Bayonne vs Montpellier Herault
What the model says about this fixture, at either ground and at a neutral one, with the record behind it. Ratings as at 2026-08-24.
Montpellier Herault 81.5%draw 2.5%16.0% BayonneEach figure is 5,000 simulations of this fixture. The score is where those runs are centred, not a likely scoreline — a rugby result is spread far too widely for any single one to be likely.
How the runs finished
Winning margin across 5,000 simulations of this fixture. Bars right of the centre are Bayonne wins; the faded bars at each end hold everything beyond 75 points.
Margin percentiles
Eight runs in ten finished between 23 Montpellier Herault and 14 Bayonne, with the middle run at 5 Montpellier Herault.
Those are gaps rather than scorelines. How much scoring the match produces is drawn separately — between 32 and 74 points in eight runs out of ten — so one bar covers many scores. The middle run above is 23-18 on a tight afternoon and 34-29 on an open one.
Why Bayonne are not favoured
Starting from an even match and adding what the model reads, in percentage points of win probability. Positive favours Bayonne, negative favours Montpellier Herault.
Bayonne 1413, Montpellier Herault 1882 — 468 points, favouring Montpellier Herault
Bayonne are at home
Recent results, head-to-head record and rest days, as the XGBoost heads weigh them. Deliberately capped, so it can nudge the rating but never overrule it.
Head to head
15 meetings, 2015 to 2026 · Montpellier Herault by 9.1 on average
Last 10: 5–0–5 · Montpellier Herault by 7.1 on average
| Date | Match | Competition | Venue |
|---|---|---|---|
| 2026-01-03 | Montpellier Herault 62–22 Bayonne | Top 14 | Septeo Stadium |
| 2025-09-13 | Bayonne 26–23 Montpellier Herault | Top 14 | Stade Jean Dauger |
| 2025-01-04 | Montpellier Herault 42–10 Bayonne | Top 14 | Septeo Stadium |
| 2024-09-28 | Bayonne 28–27 Montpellier Herault | Top 14 | Stade Jean Dauger |
| 2024-02-24 | Montpellier Herault 28–23 Bayonne | Top 14 | Septeo Stadium |
The 5 most recent meetings; the record above covers all 15. None of this feeds the forecast: both ratings already contain these results.