Since June we have measured the real 1X2 margin of every bookmaker we track — the gap between what its three prices imply and 100% — from our own odds snapshots, match by match. This week we asked a simple question of the data: over the summer, did bookmakers get more expensive?
Ensimmäinen vastaus oli kyllä, selvästi. Toinen vastaus, kun asia tehtiin oikein, oli ei. Näiden kahden välinen ero on tämän artikkelin aihe, koska sama virhe on sisäänrakennettuna useimpiin väitteisiin tyyliin "bookmaker X nosti marginaalejaan", joita tulet lukemaan — myös siihen taulukkoon, jonka aiomme julkaista tällä viikolla.
Average margin of the 31 bookmakers with at least 50 priced matches in both June and August:
| June | July | August | Change | |
|---|---|---|---|---|
| Market average (31 bookmakers) | 7.63% | 8.55% | 8.67% | +1.04 pp |
| Pinnacle (sharp benchmark) | 5.45% | 6.51% | 6.87% | +1.42 pp |
Ranked by the change, the top of the table looks like a story: Unibet (FR) +5.06 pp, PMU (FR) +4.29 pp, Winamax (FR) +3.24 pp, Winamax (DE) +2.97 pp. Even the sharpest book in the set drifted up by 1.4 points.
It would have made a good headline. It is also almost entirely false.
Margin depends heavily on the competition. On the matches we priced in the last 30 days, the market's average margin runs from about 5.8% in the Premier League (La Liga 5.89%, Champions League 5.92%, Ligue 1 5.98%, Serie A 6.08%) up to 11–12% in the smallest competitions we cover (Argentina's Primera C 11.60%, Hungary NB II 11.39%, Lithuania's 1 Lyga 11.39%, Bulgaria's Second League 11.18%). A bookmaker prices a fourth-division match in Argentina at roughly double the margin it charges on Arsenal.
In June our coverage was a summer-only set of leagues. In July we added dozens of smaller competitions, and in August the big European seasons started. The mix of matches behind the "average margin" changed completely between the first column and the last — so the average changed, whether or not any single price did.
To ask the real question you have to hold the league constant: compare each bookmaker's margin on the same competition in June and in August. Only four leagues have enough matches in both months to do that (the summer leagues: Norway's 1. Division, Sweden's Superettan, Argentina's Primera Nacional and one more), giving 41 bookmaker-league pairs with at least 20 priced matches on each side.
| Naive (all matches) | Like-for-like (same leagues) | |
|---|---|---|
| Median change, June → August | +1.04 pp | +0.06 pp |
| Mean change | — | +0.19 pp |
| Pairs that went up | — | 32 of 41, most by less than 0.2 pp |
On the same leagues, the median bookmaker's margin moved by six hundredths of a point. Norway's 1. Division drifted up 0.15 pp across 12 bookmakers; Superettan and Primera Nacional by 0.04 pp. The one bookmaker that genuinely moved is Bet365, +1.37 pp on its three comparable leagues — a real change, on a small base. Everyone else's dramatic "+3 to +5 points" was the calendar.
So: did bookmakers get more expensive this summer? Within our measurement error, no. Did betting get more expensive for someone who followed our coverage from June to August? Yes — because they were now looking at Lithuanian and Argentine lower-league prices next to Premier League ones.
Every comparison of margins across time, across sites or across "offers" has this problem. A book that markets itself on low margins can be genuinely cheap on the Premier League and among the most expensive on everything else; a "margins are rising" chart can be a coverage change; a "we beat bookmaker X by 2 points" claim depends entirely on which matches were counted. The spread between bookmakers on the same league is enormous — on the Premier League this month, 21 bookmakers range from 1.55% to 10.15%; on Superettan from about 0% to 12.7% — and it dwarfs the change of any bookmaker over time.
That is why each bookmaker now has its own page here, with its margin broken down by league and its 90-day trend measured on a fixed window: bet365, Pinnacle, Unibet, William Hill, Matchbook — all 33 are linked from the Bookmaker Margin Index. The number that costs you money is the margin on the league you actually bet, at the book you actually use. The average is for headlines.
Method notes: margin per bookmaker per match = sum of the implied probabilities of its most recent pre-match 1X2 prices, minus 1; only complete three-way prices with an overround between 0.9 and 1.4 are counted. Monthly figures use the matches whose last price was captured in that month. Like-for-like pairs require at least 20 priced matches for the bookmaker on that league in both months. Exchanges are excluded (they charge commission, not margin). All figures recomputed from the public odds snapshots; the live index updates hourly.