You can do everything right (find real 3% edges, size them sensibly, never chase) and still be down after a month. That is not unusual bad luck. It is exactly what the math predicts. Understanding variance is what keeps good bettors from quitting a winning process, and keeps everyone else from mistaking a hot streak for skill.
The numbers behind a downswing
Take a bettor who places bets at even money (+100) with a true 51.5% win chance: a 3% edge.
- Expected profit per bet: 0.03 units
- Standard deviation per bet: about 1 unit
Over n bets, expected profit grows with n, but the swings grow only with the square root of n:
| Bets | Expected profit | 1 std. dev. range | Chance you are losing |
|---|---|---|---|
| 100 | +3u | ±10u | ~38% |
| 500 | +15u | ±22u | ~25% |
| 1,000 | +30u | ±32u | ~17% |
| 3,000 | +90u | ±55u | ~5% |
Even after 1,000 genuinely +EV bets, roughly one bettor in six is still in the red. After 100 bets, the result is close to a coin flip.
Underdogs swing harder
Longer odds mean more variance. A 3% edge on +300 underdogs wins only about 26% of the time, so losing streaks of 10–15 in a row are routine. Player props at plus money and long parlays are the most volatile of all. If you bet mostly underdogs, size smaller. (Bankroll management.)
How many bets until you know?
Proof from results alone takes a long time. A rough rule at even money: the number of bets needed to be about 95% sure your edge is real is around 4 ÷ edge². For a 3% edge that is about 4,400 bets. For 2%, around 10,000.
That is why serious bettors grade themselves on closing line value instead of profit. CLV has far less noise: within a few hundred bets you can see whether you consistently get better prices than the market settles at. (CLV explained.)
Signs it is variance, not a broken process
- Your average CLV is still positive
- Your bets still meet your edge and confidence thresholds
- Losses are spread across sports and markets rather than concentrated in one place
- Your bet size has stayed consistent
Signs it might be the process
- Your CLV is negative or near zero across 300+ bets
- You have been taking edges from a single stale source
- You have drifted into bigger bets, more parlays or more "feel" plays
- One market (say, live props) accounts for most of the losses
A bet tracker that breaks results down by sport, market and book turns this diagnosis from a guess into a glance.
Surviving the swings
- Bet small. At 1% of bankroll per bet, a 30-unit downswing costs 30% of your roll: painful but survivable. At 5% per bet, it is fatal.
- Use fractional Kelly. Full Kelly maximizes growth on paper but produces brutal drawdowns when edges are overestimated. Quarter Kelly is the common compromise. (Fractional Kelly.)
- Do not change your process mid-downswing. Raising stakes to "get back to even" is chasing. (Tilt and chasing losses.)
- Zoom out. Review results monthly, not after every bet.
Frequently asked questions
Is it normal to lose money with +EV betting?
Yes. Over a few hundred bets, a real 2–3% edge still loses money a large share of the time. The edge becomes dependable only over thousands of bets.
How long does a betting downswing last?
It can last weeks or months. A 30–50 unit drawdown is not unusual over a few thousand bets, especially if you bet underdogs and props.
What is the best way to know if my betting is working?
Track closing line value. Beating the closing line on average is the strongest early evidence of a real edge.