HT/FT by Elo Rating Gap: How All 9 Results Split
I split 133,988 matches by Elo rating gap to see how the nine half-time/full-time results move as one side grows stronger, and where the fair prices sit.
- 54.2%Home/Home with a 200+ home Elo edge
- 18.8%Draw/Draw in evenly rated matches
- 37.9%Away/Away with a 200+ away Elo edge
- 133,988Matches with Elo ratings for both sides
HT/FT is a market about how a match unfolds, so the first thing I want to know is how much strength decides that shape. The approach behind my half time/full time predictions page starts with one question: as the Elo gap between two sides widens, which of the nine results gains and which fades away? I put that question to 133,988 matches across 22 European divisions.
The headline is clear. With a home edge of 200 points or more, Home/Home lands 54.2% of the time. With the same edge for the visitors, Away/Away reaches 37.9%. In evenly rated games, Draw/Draw peaks at 18.8%. Strength shapes the half-time picture strongly, while full reversals barely move at all.
Q1How do all nine HT/FT results move as the gap widens?
Read the heatmap left to right, from the away side much stronger to the home side much stronger. Home/Home climbs steadily from 9.0% to 54.2%, and Away/Away mirrors it, falling from 37.9% to 3.7%. Those two rows tell you most of what the rating gap does.
The quieter rows interest me more. Draw/Draw peaks at 18.8% in even matches, then drops to 8.5% with a big home edge and 13.0% with a big away edge. Draw/Home rises to 21.4% in the strongest home band, and Draw/Away reaches 19.9% at the opposite end. A favourite level at the break is a common path to a win, not a sign that something has gone wrong.
View the data: all 9 ht/ft results by elo gap
| Below −200 | −200 to −100 | −100 to −30 | −30 to +30 | +30 to +100 | +100 to +200 | +200 or more | |
|---|---|---|---|---|---|---|---|
| Home/Home | 9.0% | 16.0% | 21.5% | 25.4% | 30.6% | 38.5% | 54.2% |
| Home/Draw | 4.2% | 5.3% | 5.8% | 5.7% | 5.3% | 4.9% | 3.3% |
| Home/Away | 3.3% | 2.8% | 2.3% | 2.0% | 1.6% | 1.2% | 0.5% |
| Draw/Home | 6.5% | 11.3% | 13.9% | 16.0% | 18.0% | 20.2% | 21.4% |
| Draw/Draw | 13.0% | 17.3% | 18.6% | 18.8% | 17.3% | 14.0% | 8.5% |
| Draw/Away | 19.9% | 15.3% | 12.2% | 10.5% | 8.2% | 5.9% | 2.6% |
| Away/Home | 1.3% | 2.0% | 2.3% | 2.4% | 2.9% | 3.4% | 3.0% |
| Away/Draw | 4.9% | 5.4% | 5.6% | 5.3% | 5.1% | 4.1% | 2.8% |
| Away/Away | 37.9% | 24.6% | 17.9% | 13.9% | 11.2% | 7.8% | 3.7% |
For context, across every match Home/Home was the most common result at 26.9%, ahead of Draw/Draw at 16.6% and Away/Away at 15.6%. Once you split by rating gap, those averages hide a great deal.
Q2How often is the stronger side already ahead at half-time?
This chart shows the half-time state behind every HT/FT bet. With a 200-point edge, the home side led at the break in 58.1% of matches, compared with 16.5% when it was 200 points weaker. A strong away side led at half-time 44.1% of the time.
The level share is what I watch. It ranges from 32.5% to 45.2% and peaks in evenly rated matches. In those even games, the home side still led at the break 33.1% of the time against 21.7% for the visitors. The method adds no home-advantage adjustment to the ratings, so a gap of 0 still leans towards the hosts.
View the data: who leads at half-time, by elo gap
| Home leads | Level | Away leads | |
|---|---|---|---|
| Below −200 | 16.5% | 39.4% | 44.1% |
| −200 to −100 | 24.1% | 43.8% | 32.1% |
| −100 to −30 | 29.6% | 44.6% | 25.8% |
| −30 to +30 | 33.1% | 45.2% | 21.7% |
| +30 to +100 | 37.4% | 43.5% | 19.1% |
| +100 to +200 | 44.6% | 40.1% | 15.3% |
| +200 or more | 58.1% | 32.5% | 9.4% |
The practical lesson is that even with a large gap, a big share of first halves still finish level. That is why Draw/Home sits at 20.2% in the +100 to +200 band. Patient favourites are a real pattern, and the market sometimes forgets it.
Q3Does a bigger rating gap make a full reversal less likely?
Full reversals, where Home/Away or Away/Home swing the result, happened in 4.5% of all matches. Across the bands they ranged only from 3.5% to 4.8%. Most bands sit in a narrow strip around the overall figure.
