Cycling race predictor Β· Tour of Huangshan

Stage 1 β€” 2026-09-18

The same production stack priced under 1 course readings, against the bookmaker's pre-stage line. Each scenario is one opinion about what kind of day this is; the results section settles the argument.

1 scenarios 85 riders priced models retrained d911de60f7 book snapshot no capture

The fork

1 readings of the same road

Course profile is an input the model can't see from a map β€” it's set from reference races (or overridden by hand). Different readings redistribute the same probability mass across rider types.

Hindsight (post-hoc)

Hindsight (post-hoc) β€” oracle-matched references
sprint0.42
mountain0.08
hilly0.36
rolling0.04
break0.10
TT0.00
keyed: bunch sprint (board order by rider-type retrieval; prices unchanged)
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): Cavia Daniel 32.6%; Scott Cameron 7.0%; Munoz Hodei 4.9%; Isasa Xabier 4.1%; Lopez Jordi 4.0%

The board

Win probabilities, scenario by scenario

Blend win probability per scenario (cascade Γ— LambdaRank, DNF-adjusted at the pricing layer), next to the devigged Kambi (Unibet) win line captured the morning before the stage. "Max edge" compares the most favourable scenario against the book β€” treat it as scenario-conditional, not a free bet.

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 (post-hoc) 74% β€” while the devigged book column sums to exactly 100%. Putting the two side by side flatters the model, and Max edge is inflated by about that factor: divide a rider's probability by his reading's total before calling anything an edge. Of our heads only the LambdaRank one is natively book-preserving.

RiderTeamHindsight (post-hoc)Book oddsBook devigMax edge
Cavia DanielBurgos Burpellet BH32.6%n/a––
Scott CameronLi Ning Star7.0%n/a––
Munoz HodeiEuskaltel-Euskadi4.9%n/a––
Isasa XabierEuskaltel-Euskadi4.1%n/a––
Lopez JordiEuskaltel-Euskadi4.0%n/a––
Ugarte GariEuskaltel-Euskadi2.7%n/a––
Rikunov PetrWheeltop Rotor Chengdu Team1.9%n/a––
Quartucci LorenzoBurgos Burpellet BH1.8%n/a––
BoΓΊglas GeorgiosBurgos Burpellet BH1.6%n/a––
Rasch JesperChina Anta - Mentech CT1.2%n/a––
Hindsight (post-hoc) β€” oracle-matched references β€” top 30, all heads
#RiderBlendCascade LambdaRankPodiumTop 15 Book odds
1Cavia Daniel32.6%25.3%45.9%46.4%74.3%n/a
2Scott Cameron7.0%6.6%9.5%14.7%36.2%n/a
3Munoz Hodei4.9%9.7%5.4%22.5%51.5%n/a
4Isasa Xabier4.1%4.5%5.5%15.2%47.5%n/a
5Lopez Jordi4.0%2.1%5.8%9.5%54.6%n/a
6Ugarte Gari2.7%2.2%3.7%8.3%34.9%n/a
7Rikunov Petr1.9%2.1%2.5%9.1%47.8%n/a
8Quartucci Lorenzo1.8%1.5%2.4%6.4%40.6%n/a
9BoΓΊglas Georgios1.6%1.6%2.1%6.0%27.2%n/a
10Rasch Jesper1.2%2.4%1.4%7.7%25.7%n/a
11Da Silva Tiano1.0%1.1%1.3%4.8%30.4%n/a
12Álvarez Rodrigo0.9%1.2%1.2%5.9%39.0%n/a
13Smit Willie0.9%1.5%1.0%4.3%35.3%n/a
14Sanikwathi Thanawut0.8%1.9%0.9%4.6%16.3%n/a
15Hopkins Dylan0.8%1.5%1.0%4.9%28.1%n/a
16Miltiadis Andreas0.7%1.3%0.8%4.0%25.9%n/a
17Stepanov Andrei0.5%0.3%0.7%1.4%23.2%n/a
18Fitsumbrhan Daniyom0.3%0.2%0.5%1.1%16.1%n/a
19Jobber James0.3%0.5%0.4%1.4%27.1%n/a
20Sydor Serhii0.3%0.2%0.5%1.0%11.0%n/a
21Pronskiy Vadim0.3%0.2%0.4%1.7%20.1%n/a
22Vandenbulcke Alex0.3%0.1%0.4%0.6%18.2%n/a
23Verburg Luke0.2%0.3%0.3%3.3%19.0%n/a
24Yang Mingsen0.2%0.2%0.3%0.6%16.4%n/a
25Li Tiancheng0.2%0.1%0.3%2.4%12.5%n/a
26Iderbold Bold0.2%0.4%0.2%1.5%17.2%n/a
27Labib Shotorban Ali0.2%0.2%0.3%1.2%13.3%n/a
28Alustiza Nicolas0.2%0.2%0.3%0.8%13.8%n/a
29Delgado David0.2%0.1%0.3%0.6%11.4%n/a
30Meijers Jeroen0.2%0.2%0.2%0.8%24.3%n/a

After the stage

Results β€” won by Cavia Daniel

PosRiderTeamHindsight (post-hoc) win p
1Cavia DanielBURGOS-BURPELLET-BH32.6%
2Mudgway LukeLI NING STAR0.2%
3Rasch JesperCHINA ANTA - MENTECH CYCLING TEAM1.2%
4Verburg LukeAZERION / VILLA VALKENBURG0.2%
5Scott CameronLI NING STAR7.0%
6Meijers JeroenVICTORIA SPORTS PRO CYCLING0.2%
7Álvarez RodrigoBURGOS-BURPELLET-BH0.9%
8Ugarte GariEUSKALTEL-EUSKADI2.7%
9Rikunov PetrWHEELTOP ROTOR CHENGDU TEAM1.9%
10Da Silva TianoTHE JOYRUN & HURRICANE CYCLING TEAM1.0%

3 riders in the top 15 are shown under a provisional identity pending reconciliation: SMIT Willem Jakobus, SMIT Willem Jakobus, PAWSON Edward.

Scenario scoring β€” probability given to the actual winner

ScenarioBlend win pBook devig
Hindsight (post-hoc)32.62%–