Cycling race predictor · Tour of Huangshan

Stage 2 — 2026-09-19

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 95 riders priced models retrained 75c6e7af51 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 — 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) — oracle-matched references
sprint0.11
mountain0.05
hilly0.81
rolling0.02
break0.01
TT0.00
keyed: bunch sprint (board order by rider-type retrieval; prices unchanged)
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): Scott Cameron 20.1%; Munoz Hodei 15.1%; Cavia Daniel 13.7%; Rasch Jesper 4.6%; Ugarte Gari 4.4%

The board

Win probabilities, scenario by scenario

Blend win probability per scenario (cascade × 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 — 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 — across this field each reading totals Hindsight (post-hoc) 86% — 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
Scott CameronLi Ning Star20.1%n/a––
Munoz HodeiEuskaltel-Euskadi15.1%n/a––
Cavia DanielBurgos Burpellet BH13.7%n/a––
Rasch JesperChina Anta - Mentech CT4.6%n/a––
Ugarte GariEuskaltel-Euskadi4.4%n/a––
Mudgway LukeLi Ning Star3.6%n/a––
Rikunov PetrWheeltop Rotor Chengdu Team2.5%n/a––
Lopez JordiEuskaltel-Euskadi2.5%n/a––
Smit WillieChina Anta - Mentech CT2.1%n/a––
Álvarez RodrigoBurgos Burpellet BH1.3%n/a––
Verburg LukeAzerion / Villa Valkenburg1.3%n/a––
Isasa XabierEuskaltel-Euskadi1.2%n/a––
Meijers JeroenVictoria Sports Pro Cycling1.1%n/a––
Da Silva TianoThe JoyRun & Hurricane Cycling Team1.1%n/a––
Boúglas GeorgiosBurgos Burpellet BH1.0%n/a––
Hindsight (post-hoc) — oracle-matched references — top 30, all heads
#RiderBlendCascade LambdaRankPodiumTop 15 Book odds
1Scott Cameron20.1%21.5%23.0%37.3%59.0%n/a
2Munoz Hodei15.1%19.8%16.4%38.4%67.2%n/a
3Cavia Daniel13.7%9.5%17.0%35.2%66.7%n/a
4Rasch Jesper4.6%7.4%4.6%25.5%48.4%n/a
5Ugarte Gari4.4%2.4%5.6%12.2%52.8%n/a
6Mudgway Luke3.6%3.1%4.4%13.5%56.5%n/a
7Rikunov Petr2.5%2.5%3.0%10.7%57.9%n/a
8Lopez Jordi2.5%1.6%3.1%6.2%41.3%n/a
9Smit Willie2.1%2.1%2.4%7.6%46.8%n/a
10Álvarez Rodrigo1.3%1.4%1.5%7.6%48.1%n/a
11Verburg Luke1.3%1.5%1.5%5.7%38.6%n/a
12Isasa Xabier1.2%2.2%1.2%6.9%33.9%n/a
13Meijers Jeroen1.1%0.8%1.3%5.7%52.3%n/a
14Da Silva Tiano1.1%0.9%1.3%5.1%43.4%n/a
15Boúglas Georgios1.0%0.3%1.3%2.0%10.6%n/a
16Quartucci Lorenzo0.9%0.8%1.1%4.9%34.3%n/a
17Stepanov Andrei0.8%0.5%1.0%2.0%27.5%n/a
18Yang Mingsen0.5%0.1%0.6%2.4%38.6%n/a
19Sydor Serhii0.4%0.1%0.6%0.9%8.1%n/a
20Hopkins Dylan0.4%0.4%0.5%1.7%16.4%n/a
21Tamm Lauri0.3%0.3%0.4%1.3%14.1%n/a
22Forbes James0.3%0.2%0.4%1.6%12.3%n/a
23Pawson Edward0.3%0.3%0.4%3.0%20.0%n/a
24Jeong Seunghwa0.3%0.2%0.4%0.9%17.5%n/a
25Pronskiy Vadim0.3%0.2%0.4%1.4%25.9%n/a
26Fitsumbrhan Daniyom0.3%0.3%0.4%1.1%9.3%n/a
27Iderbold Bold0.3%0.3%0.3%1.4%21.8%n/a
28Sepulveda Eduardo0.2%0.1%0.3%1.0%11.8%n/a
29Raileanu Cristian0.2%0.2%0.2%0.8%17.0%n/a
30Desriac Loïc0.2%0.1%0.2%0.6%16.2%n/a

After the stage

Results — won by Tamm Lauri

PosRiderTeamHindsight (post-hoc) win p
1Tamm LauriCHINA ANTA - MENTECH CYCLING TEAM0.3%
2Rikunov PetrWHEELTOP ROTOR CHENGDU TEAM2.5%
3Munoz HodeiEUSKALTEL-EUSKADI15.1%
4Álvarez RodrigoBURGOS-BURPELLET-BH1.3%
5Cavia DanielBURGOS-BURPELLET-BH13.7%
6Meijers JeroenVICTORIA SPORTS PRO CYCLING1.1%
7Isasa XabierEUSKALTEL-EUSKADI1.2%
8Mudgway LukeLI NING STAR3.6%
9Ugarte GariEUSKALTEL-EUSKADI4.4%
10Küüt OskarQUICK PRO TEAM0.2%

Scenario scoring — probability given to the actual winner

ScenarioBlend win pBook devig
Hindsight (post-hoc)0.35%–