Which Match Stat Predicts Results Best? Shots, On Target or Corners
Shots, shots on target, corners, goals or points: I tested which pre-match average best picks a winner, and how each compares with the market's own favourite.
- 63.1%Decisive games won by side with better last-10 SOT diff
- 60.7%Same test using corner difference
- 68.7%Decisive games won by the odds favourite
- 60,274Matches tested
Every bettor has a favourite pre-match number, and most of us defend it more fiercely than the evidence allows. I wanted to settle a simple question: which recent average, whether shots, shots on target, corners, goals or points, does the best job of naming the side that goes on to win? The test covers 21 European divisions from 2016/17-2025/26, using 60,274 matches where both teams had a long enough history and a pre-match price.
The short answer is that shots-on-target difference over the last 10 matches comes out on top, picking the winner in 63.1% of decisive games, with corners clearly last at 60.7%. The longer answer, and the one that matters more if you use these numbers alongside our 1X2 betting tips, is that the pre-match odds favourite beat every one of them at 68.7%. The stats are useful, but the price already knows most of what they know.
Q1Which last-10 stat picks the winner most often?
The bars show, for each measure, how often the side with the better last-10 figure won when the match had a winner. Shots-on-target difference leads at 63.1%, with total shot difference at 62.8% and goal difference at 62.4% close behind. Points manage 61.6% and corner difference trails at 60.7%.
One caution belongs here and applies to everything that follows. Draws are stripped out of these figures, so none of them is a strike rate for a 1X2 bet; they measure ranking quality, not profit.
View the data: better last-10 figure: share of decisive games won
| Matches with a winner; side with the better previous-10 average | |
|---|---|
| Points | 61.6% |
| Goal difference | 62.4% |
| Shot difference | 62.8% |
| Shots-on-target difference | 63.1% |
| Corner difference | 60.7% |
| Pre-match odds favourite | 68.7% |
What strikes me is how tightly the shot-based measures bunch together, and how far the odds favourite sits above them. The gap between the best stat and the worst is small; the gap between the best stat and the market is not. If your edge rests on a single average, the price has probably absorbed it already.
Q2Should I use three, five, ten or twenty games?
This chart tracks accuracy as the sample grows from the last 3 matches to the last 20. Shots-on-target difference climbs from 59.5% to 63.3%, but nearly all of that gain has arrived by the last 10, where it sits at 63.1%. Goal difference starts lower at 59.0% and keeps rising to 63.2%, almost catching up.
View the data: does a longer sample help? accuracy by window
| Shots-on-target difference | Goal difference | Corner difference | |
|---|---|---|---|
| Last 3 | 59.5% | 59.0% | 57.4% |
| Last 5 | 61.4% | 60.8% | 59.2% |
| Last 10 | 63.1% | 62.4% | 60.7% |
| Last 20 | 63.3% | 63.2% | 61.6% |
My reading is that shots on target settle quickly while goals need more time to wash out luck, which is exactly what you would expect from a noisier stat. Corners improve too, from 57.4% to 61.6%, but never close the distance. A three-game snapshot is the weakest view on every line, and it is the one most previews lean on.
Q3How much does a big shots-on-target edge really matter?
Here every match is split into fifths by the home side's last-10 edge, and each dot shows the home win rate. When the home team's shots-on-target edge sat in the top fifth, it won 63.0% of matches; in the bottom fifth that fell to 26.8%. Corners run from 28.8% to 59.3% across the same split.
View the data: home win % by last-10 edge, quintile 1 (worst) to 5
| Corner edge | Shots-on-target edge | |
|---|---|---|
| Q1 (12,169 games) | 28.8% | 26.8% |
| Q2 (12,180 games) | 37.9% | 36.6% |
| Q3 (12,314 games) | 43.6% | 42.3% |
| Q4 (11,964 games) | 47.9% | 49.6% |
| Q5 (11,647 games) | 59.3% | 63.0% |
The spread is the real story: 36.2 points from bottom to top for shots on target against 30.5 for corners. Both measures sort teams sensibly, but shots on target separate the extremes more sharply, particularly at the strong end where it matters most to anyone backing a home side.
Q4How closely does each pre-match stat track the final scoreline?
The table adds a second lens: the correlation between each pre-match difference and the final goal difference. Shot difference and shots-on-target difference both reach 0.33, goal difference 0.32, points 0.29 and corners 0.28. The odds-based measure stands at 0.42.
That ordering mirrors the accuracy column almost exactly, which gives me more confidence that the result is not a quirk of how winners were counted. Shot quality and volume carry slightly more signal than results alone, corners carry the least, and the market carries the most.
| Pre-match measure | Decisive games | Picked winner | r with goal diff |
|---|---|---|---|
| Points | 42,105 | 61.6% | 0.29 |
| Goal difference | 42,746 | 62.4% | 0.32 |
| Shot difference | 44,194 | 62.8% | 0.33 |
| Shots-on-target difference | 43,720 | 63.1% | 0.33 |
| Corner difference | 43,801 | 60.7% | 0.28 |
| Pre-match odds (home minus away win chance) | 44,290 | 68.7% | 0.42 |
The playbook
How we would use these numbers before the next match.
- Rank by shots on target. When comparing two sides, start with last-10 shots-on-target difference, which picked 63.1% of decisive winners against 60.7% for corners. Treat corners as context, not a verdict.
- Ignore three-game form. The last 3 matches were the weakest window for every stat, with shots on target at just 59.5%. Use at least the last 10 before forming a view.
- Start from the price. The odds favourite won 68.7% of decisive games and correlated at 0.42 with the final margin. Use the stats to question a price, not to replace it.
Data & method
- Data: 21 European divisions, 2016/17-2025/26, 67,047 matches with full match statistics.
- Each team's averages use only its previous matches (shifted by one game) within the 2016/17-2025/26 window; seasons are not reset.
- Differences are home average minus away average; "picked the winner" counts decisive matches (no draw) in which the side with the higher figure won; ties on a measure are excluded.
- All windows (3, 5, 10, 20) are compared on the same set of matches: both teams needed 20 previous matches and a pre-match price.
- The odds measure removes the margin proportionally from a major bookmaker's 1X2 prices.
- r is the Pearson correlation between the pre-match difference and the final home goal difference.
- Caveat: Draws are left out of the accuracy measure, so the percentages are not win rates for a bet.
- Caveat: Averages carry across a summer break and divisional moves, which slightly blurs early-season figures.
Quick Q&A
- Is shots on target a better predictor than goals in football?
- Slightly, over a last-10 window: shots-on-target difference picked 63.1% of decisive winners against 62.4% for goal difference. Over the last 20 matches the two almost converge, at 63.3% and 63.2%.
- Do corners predict who wins a football match?
- Only weakly compared with the alternatives. The side with more corners over its last 10 matches won 60.7% of decisive games, the lowest of the stats tested, with a correlation of 0.28 to the final goal difference.
- Can match stats beat the bookmaker's favourite?
- Not on their own in this study. The pre-match odds favourite won 68.7% of decisive games, comfortably ahead of the best single stat, shots-on-target difference, at 63.1%.
I'm Eleanor Crow, editor at Soccers Tips. After a decade writing football betting previews I now lead the team and keep our analysis honest across every market we cover.
More from Eleanor Crow →