Premier League Over/Under 2.5: Do the Goals Odds Get It Right?
Twenty-one seasons of Premier League over/under 2.5 prices set against what landed: where the market reads the game well, where it drifts, and what backing each side cost.

- Premier League matches over 2.5 goals (7,980)52.3%
- Average priced over 2.5 chance, margin removed51.8%
- Flat-stake return, every over 2.5−4.7%
- Flat-stake return, every under 2.5−6.5%
Every Premier League goals price is a judgement on one question: will this game open up or close down? I wanted to know how good that judgement has been over the long run. So I set the margin-free over 2.5 chance from a major bookmaker against the final scores of 7,980 league matches from 2005/06-2025/26.
The short answer is very close, but not free. Over 2.5 landed in 52.3% of matches against an average priced chance of 51.8%, yet backing every over returned −4.7% and every under −6.5%. For match-by-match reading, my over/under betting tips page is where that work lives. This file is the background I check before any of it.
§ 1The market's goals prices track reality across most bands
Each row groups matches by their over 2.5 price and compares the chance the price implied with how often three or more goals actually came. In the shortest band, under 1.50 across 629 matches, the market implied 67.4% and over landed 69.2% (±3.6). In the 1.90–2.09 band, the most crowded with 2,339 matches, priced and actual sit almost on top of each other.
The one band that stands out among those with 300+ matches is 1.70-1.89. There the priced chance was 52.8% and over landed 55.5%, a gap of +2.7 points. That is the market slightly underestimating games it considered roughly even.
View the data: premier league over 2.5: priced vs landed
| Priced chance | Actually went over | 95% range | |
|---|---|---|---|
| Under 1.50 (629 matches) | 67.4% | 69.2% | ±3.6 |
| 1.50–1.69 (1,567 matches) | 59.0% | 57.6% | ±2.4 |
| 1.70–1.89 (2,068 matches) | 52.8% | 55.5% | ±2.1 |
| 1.90–2.09 (2,339 matches) | 47.2% | 47.7% | ±2 |
| 2.10–2.39 (1,245 matches) | 43.0% | 42.2% | ±2.7 |
| 2.40+ (130 matches) | 37.9% | 33.1% | ±8.1 |
At the long end, 2.40 or bigger, the price implied 37.9% and only 33.1% went over. With just 130 matches the range is a wide ±8.1, so I treat that as a hint, not a finding. What I take from the chart is simple: the market reads the shape of a Premier League game well, and any edge sits in the margins.
§ 2Backing every over or under has lost money
Close pricing still leaves the margin to pay. Backing over 2.5 in all 7,978 priced matches at flat stakes returned −4.7%; backing under returned −6.5%. Overs have been the slightly cheaper side to back blind, which fits with the market pricing the over chance a touch below what landed.
The bands follow the first chart. The 1.70-1.89 band, where overs were underpriced, was the least costly place to back them at −0.5% across 2,068 matches. For unders, the only positive figure came where the over price was 2.40 or bigger, at +2.0%.
View the data: backing every over or under in the premier league
| Back over 2.5 | Back under 2.5 | |
|---|---|---|
| Under 1.50 | −2.7% | −9.6% |
| 1.50–1.69 | −7.6% | −2.0% |
| 1.70–1.89 | −0.5% | −10.5% |
| 1.90–2.09 | −4.9% | −6.8% |
| 2.10–2.39 | −7.5% | −4.6% |
| 2.40+ | −17.9% | +2.0% |
| All matches | −4.7% | −6.5% |
Here is the caveat I hold myself to. Returns a few points either side of zero can easily come from chance, and none of these bands is a betting system. The positive under figure rests on that same thin 130-match band. I read this chart as a map of where the margin hurts least, not where profit is waiting.
§ 3Overs have climbed, and 2023/24 caught the market out
The Premier League has become a more open league. In 2005/06 45.0% of matches went over 2.5; in 2025/26 it was 55.0% across 380 matches. The peak was 64.7% in 2023/24, the high point of an era of quicker transitions and longer stoppage time.
Over 2.5 beat its margin-free price in 11 of 21 seasons, so the market is not systematically behind. But the widest miss was 2023/24, priced at 59.3% and landing 64.7%. The market adjusted afterwards, and the priced line has stayed higher since.
