How win rates work
A published win rate can be true and still say little about the bets behind it: it depends on what is counted as a win, and at what price. Here are common ways a rate is made to look better than the bets, each checked on Oddsfolio’s own soccer archive, and then how Oddsfolio counts its own.
Several tips a match, counted as won if any one lands
Give several tips on every match, tell readers to pick the one they like, and count the match as won if any one of them comes in. We ran one fixed rule, with no model at all, over every archived match with match-result prices: 46,955 matches, 10 Mar 2012 to 30 Oct 2026. Four tips a match: favourite to win; over 1.5 goals; both teams to score; favourite to win & over 1.5 goals. The favourite is the team with the shorter archived match-result price (the home side when the two are level).
| Tip | Landed | Rate |
|---|---|---|
| Any one of the four | 40,496 of 46,955 | 86.2% |
| Favourite to win alone | 23,939 of 46,955 | 51% |
| Over 1.5 goals alone | 35,132 of 46,955 | 74.8% |
| Both teams to score alone | 24,917 of 46,955 | 53.1% |
| Favourite to win & over 1.5 goals alone | 18,575 of 46,955 | 39.6% |
The four overlap: both teams scoring already means at least two goals, and the last tip is the favourite’s win with a second goal. So the set loses only on these scores, with the favourite at home (mirrored when the favourite is away): 0-0, 0-1. Nobody can bet “any one of four”. A reader backs one tip, and it lands only as often as its own row says.
Tips that cannot all lose
Some sets of tips cover every possible score: whatever the result, at least one of them lands, so a match given all of them is a “win” before it kicks off. Two such sets, checked by trying every score:
Over 1.5 goals + Not both teams to score
If over 1.5 goals fails, the match has at most one goal, so at most one team scored and “not both teams to score” lands. Every score from 0-0 to 9-9 tried (100): no score loses them all.
Under 3.5 goals + Both teams to score + Either team to win & over 1.5 goals
Three goals or fewer: under 3.5 lands. Four or more in a draw: each side scored at least twice, so both teams scored. Four or more with a winner: a win with over 1.5 goals. Every score from 0-0 to 9-9 tried (100): no score loses them all.
A slip graded match by match
A slip wins only if every leg wins. Some records list each match of a slip as its own result, so a slip that lost on one leg still adds wins to the count. Even at the any-one-of-four rate above (86.2% a match), three independent matches all landing together is 64.1%.
Oddsfolio’s bankers slip, graded as a slip: No slips graded yet.
The same bankers graded one by one: 2–0 (100%) · average price -434 (needs 81.3% to break even) · +$46.03 on $100 a pick
The slip line is the one a bettor of the slip actually had. Both are on Daily picks.
Very short prices
A high hit rate can still lose money. The price sets the rate a tip needs just to break even, and how many wins one loss wipes out:
| Price | Decimal | Needs to land | One loss costs |
|---|---|---|---|
| -900 | 1.11 | 90.0% | 9 wins |
| -500 | 1.20 | 83.3% | 5 wins |
| -250 | 1.40 | 71.4% | 2.5 wins |
| -110 | 1.91 | 52.4% | 1.1 wins |
A tipster landing 90.0% of picks at -900 has only broken even; anything less loses money, however high the rate looks.
One tip on every archived match, at the archived bookmaker price (the market average where the archive has one):
| Tip | Matches | Landed | Average price | Return per $100 staked |
|---|---|---|---|---|
| Favourite to win 10 Mar 2012 to 30 Oct 2026 | 46,955 | 51% | 1.98 | −$3.65 |
| Over 2.5 goals 23 Jul 2021 to 29 Oct 2026 | 35,013 | 50.9% | 1.92 | −$5.05 |
Each loses money at those prices: the book’s margin is in every price, so a tip that lands often can still lose.
How Oddsfolio counts
- One call, one result: every call is graded on its own against the final score, at the line it was made at, and stays on the Record. Nothing is picked after the game.
- No tip sets: a game gets only the calls that cleared their own tested filter, and each is graded separately, so two calls on one game can both lose.
- Slips are graded as slips: the bankers slip and the 5 odds and 10 odds slips win only when no leg loses.
- Daily picks keep the price they were taken at, so each record shows the rate it needs to break even and the profit of a flat $100.
- Every rate on the Record shows its sample size, losses included.
Bankers slip: No slips graded yet.
Banker legs: 2–0 (100%) · average price -434 (needs 81.3% to break even) · +$46.03 on $100 a pick
5 odds slip: No slips graded yet.
10 odds slip: No slips graded yet.
Not advice. 21+. Bet only what you can afford to lose.