How often does the favourite win in football?
Published
Search this question and you get a different answer every time. One widely-shared figure says favourites win 57.9% of the time; another says barely half; a third quotes 55% but is counting World Cup matches only; several of the top results answer for American football and never mention it. The disagreement is not really disagreement — it is people measuring different competitions over different periods and dropping the qualifier.
So here is our answer, with the qualifier attached: across 16,628 matches in the nine leagues we model, each one carrying a bookmaker’s closing odds, the favourite won 53.1% of the time.
What “favourite” means here
The definition matters more than the number. Throughout this article the favourite is the outcome with the shortest closing price in the match-result market — the last price before kick-off, after the market has absorbed team news, weather and money. It is not the bigger club, not the home side, and not our model’s pick.
Closing odds are the strictest available benchmark because they are the market’s final word. Earlier prices are noisier and easier to look clever against.
Two things follow immediately from that definition:
- The favourite is the home team 67.8% of the time. Home advantage does most of the work in deciding who gets the short price.
- The average favourite goes off at about 1.97 — implying, before margin, that the typical “favourite” is roughly a coin flip rather than a strong bet.
That second point explains most of the confusion. People picture a favourite as a heavy one, but the average favourite across a full season of fixtures is priced near evens, and near-evens outcomes lose all the time.
Home favourites and away favourites
Splitting by venue shows a smaller gap than folklore suggests:
- Home favourites win 53.9%
- Away favourites win 51.5%
An away favourite is priced short precisely because the market thinks the team is good enough to overcome the venue, and that judgement is mostly sound — the 2.4-point gap is real but modest. If you have absorbed the idea that away favourites are traps, the data does not support it strongly.
The number moves with the price
A single average hides the structure. Grouping every match by the favourite’s closing price:
| Favourite’s closing odds | Matches | Favourite won |
|---|---|---|
| Under 1.30 | 1,199 | 83.6% |
| 1.30 – 1.59 | 2,776 | 68.4% |
| 1.60 – 1.99 | 4,686 | 54.2% |
| 2.00 – 2.49 | 5,553 | 45.0% |
| 2.50 and above | 2,414 | 36.7% |
Two readings are worth taking from this.
First, the market is broadly right. Win rates fall smoothly as prices lengthen, in roughly the order the prices imply. Whatever else is true of bookmakers, their ranking of outcomes is not noise.
Second — and this is the part that surprises people — most favourites are not favourites in the sense you imagine. Nearly half of all matches (7,967 of 16,628) have a favourite priced at 2.00 or longer, and in that group the favourite loses more often than it wins. “Backing the favourite” as a strategy spends most of its time on outcomes that are odds-against.
A caution on comparing these win rates to the odds themselves: a bookmaker’s implied probability includes their margin, so it will sit above the true probability by construction. Working out what a price really claims means stripping the margin out first, and the margin is not spread evenly — it leans on the longer prices, which is the favourite-longshot bias.
The league changes the answer by nine points
If you take one thing from this article, take this. The same question, asked of nine different competitions, returns answers nearly ten points apart:
| League | Matches | Favourite won | Draws |
|---|---|---|---|
| Primeira Liga | 1,530 | 56.9% | 24.8% |
| Eredivisie | 1,530 | 56.0% | 24.2% |
| Premier League | 1,900 | 55.5% | 23.9% |
| La Liga | 1,900 | 54.2% | 26.2% |
| Bundesliga | 1,530 | 54.1% | 24.9% |
| Serie A | 1,900 | 53.9% | 27.2% |
| Ligue 1 | 1,678 | 52.9% | 24.6% |
| Brasileirão | 1,900 | 50.8% | 27.3% |
| Championship | 2,760 | 47.4% | 26.1% |
In the Championship, the favourite loses more often than it wins. England’s second tier is 46 rounds of closely-matched teams with heavy fixture congestion, and the market’s best guess comes in under half the time — nine and a half points below the Primeira Liga, where a small number of dominant clubs make short prices reliable.
Notice what tracks with it: the leagues where favourites do worst are, broadly, the leagues that draw most often. Every draw is a match the favourite did not win, and competitions that produce draws at 26–27% mathematically cap how often anyone’s favourite can come in.
The practical consequence is that any single figure for “how often the favourite wins” is close to useless without naming the competition. A record built in the Primeira Liga and a record built in the Championship are not comparable, and a tipster moving between them can change their apparent hit rate by ten points without changing anything about their skill.
Why a high hit rate still isn’t the point
53.1% sounds respectable until you notice that the average favourite is priced at 1.97. Backing every favourite in this dataset at that kind of price is close to a break-even proposition before the bookmaker’s margin, and comfortably losing after it. A hit rate is not a return, and the two frequently point in opposite directions — which is why we score our own predictions with the Brier score rather than a win percentage, and why a bare accuracy figure from any prediction service tells you almost nothing.
About these numbers
Every figure here comes from our own database: finished matches in the nine leagues we cover, joined to closing prices from football-data.co.uk. Not every match we hold carries closing odds, so the per-league counts are slightly below our full match totals — the Brasileirão, for instance, contributes 1,900 priced matches out of a larger set of results. The figures are a snapshot as this was written, and they will drift as more matches finish.
None of this is betting advice, and a favourite winning 53.1% of the time is not a system. Treat probabilities as probabilities, and stake only what you can afford to lose.