Skip to content

Downhill forecast

Current forecast: 2026 Men Elite.

Model run 1287 (2026-09-27), with data through Whistler Timed Training and the Q1 start list. The model is fit on everything before the Whistler weekend, using 2021–2026 elite and junior history. How accurate the model has been: Downhill evaluation. For Formula 1, see F1 forecast.

2026 title odds with 2 rounds left (Whistler in progress, then one more round): Williams 71.4%, Vermette 18.6%, Iles 4.3%, Pierron 4.0% (full forecast).

Placeholder points

Expected points, standings, champion odds and rank-movement probabilities use placeholder points tables. Win, podium, top-10 and make-Final odds don't.

Next round: Whistler (DHI #8, 25–27 Sep), in progress

There are 109 starters from the Q1 start list. Timed Training was ignored by the safety check (residual IQR 0.71): riders were held on track, so finish times don't reflect pace. These odds therefore carry no Whistler-specific information.

Rider Win Podium Top 10 Make Final Exp pts Rank now Rank up Rank down
WILLIAMS Jordan 6.7% 16.8% 40.2% 75.3% 91 1 0.0% 18.9%
VERMETTE Asa 5.6% 14.9% 37.3% 73.7% 84 2 13.9% 34.8%
ALRAN Max 5.6% 15.1% 39.2% 73.1% 86 6 28.7% 49.8%
BRUNI Loic 5.1% 13.4% 35.4% 71.0% 80 26 48.4% 33.0%
ILES Finn 5.0% 12.6% 34.1% 69.3% 76 3 20.1% 45.5%
PINKERTON Ryan 4.5% 12.5% 33.5% 69.0% 75 8 46.7% 24.9%
PIERRON Amaury 4.3% 12.4% 33.9% 69.5% 76 4 33.5% 30.7%
ALRAN Till 4.3% 12.5% 34.5% 69.5% 77 11 39.8% 38.7%

Projected final standings

Rider Rank now Pts now Exp final pts p10–p90 Champion Top 3
WILLIAMS Jordan 1 1177 1364 1209–1532 71.4% 99.2%
VERMETTE Asa 2 1042 1215 1065–1383 18.6% 82.6%
ILES Finn 3 962 1091 962–1252 4.3% 40.3%
PIERRON Amaury 4 928 1085 944–1243 4.0% 40.4%
GOLDSTONE Jackson 5 822 962 822–1122 0.7% 12.4%
ALRAN Max 6 773 952 798–1118 0.5% 12.2%
BROSNAN Troy 7 760 905 770–1060 0.2% 5.4%
PINKERTON Ryan 8 743 898 759–1061 0.2% 5.6%

The remaining unknown round uses the attendance model (riders who started the last 3 completed rounds).

2026 so far: how this season's rounds were called

Each round was predicted from everything before it (walk-forward backtest run 938, the same numbers as Downhill evaluation):

Round Spearman Brier make-Final (model / base) Top-10 hits Winner (win prob)
Mona Yongpyong 0.646 0.135 / 0.231 6 Vermette (21.3%)
Loudenvielle 0.611 0.128 / 0.210 4 Shaw (3.1%)
Leogang 0.671 0.113 / 0.202 4 Iles (2.6%)
Lenzerheide 0.679 0.111 / 0.204 5 Iles (4.9%)
La Thuile 0.674 0.100 / 0.194 5 Williams (4.0%)
Pal Arinsal 0.679 0.121 / 0.222 6 Williams (4.6%)
Les Gets 0.701 0.106 / 0.201 7 M. Alran (4.5%)

Holdout: rounds 6–7 and the standings, predicted from data up to round 5

Rider Pts after R5 Exp pts after R7 Leader after R7 Top 3 Actual pts Actual rank
PIERRON Amaury 844 1027 51.0% 93.7% 928 4
VERMETTE Asa 755 926 20.7% 67.1% 1042 2
WILLIAMS Jordan 752 916 18.7% 64.1% 1177 1
ILES Finn 782 866 6.7% 45.7% 962 3
BROSNAN Troy 539 681 0.8% 7.3% 760 7
VERGIER Loris 508 669 0.9% 6.8% 627 9

Standings Spearman: 0.962. Williams took the lead by scoring 425 points in rounds 6–7 (a win at Pal Arinsal and 7th at Les Gets), which the model gave an 18.7% chance.

This table is from the 2026-09-26 forecast run, before the database was rebuilt; per-rider holdout rows aren't stored. Backtest run 938's holdout of the same rounds gives the same standings Spearman (0.962) and Pierron 51.0%, and Williams 17.7% (see Downhill evaluation).

How to refresh it

After each session: re-download, ingest and re-run (see Web app).

racinglines data ingest mtb_dh data/raw/mtb_dh/chronorace
racinglines model forecast mtb_dh --db --save --backtest 0 --remaining 2   # 10k sims, seed 42