The exception is the strongest home band, at 3.5%, the lowest of any. A strong away side tells a different story: reversals ran at 4.6% there. Home/Away alone was 3.3% when the visitors were 200+ points stronger, against 0.5% with a 200+ home edge. A home side that nicks a lead against a much better opponent is often caught.
View the data: full reversals (home/away + away/home) by elo gap
| % of all matches in each band | |
|---|---|
| Below −200 | 4.6% |
| −200 to −100 | 4.8% |
| −100 to −30 | 4.6% |
| −30 to +30 | 4.5% |
| +30 to +100 | 4.5% |
| +100 to +200 | 4.6% |
| +200 or more | 3.5% |
I treat reversals as long shots whose likelihood barely depends on the mismatch. That steadiness is useful. It means the price, rather than the size of the gap, decides whether a reversal is worth a look.
Q4What fair price does each HT/FT result deserve by band?
Fair odds here are simply 1 divided by the observed rate, with no margin. They mark the break-even line, not a target. Home/Home with a 200-point home edge is fair at 1.85, and Away/Away with the same away edge is fair at 2.64. Step down to the 100 to 200 bands and those become 2.60 and 4.07.
Draw/Draw is fair at 5.32 in even matches, but drifts out to 11.76 when the home side is much stronger. My main caveat is that Elo is a strength estimate, not a price. The extreme bands also lump together a 210-point gap and a 450-point gap, and they mix leagues and eras.
| Elo gap | Matches | Home/Home | Draw/Draw | Away/Away |
|---|---|---|---|---|
| Below −200 | 8,641 | 9.0% · 11.11 | 13.0% · 7.69 | 37.9% · 2.64 |
| −200 to −100 | 17,137 | 16.0% · 6.25 | 17.3% · 5.78 | 24.6% · 4.07 |
| −100 to −30 | 26,594 | 21.5% · 4.65 | 18.6% · 5.38 | 17.9% · 5.59 |
| −30 to +30 | 29,230 | 25.4% · 3.94 | 18.8% · 5.32 | 13.9% · 7.19 |
| +30 to +100 | 26,663 | 30.6% · 3.27 | 17.3% · 5.78 | 11.2% · 8.93 |
| +100 to +200 | 17,083 | 38.5% · 2.60 | 14.0% · 7.14 | 7.8% · 12.82 |
| +200 or more | 8,640 | 54.2% · 1.85 | 8.5% · 11.76 | 3.7% · 27.03 |
The playbook
How we would use these numbers before the next HT/FT bet.
- Check the band first. Before pricing any HT/FT bet, place the match in its Elo band. Home/Home moves from 25.4% in even games to 54.2% with a 200-point home edge.
- Respect the level half. Do not assume a favourite leads at the break. Even with a 100 to 200 home edge, Draw/Home ran at 20.2%, so price that path properly.
- Demand value on reversals. Reversals stayed close to 4.5% in nearly every band. Only back one when the offered price clearly beats break-even, and be most sceptical against strong home sides, where they fell to 3.5%.
Data & method
- Data: 22 European divisions, 2000/01–2025/26, 133,988 matches.
- HT/FT outcomes are written half-time result / full-time result from the home side's view (Home/Draw = home led at the break, match drawn).
- Elo gap = home club Elo rating − away club Elo rating before the match; no home-advantage adjustment is added, so a gap of 0 still favours the home side.
- Matches: 22 European divisions, 2000/01–2025/26, where both Elo ratings and the half-time score are recorded.
- Bands use the lower bound inclusive (e.g. −30 to +30 means −30 ≤ gap < +30).
- Fair odds are 1 ÷ observed frequency (no margin); they show the price at which a bet would only break even.
- Caveat: Elo ratings are a strength estimate, not a price; bands with the same gap can still differ in league and era.
- Caveat: The extreme bands mix very different mismatches (a 210-point gap and a 450-point gap sit together).
Quick Q&A
- How often does Home/Home win when the home team is much stronger?
- When the home side was rated 200 or more Elo points stronger, Home/Home landed 54.2% of the time across 8,640 matches. That works out to a fair price of 1.85.
- Is Draw/Draw more likely in evenly matched games?
- Yes. Draw/Draw peaked at 18.8% when the Elo gap was within 30 points, compared with 8.5% when the home side had a 200-point edge.
- How common are HT/FT reversals like Home/Away or Away/Home?
- Full reversals happened in 4.5% of all matches and ranged only from 3.5% to 4.8% across the Elo bands. The lowest rate came when the home side was 200 or more points stronger.
I'm Imogen Vance, and I focus on Half-Time/Full-Time markets, where you can't just ask who wins but have to picture how the ninety minutes actually unfolds.
More from Imogen Vance →
