Cycling race predictor · one-day race

GP Industria & Artigianato (no prediction made) settled

One day, one result: the production stack priced the whole field under 1 course reading, against the market's line where one was captured. A one-day race is its own best reference — the main reading is built from this race's own recent editions.

Raced 6 September 2026 1 reading 155 riders priced models retrained d911de60f7 market snapshot no capture

The fork

1 reading of the same road

Course profile is an input the model cannot see from a map — it is set from reference races. The main reading averages this race's own last editions, so the terrain and the kind of rider it selects come from the race's own history; any alternate reading swaps in an archetype set to ask what a different day would do to the same field.

Hindsight (post-hoc)

Hindsight (post-hoc) — oracle-matched references
sprint0.00
mountain0.20
hilly0.00
rolling0.79
break0.01
TT0.00
References: Faun Drome Classic (2023); Trofeo Andratx - Pollença - Trofeo Andratx - Pollença (2020); Liège-Bastogne-Liège (2023); Faun-Ardèche Classic (2022); GP Montreal (2023)
RETROSPECTIVE ONLY — no pre-race prediction was made. This board uses the settled result as the oracle and the nearest historical realized race shapes. It is corpus completion, not evidence of forecasting skill.
Top 5 (blend): Martinez Lenny 38.6%; Pidcock Tom 22.6%; Grégoire Romain 12.3%; Scaroni Cristian 7.8%; Christen Jan 6.3%

The board

Win probabilities across the heads

Blend is the production price (cascade × LambdaRank, DNF-adjusted at the pricing layer); Cascade and LambdaRank are the two underlying heads shown for the main reading. The market columns are the devigged win line, present as the information benchmark the model is scored against.

Read "Max edge" as directional, not as a price. The blend is a set of per-rider marginals and does not sum to one book — across this field each reading totals hindsight 123% — while the devigged market column sums to exactly 100%. Putting the two side by side flatters the model, and Max edge is inflated by about that factor. Of our heads only the LambdaRank one is natively book-preserving.

RiderTeamBlendCascadeLambdaRankMarket oddsMarket devigMax edge
Martinez LennyBAHRAIN VICTORIOUS38.6%35.4%32.1%n/a––
Pidcock TomPINARELLO-Q36.5 PRO CYCLING TEAM22.6%13.6%20.6%n/a––
Grégoire RomainGROUPAMA-FDJ UNITED12.3%13.4%9.7%n/a––
Scaroni CristianXDS ASTANA TEAM7.8%11.1%5.5%n/a––
Christen JanUAE TEAM EMIRATES XRG6.3%4.4%5.6%n/a––
Aranburu AlexCOFIDIS2.5%2.4%2.0%n/a––
Powless NeilsonEF EDUCATION - EASYPOST2.4%2.7%1.9%n/a––
Rondel MathysTUDOR PRO CYCLING TEAM2.1%1.3%1.9%n/a––
Morgado AntónioUAE TEAM EMIRATES XRG1.5%2.2%1.0%n/a––
Izagirre IonCOFIDIS1.4%1.5%1.1%n/a––
Storer MichaelTUDOR PRO CYCLING TEAM1.2%2.4%0.7%n/a––
Kulset JohannesUNO-X MOBILITY1.2%1.2%0.9%n/a––
Crescioli LudovicoTEAM POLTI VISITMALTA1.1%1.4%0.8%n/a––
Breuillard NicolasTOTALENERGIES1.0%1.7%0.7%n/a––
Tesfazion NatnaelMOVISTAR TEAM1.0%1.1%0.8%n/a––
Oldani StefanoCAJA RURAL-SEGUROS RGA1.0%1.3%0.7%n/a––
Hindsight (post-hoc) — oracle-matched references — top 30, all heads
#RiderBlendCascade LambdaRankPodiumTop 15 Market odds
1Martinez Lenny38.6%35.4%32.1%56.9%85.4%n/a
2Pidcock Tom22.6%13.6%20.6%39.4%79.1%n/a
3Grégoire Romain12.3%13.4%9.7%31.8%72.2%n/a
4Scaroni Cristian7.8%11.1%5.5%25.0%66.3%n/a
5Christen Jan6.3%4.4%5.6%16.7%55.8%n/a
6Aranburu Alex2.5%2.4%2.0%11.0%52.5%n/a
7Powless Neilson2.4%2.7%1.9%9.2%40.5%n/a
8Rondel Mathys2.1%1.3%1.9%6.3%54.4%n/a
9Morgado António1.5%2.2%1.0%7.6%32.3%n/a
10Izagirre Ion1.4%1.5%1.1%6.4%39.3%n/a
11Storer Michael1.2%2.4%0.7%11.7%43.0%n/a
12Kulset Johannes1.2%1.2%0.9%5.8%47.3%n/a
13Crescioli Ludovico1.1%1.4%0.8%5.3%38.4%n/a
14Breuillard Nicolas1.0%1.7%0.7%4.3%30.6%n/a
15Tesfazion Natnael1.0%1.1%0.8%3.9%46.9%n/a
16Oldani Stefano1.0%1.3%0.7%5.3%34.1%n/a
17Tiberi Antonio0.9%1.6%0.6%4.6%34.3%n/a
18Delettre Alexandre0.8%0.7%0.7%5.1%42.8%n/a
19Fancellu Alessandro0.8%1.1%0.6%8.8%33.7%n/a
20Castellon Jan0.8%1.2%0.5%4.3%24.0%n/a
21Arrieta Igor0.6%0.7%0.5%2.6%22.8%n/a
22Higuita Sergio0.6%0.7%0.5%3.0%25.5%n/a
23Vercher Mattéo0.6%0.9%0.4%3.4%27.3%n/a
24Jegat Jordan0.6%0.5%0.5%4.2%24.8%n/a
25Champoussin Clément0.5%0.7%0.4%4.1%23.2%n/a
26O'Connor Ben0.5%0.7%0.3%2.4%27.5%n/a
27Gachignard Thomas0.5%0.6%0.3%1.7%12.5%n/a
28Hermans Quinten0.4%0.3%0.4%1.1%10.2%n/a
29Ulissi Diego0.4%0.3%0.4%1.3%24.3%n/a
30Carr Simon0.4%0.8%0.2%2.0%16.0%n/a

After the race

Result — won by Pidcock Tom

PosRiderTeamHindsight (post-hoc) win p
1Pidcock TomPINARELLO-Q36.5 PRO CYCLING TEAM22.6%
2Scaroni CristianXDS ASTANA TEAM7.8%
3Charmig AnthonUNO-X MOBILITY0.1%
4Adrià RogerMOVISTAR TEAM0.3%
5Fancellu AlessandroMBH BANK CSB TELECOM FORT0.8%
6Tiberi AntonioBAHRAIN VICTORIOUS0.9%
7Gachignard ThomasTOTALENERGIES0.5%
8Rondel MathysTUDOR PRO CYCLING TEAM2.1%
9O'Connor BenTEAM JAYCO ALULA0.5%
10Grégoire RomainGROUPAMA-FDJ UNITED12.3%

Scoring — probability given to the actual winner

ReadingBlendCascade LambdaRankMarket devig
Hindsight (post-hoc)22.63%13.61%20.59%–