View the data: premier league goals lines season by season
| Over 2.5 actual | Over 2.5 priced | |
|---|---|---|
| 05/06 | 45.0% | 46.3% |
| 06/07 | 46.8% | 46.3% |
| 07/08 | 45.0% | 48.0% |
| 08/09 | 47.9% | 48.4% |
| 09/10 | 52.6% | 50.0% |
| 10/11 | 54.2% | 50.3% |
| 11/12 | 53.7% | 52.3% |
| 12/13 | 55.8% | 53.4% |
| 13/14 | 51.6% | 51.8% |
| 14/15 | 48.4% | 49.9% |
| 15/16 | 52.9% | 49.4% |
| 16/17 | 54.2% | 51.6% |
| 17/18 | 51.1% | 51.3% |
| 18/19 | 53.7% | 53.4% |
| 19/20 | 52.1% | 54.2% |
| 20/21 | 50.0% | 52.2% |
| 21/22 | 53.9% | 52.6% |
| 22/23 | 52.6% | 53.2% |
| 23/24 | 64.7% | 59.3% |
| 24/25 | 56.6% | 59.1% |
| 25/26 | 55.0% | 54.1% |
For me the lesson is about timing. When the tempo of the league shifts, prices follow, but they can lag for a season. Watching early-season goal rates against what is being priced is worth more than any long-run average.
§ 4Tottenham games run open, Everton games stay tight
Across 2021/22-2025/26 the league over 2.5 rate was 56.6%, and clubs spread widely around it. Tottenham matches went over most often at 64.7%, with 3.22 goals per game, and they got there by conceding plenty as well as scoring. Liverpool and Manchester City both sat at 62.6%, but their overs came mainly from their own attacks.
At the other end, Everton matches went over in just 44.2% of 190 games, averaging 2.49 goals. Crystal Palace at 48.9% and Burnley at 49.1% followed. These are sides whose games tend to settle into a slower, more controlled rhythm.
| Club | Matches | Over 2.5 | Goals/game | Scored | Conceded | vs league |
|---|---|---|---|---|---|---|
| Tottenham | 190 | 64.7% | 3.22 | 1.71 | 1.51 | +8.1 |
| Liverpool | 190 | 62.6% | 3.22 | 2.13 | 1.09 | +6.0 |
| Manchester City | 190 | 62.6% | 3.21 | 2.31 | 0.91 | +6.0 |
| Leeds | 114 | 62.3% | 3.09 | 1.22 | 1.87 | +5.7 |
| Leicester | 114 | 62.3% | 3.10 | 1.28 | 1.82 | +5.7 |
| Nottingham Forest | 152 | 50.7% | 2.80 | 1.27 | 1.53 | −5.9 |
| Wolves | 190 | 50.0% | 2.65 | 1.05 | 1.59 | −6.6 |
| Burnley | 114 | 49.1% | 2.80 | 0.99 | 1.81 | −7.5 |
| Crystal Palace | 190 | 48.9% | 2.60 | 1.26 | 1.34 | −7.7 |
| Everton | 190 | 44.2% | 2.49 | 1.08 | 1.41 | −12.4 |
League over 2.5 rate in the same seasons: 56.6%. 21 clubs qualified; top 5 and bottom 5 shown.
Club tendencies are well known, so I expect them to be largely priced in. I use this table to understand why a game might open up or stay closed, such as a leaky defence or a dominant attack, rather than as a shortcut to a bet.
Almanac verdict
- Respect the margin first
- Backing every over returned −4.7% and every under −6.5%, so a view needs to be sharper than the price before it is worth staking.
- Look hardest near evens
- The 1.70-1.89 band was the only large band where overs landed clearly above the price, 55.5% against 52.8%. That is where I test a strong tempo read.
- Watch the season's tempo
- In 2023/24 overs landed 64.7% against a 59.3% price. Compare early-season goal rates with the priced line before trusting long-run averages.
Questions bettors ask
How often do Premier League matches go over 2.5 goals?
Across 7,980 Premier League matches from 2005/06-2025/26, 52.3% went over 2.5 goals. That makes fair odds about 1.91, with an average of 2.74 goals per game.
Is it profitable to back over 2.5 in every Premier League game?
Not historically. Flat-stake backing of every over 2.5 returned −4.7% across 7,978 priced matches, and every under returned −6.5%.
Which Premier League teams have the most over 2.5 games?
From 2021/22-2025/26, Tottenham matches went over 2.5 most often at 64.7%, ahead of Liverpool and Manchester City at 62.6%. Everton had the lowest rate at 44.2%.
