Cycling race predictor Β· one-day race

GP Montreal (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 β€” the main reading is built from this race's own recent editions.

Raced 13 September 2026 1 reading 149 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 β€” 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) β€” oracle-matched references
sprint0.00
mountain0.16
hilly0.02
rolling0.68
break0.15
TT0.00
References: GP Montreal (2025); GP Montreal (2024); GP Montreal (2023); GP Montreal (2022); GP Montreal (2019)
RETROSPECTIVE ONLY β€” 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): Seixas Paul 34.0%; Evenepoel Remco 33.3%; del Toro Isaac 23.3%; Pidcock Tom 10.8%; Ayuso Juan 3.3%

The board

Win probabilities across the heads

Blend is the production price (cascade Γ— 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 β€” across this field each reading totals hindsight 142% β€” 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
Seixas PaulDecathlon CMA CGM Team34.0%38.7%22.9%n/a––
Evenepoel RemcoRed Bull-BORA-hansgrohe33.3%33.9%23.4%n/a––
del Toro IsaacUAE Team Emirates-XRG23.3%18.7%17.6%n/a––
Pidcock TomPinarello-Q36.5 Pro Cycling Team10.8%5.4%8.9%n/a––
Ayuso JuanLidl-Trek3.3%2.5%2.5%n/a––
Vermeersch FlorianUAE Team Emirates-XRG2.1%3.2%1.2%n/a––
McNulty BrandonUAE Team Emirates-XRG1.9%2.9%1.1%n/a––
Jorgenson MatteoTeam Visma | Lease a Bike1.9%1.7%1.4%n/a––
Ciccone GiulioLidl-Trek1.9%1.5%1.4%n/a––
Matthews MichaelTeam Jayco-AlUla1.8%3.5%0.9%n/a––
Hirschi MarcTudor Pro Cycling Team1.6%1.5%1.1%n/a––
Lapeira PaulDecathlon CMA CGM Team1.5%2.1%0.9%n/a––
Van Eetvelt LennertLotto-IntermarchΓ©1.5%1.3%1.1%n/a––
Silva ThomasXDS Astana Team1.3%1.6%0.9%n/a––
Van Wilder IlanSoudal Quick-Step1.0%0.9%0.7%n/a––
Hindsight (post-hoc) β€” oracle-matched references β€” top 30, all heads
#RiderBlendCascade LambdaRankPodiumTop 15 Market odds
1Seixas Paul34.0%38.7%22.9%58.5%83.0%n/a
2Evenepoel Remco33.3%33.9%23.4%58.6%83.9%n/a
3del Toro Isaac23.3%18.7%17.6%34.3%68.7%n/a
4Pidcock Tom10.8%5.4%8.9%20.2%64.1%n/a
5Ayuso Juan3.3%2.5%2.5%8.2%35.4%n/a
6Vermeersch Florian2.1%3.2%1.2%7.3%25.2%n/a
7McNulty Brandon1.9%2.9%1.1%6.9%30.0%n/a
8Jorgenson Matteo1.9%1.7%1.4%6.3%32.4%n/a
9Ciccone Giulio1.9%1.5%1.4%6.5%45.4%n/a
10Matthews Michael1.8%3.5%0.9%7.1%36.2%n/a
11Hirschi Marc1.6%1.5%1.1%6.2%43.7%n/a
12Lapeira Paul1.5%2.1%0.9%6.5%33.4%n/a
13Van Eetvelt Lennert1.5%1.3%1.1%5.5%42.1%n/a
14Silva Thomas1.3%1.6%0.9%5.1%40.7%n/a
15Van Wilder Ilan1.0%0.9%0.7%2.3%26.0%n/a
16Baudin Alex0.9%1.5%0.5%5.5%38.6%n/a
17Berckmoes Jenno0.8%0.6%0.7%4.7%35.8%n/a
18Van Gils Maxim0.8%0.8%0.5%2.6%20.5%n/a
19Schachmann Max0.7%0.7%0.5%2.4%27.1%n/a
20Wellens Tim0.7%0.6%0.5%2.1%20.9%n/a
21Benoot Tiesj0.7%0.7%0.5%2.0%12.3%n/a
22Gruel Thibaud0.7%1.6%0.2%2.6%22.4%n/a
23Zambanini Edoardo0.6%0.5%0.5%3.7%26.1%n/a
24Yates Adam0.6%0.7%0.4%2.2%24.7%n/a
25Schmid Mauro0.6%0.4%0.5%1.8%15.7%n/a
26Simmons Quinn0.6%0.5%0.4%3.4%22.9%n/a
27Strong Corbin0.5%0.7%0.3%3.0%19.1%n/a
28Segaert Alec0.5%0.8%0.3%2.5%13.4%n/a
29Lemmen Bart0.5%0.5%0.4%1.6%9.8%n/a
30Arensman Thymen0.5%0.7%0.3%1.8%15.7%n/a

After the race

Result β€” won by del Toro Isaac

PosRiderTeamHindsight (post-hoc) win p
1del Toro IsaacUAE Team Emirates-XRG23.3%
2Seixas PaulDecathlon CMA CGM Team34.0%
3McNulty BrandonUAE Team Emirates-XRG1.9%
4Simmons QuinnLidl-Trek0.6%
5Evenepoel RemcoRed Bull-BORA-hansgrohe33.3%
6Pidcock TomPinarello-Q36.5 Pro Cycling Team10.8%
7Ciccone GiulioLidl-Trek1.9%
8Healy BenEF Education-EasyPost0.3%
9Eulalio AfonsoBahrain Victorious0.2%
10Van Eetvelt LennertLotto-IntermarchΓ©1.5%

Scoring β€” probability given to the actual winner

ReadingBlendCascade LambdaRankMarket devig
Hindsight (post-hoc)23.31%18.71%17.61%–

Result rows are joined to the boards by FirstCycling rider id through the riders table. 0 row(s) needed the normalised-name fallback; 1 finisher(s) could not be matched at all.