Skill or luck?
Most winning records are luck that hasn't run out yet. Type in yours and get the same kind of test our own picks are graded by in public: one that stays honest however often you check it.
Leave pushes out: they hand the stake back. One price stands in for every bet, so a record at very mixed prices or full of parlays gets a rough answer.
This is an example record. Type in yours.
Can’t tell from luck yet.
You won 58.0% of 100 bets; at −110 you need 52.4% to break even. Your long-run rate could be anywhere from 42.2% to 73.8%, and break-even sits inside that, so this record fits a bettor with no edge at all. If you kept winning 58.0%, about 651 more bets would settle it.
Luck alone
A bettor with no edge at all, over 100 bets at −110, does at least this well 15% of the time. That is for one look at the record; the verdict also allows for checking after every bet, which is why it asks for more.
How many bets does it take?
If your true win rate were each of these, at −110, this is how many bets the test would need before it could tell you from luck. A real 55% bettor at −110 is a very good bettor, and still needs about 3,916 bets: years of betting before anyone, the bettor included, can know.
| True win rate | Return per bet | Bets to tell it from luck |
|---|---|---|
| 53% | +1.2% | more than 20,000 |
| 54% | +3.1% | 11,182 |
| 55% | +5.0% | 3,916 |
| 56% | +6.9% | 1,936 |
| 58% | +10.7% | 751 |
| 60% | +14.5% | 398 |
The price takes its cut first
At −110 you risk $110 to win $100, so you must win 52.4% of your bets just to stay even. Winning more than half is not the bar; beating the price is.
Why checking breaks most tests
A textbook significance test assumes you look once. Bettors look after every bet, and stopping the moment a record looks good makes luck look like skill. This test (a confidence sequence) holds at 95% however often you look, so it asks for more bets.
The faster signal is the close
Wins and losses are slow to say anything. Whether your price beat the line the book closed at before tip-off says more per bet, which is why serious bettors track it, and why our record does.
Every pick our model makes is logged before tip-off and graded after the final, and the season's record is judged by this same kind of test, whether the answer flatters us or not.
See the track recordElite's edge audit runs this test on each bet you log, at the price you actually took: where your edge is by market and book, whether you beat the closing line, and whether you chase losses.
See plansIf your record has become a source of stress, the National Problem Gambling Helpline is free and confidential, any time: 1-800-GAMBLER. More help.