of bankroll. Note what happens with no edge: at p = 45.5% (the break-even rate for +120), Kelly is zero. No edge, no bet — the formula enforces the discipline for you.
Why nobody bets full Kelly
Full Kelly assumes p is exactly right. It never is. Overestimate your edge by a little and full Kelly has you systematically overbetting, which is how bankrolls die even with a real edge. The standard fix is fractional Kelly: half Kelly keeps roughly three-quarters of the growth rate with half the volatility. Most professionals run half or quarter.
What Kelly is really for
Even if you never compute it at the window, Kelly's logic should shape your sizing: bigger edges deserve bigger bets, small edges deserve small ones, and "I feel great about this one" is not a variable in the formula. Big Mike's picks carry Kelly-based sizing language for exactly this reason — the units aren't vibes.
