Observed market data
Do Premier League 1X2 Margins Tighten Before Kick-Off?
Using the same 10 opening-weekend matches from the 2025/26 Premier League season, we compared opening and closing 1X2 book percentages to see whether the market consistently became tighter by kick-off.
The question
It is common to hear that football markets become more efficient or competitive as kick-off approaches. That can be true in some contexts, but it is easy to turn a plausible idea into an unsupported rule.
We therefore asked a narrower, testable question: did the 1X2 book percentage decrease between the opening and closing price snapshots on Premier League opening weekend 2025/26?
How we measured it
For each match, we calculated the book percentage from three decimal prices at the opening snapshot and again at the closing snapshot.
Book percentage = (1 ÷ home odds + 1 ÷ draw odds + 1 ÷ away odds) × 100
Change = closing book percentage − opening book percentage
A negative change means the book became tighter. A positive change means the book percentage increased.
Bet365 opening vs closing books
| Match | Opening | Closing | Change |
|---|---|---|---|
| Liverpool v Bournemouth | 105.35% | 104.63% | -0.72 pp |
| Aston Villa v Newcastle | 107.50% | 105.94% | -1.56 pp |
| Brighton v Fulham | 105.13% | 105.68% | +0.54 pp |
| Sunderland v West Ham | 104.63% | 105.11% | +0.48 pp |
| Tottenham v Burnley | 105.28% | 106.16% | +0.88 pp |
| Wolves v Man City | 106.02% | 105.17% | -0.85 pp |
| Chelsea v Crystal Palace | 107.05% | 105.32% | -1.73 pp |
| Nott'm Forest v Brentford | 104.33% | 105.01% | +0.68 pp |
| Man United v Arsenal | 106.84% | 105.70% | -1.14 pp |
| Leeds v Everton | 105.65% | 105.58% | -0.07 pp |
Six of the 10 Bet365 books were lower at the closing snapshot and four were higher. The average change was -0.35 percentage points.
The market-average picture was even flatter
The cross-bookmaker average barely moved overall. That is an important reminder that “prices moved” and “the book became much tighter” are not the same statement.
Why the best-available book behaved differently
We also calculated a “best available” book by taking the maximum listed home, draw and away price across the source bookmaker set. That book averaged 101.74% at the opening snapshot and 102.10% at closing.
Seven of the 10 best-available books increased rather than decreased. This does not contradict the single-book result because it measures something different: it combines the best price for each outcome, potentially from three different bookmakers. When the identity of the best price changes, the combined book can move in a different direction from any one bookmaker's book.
The biggest moves in the sample
- Chelsea v Crystal Palace: Bet365 fell from 107.05% to 105.32%, a 1.73-point tightening.
- Aston Villa v Newcastle: 107.50% to 105.94%, down 1.56 points.
- Man United v Arsenal: 106.84% to 105.70%, down 1.14 points.
- Tottenham v Burnley: moved the other way, from 105.28% to 106.16%, up 0.88 points.
That spread is why we do not treat “closing markets are tighter” as a universal rule from this dataset. The average moved slightly lower, but individual matches behaved differently.
What this can and cannot tell us
This 10-match snapshot supports a modest conclusion: the average 1X2 book in this sample was slightly lower at the closing snapshot, but the effect was small and not consistent across matches.
It does not establish a general law about Premier League closing lines. A stronger study would repeat the calculation across hundreds of matches, separate bookmakers, and perhaps compare high-liquidity and lower-liquidity fixtures.
Data source and reproducibility
The underlying opening and closing odds come from Football-Data.co.uk's 2025/26 Premier League CSV. Betting Maths calculated the book percentages and opening-to-closing changes.
The complete derived dataset for all 10 matches is available below so readers can reproduce every headline number in this article.
Limitations
- Only one Premier League round is included.
- Opening and closing are the snapshots supplied in the source dataset; they are not a minute-by-minute record of price movement.
- Book percentage measures the sum of implied probabilities, not realised bookmaker profit.
- The average and maximum fields depend on the bookmaker coverage in the underlying file.
- The sample is descriptive and should not be used as a betting strategy.
