Cycling race predictor Β· Giro d'Abruzzo

Stage 4 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 122 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
sprint0.00
mountain0.40
hilly0.00
rolling0.59
break0.00
TT0.00
keyed: GC day (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): Rondel Mathys 14.0%; Crescioli Ludovico 9.7%; Fancellu Alessandro 6.4%; Castellon Jan 6.1%; Vauquelin KΓ©vin 5.7%

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) 84% 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
Rondel MathysTudor Pro Cycling Team14.0%n/a––
Crescioli LudovicoTeam Polti VisitMalta9.7%n/a––
Fancellu AlessandroMBH Bank Ballan Telecom Fort6.4%n/a––
Castellon JanCaja Rural-Seguros RGA6.1%n/a––
Vauquelin KΓ©vinNetcompany INEOS5.7%n/a––
Berasategi XabierEuskaltel-Euskadi5.2%n/a––
Godon DorianNetcompany INEOS4.5%n/a––
Tolio AlexBardiani-CSF 7 Saber3.6%n/a––
Verre AlessandroMBH Bank Ballan Telecom Fort3.1%n/a––
Poels WoutUnibet Rose Rockets2.3%n/a––
Masnada FaustoMBH Bank Ballan Telecom Fort2.3%n/a––
Ulissi DiegoXDS Astana Development Team2.2%n/a––
Turconi FilippoBardiani-CSF 7 Saber1.8%n/a––
Ponomar AndriiGeneral Store-Essegibi-F.Ili Curia1.4%n/a––
Kajamini FlorianMBH Bank Ballan Telecom Fort1.1%n/a––
Pozzovivo DomenicoSolution Tech-NIPPO-Rali1.0%n/a––
Hindsight (post-hoc) β€” oracle-matched references β€” top 30, all heads
#RiderBlendCascade LambdaRankPodiumTop 15 Book odds
1Rondel Mathys14.0%11.1%17.5%29.6%85.4%n/a
2Crescioli Ludovico9.7%7.9%12.1%25.6%78.9%n/a
3Fancellu Alessandro6.4%5.3%7.9%23.8%76.0%n/a
4Castellon Jan6.1%6.3%7.2%18.9%67.1%n/a
5Vauquelin KΓ©vin5.7%12.1%5.3%26.8%70.3%n/a
6Berasategi Xabier5.2%1.9%7.1%9.0%44.8%n/a
7Godon Dorian4.5%12.9%3.2%28.2%56.6%n/a
8Tolio Alex3.6%2.1%4.7%9.5%55.7%n/a
9Verre Alessandro3.1%1.3%4.1%4.4%14.0%n/a
10Poels Wout2.3%1.7%2.9%8.2%53.5%n/a
11Masnada Fausto2.3%1.2%3.0%5.3%34.3%n/a
12Ulissi Diego2.2%2.0%2.7%9.5%63.5%n/a
13Turconi Filippo1.8%1.4%2.3%6.0%48.8%n/a
14Ponomar Andrii1.4%1.6%1.6%3.3%24.9%n/a
15Kajamini Florian1.1%0.6%1.4%2.0%6.3%n/a
16Pozzovivo Domenico1.0%0.8%1.3%5.2%56.7%n/a
17Murguialday Jokin1.0%1.5%1.0%4.5%39.8%n/a
18Jasch Lennart0.7%1.7%0.6%4.6%45.5%n/a
19Juaristi Txomin0.6%0.4%0.8%1.4%14.9%n/a
20BarcelΓ³ Fernando0.6%0.4%0.8%2.4%12.6%n/a
21Sutton Louis0.6%0.3%0.7%2.0%26.0%n/a
22Christophersen Cedrik Bakke0.6%1.0%0.5%2.3%46.0%n/a
23Bou Joan0.5%0.2%0.7%1.5%18.8%n/a
24Lastra Jonathan0.5%0.3%0.7%1.2%12.1%n/a
25Tsarenko Kyrylo0.5%0.7%0.5%1.5%13.4%n/a
26Ballabio Giacomo0.4%0.1%0.6%0.5%23.9%n/a
27Conci Nicola0.4%0.8%0.4%4.0%46.3%n/a
28Cantoni Andrea0.4%0.4%0.4%1.9%21.2%n/a
29Garibbo Nicolo0.4%0.7%0.3%3.0%28.5%n/a
30Martinelli Alessio0.3%0.1%0.4%0.9%6.9%n/a

After the stage

Results β€” won by Vauquelin KΓ©vin

PosRiderTeamHindsight (post-hoc) win p
1Vauquelin KΓ©vinNetcompany INEOS5.7%
2Rondel MathysTudor Pro Cycling Team14.0%
3Conci NicolaXDS Astana Development Team0.4%
4Godon DorianNetcompany INEOS4.5%
5Ulissi DiegoXDS Astana Development Team2.2%
6Crescioli LudovicoTeam Polti VisitMalta9.7%
7Garibbo NicoloTeam UKYO0.4%
8Turconi FilippoBardiani-CSF 7 Saber1.8%
9Ballabio GiacomoTeam Vorarlberg0.4%
10Jasch LennartTudor Pro Cycling Team0.7%

Scenario scoring β€” probability given to the actual winner

ScenarioBlend win pBook devig
Hindsight (post-hoc)5.74%–