Odds are a probability written backwards: a price of 1.50 means the bookmaker rates the outcome at roughly 67% (1 / 1.50) and adds a little on top as margin. There is one way to check how honest those estimates are: take many matches, group them by odds, and see how often the favourite actually won. The table below does exactly that on the TheAnalytic database and is recomputed every day.
The table
| Favourite odds | Matches | Favourite won | Implied by odds | Difference |
|---|---|---|---|---|
| 1.30 – 1.50 | 20 | 65.0% | 71.2% | -6.2 pp |
| 1.50 – 1.70 | 31 | 58.1% | 62.8% | -4.7 pp |
| 1.70 – 1.90 | 39 | 56.4% | 55.7% | +0.7 pp |
| 1.90 – 2.20 | 50 | 48.0% | 48.5% | -0.5 pp |
| 2.20 – 2.60 | 70 | 44.3% | 42.0% | +2.3 pp |
Finished matches of top leagues from 06.09.2025 to 06.09.2026, 221 matches; the table is recomputed daily. The favourite is the side with the lower closing odds at Pinnacle; “implied by odds” is the average 1/odds without removing the margin.
How to read it
- Favourite odds is the band of Pinnacle closing odds on the side that was cheaper. The draw is never the favourite.
- Favourite won is the share of matches in the band where the favourite won in regular time.
- Implied by odds is the average implied probability 1/odds over the band’s matches. The margin is not removed, so it is slightly inflated: in an honest market the actual win rate should be a little below the implied one.
- Difference is actual minus implied, in percentage points. Around zero or slightly negative means the market is calibrated; a clear plus means favourites in that band are underrated; a clear minus means they are overrated.
What the numbers usually show
On large samples the closing prices of low-margin bookmakers are calibrated to within one or two percentage points in every band. That is the main argument for CLV as a measure of pick quality: if the closing line is an almost honest probability, buying an outcome cheaper than the close means buying it with positive expectation.
The systematic deviations described in the literature, and sometimes visible in the table, are small: strong favourites (1.01–1.30) are slightly underrated, long-odds underdogs are overrated — the so-called favourite-longshot bias. On short samples, especially early in the season, the difference can look large in either direction; check the number of matches in the row.
What the table does not say
“A 1.50 favourite wins 65% of the time” does not mean favourites are profitable to back: 65% wins at 1.50 yields an ROI of about minus 2–3%, exactly the margin. The table answers a different question: can the closing line be trusted as a probability estimate. The answer is usually “yes” — and that is precisely why a tipster who consistently beats the close is worth something, and a tipster with a pretty ROI and negative CLV is not.
About the sample
Finished top-league matches over the period stated below the table that have a Pinnacle closing price on both sides. Bands with fewer than 20 matches are not shown. Drawn matches count as “favourite did not win”. Where the odds come from and what the closing line is are covered in the articles on data sources and on the opening and closing line.