Cycling race predictor Β· Giro d'Abruzzo

Stage 2 β€” 2026-09-16

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 127 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
mountain0.29
hilly0.00
rolling0.69
break0.02
TT0.00
keyed: split sprint (res_5) (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 22.9%; Godon Dorian 14.0%; Crescioli Ludovico 13.6%; Vauquelin KΓ©vin 10.2%; Fancellu Alessandro 5.4%

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) 105% 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 Team22.9%n/a––
Godon DorianNetcompany INEOS14.0%n/a––
Crescioli LudovicoTeam Polti VisitMalta13.6%n/a––
Vauquelin KΓ©vinNetcompany INEOS10.2%n/a––
Fancellu AlessandroMBH Bank Ballan Telecom Fort5.4%n/a––
Castellon JanCaja Rural-Seguros RGA4.7%n/a––
Berasategi XabierEuskaltel-Euskadi3.0%n/a––
Ulissi DiegoXDS Astana Development Team2.5%n/a––
Poels WoutUnibet Rose Rockets2.3%n/a––
Juaristi TxominEuskaltel-Euskadi1.8%n/a––
Verre AlessandroMBH Bank Ballan Telecom Fort1.7%n/a––
Turconi FilippoBardiani-CSF 7 Saber1.6%n/a––
Pozzovivo DomenicoSolution Tech-NIPPO-Rali1.5%n/a––
Tolio AlexBardiani-CSF 7 Saber1.5%n/a––
Conci NicolaXDS Astana Development Team1.5%n/a––
Kajamini FlorianMBH Bank Ballan Telecom Fort1.1%n/a––
Hindsight (post-hoc) β€” oracle-matched references β€” top 30, all heads
#RiderBlendCascade LambdaRankPodiumTop 15 Book odds
1Rondel Mathys22.9%15.4%23.6%33.2%86.5%n/a
2Godon Dorian14.0%26.1%10.3%41.4%62.2%n/a
3Crescioli Ludovico13.6%9.4%13.9%28.2%80.0%n/a
4Vauquelin KΓ©vin10.2%19.7%7.3%37.3%79.5%n/a
5Fancellu Alessandro5.4%2.4%5.8%16.6%69.4%n/a
6Castellon Jan4.7%3.6%4.7%11.6%62.4%n/a
7Berasategi Xabier3.0%1.4%3.2%7.3%40.2%n/a
8Ulissi Diego2.5%2.1%2.5%14.8%63.4%n/a
9Poels Wout2.3%1.6%2.3%7.2%52.6%n/a
10Juaristi Txomin1.8%1.6%1.8%4.8%29.2%n/a
11Verre Alessandro1.7%0.4%1.9%1.1%6.6%n/a
12Turconi Filippo1.6%1.1%1.6%5.2%46.7%n/a
13Pozzovivo Domenico1.5%1.2%1.5%6.9%62.9%n/a
14Tolio Alex1.5%1.0%1.5%4.0%46.4%n/a
15Conci Nicola1.5%2.1%1.2%9.7%56.0%n/a
16Kajamini Florian1.1%0.6%1.2%1.8%8.9%n/a
17Jasch Lennart0.9%1.6%0.6%4.0%40.0%n/a
18Eiking Odd Christian0.8%0.5%0.8%2.0%13.0%n/a
19Masnada Fausto0.8%0.5%0.8%1.5%17.5%n/a
20Christophersen Cedrik Bakke0.7%1.1%0.5%4.7%49.2%n/a
21Ballabio Giacomo0.6%0.6%0.6%3.5%45.4%n/a
22BarcelΓ³ Fernando0.6%0.5%0.5%2.6%23.2%n/a
23Murguialday Jokin0.6%0.2%0.6%0.8%21.2%n/a
24Tsarenko Kyrylo0.5%0.4%0.6%1.6%14.5%n/a
25Dati Tommaso0.5%0.6%0.5%2.9%15.4%n/a
26Garibbo Nicolo0.5%0.6%0.4%4.3%42.0%n/a
27Bais Mattia0.5%0.7%0.4%4.5%38.6%n/a
28Sutton Louis0.4%0.3%0.4%1.7%24.0%n/a
29Peter Jannis0.4%0.4%0.4%1.5%23.4%n/a
30Martinelli Alessio0.4%0.2%0.4%0.7%12.9%n/a

After the stage

Results β€” won by Jasch Lennart

PosRiderTeamHindsight (post-hoc) win p
1Jasch LennartTudor Pro Cycling Team0.9%
2Godon DorianNetcompany INEOS14.0%
3Scalco MatteoXDS Astana Development Team0.2%
4Dati TommasoTeam UKYO0.5%
5Garibbo NicoloTeam UKYO0.5%
6Crescioli LudovicoTeam Polti VisitMalta13.6%
7Lorello RiccardoSC Padovani Polo Cherry Bank0.0%
8Ballabio GiacomoTeam Vorarlberg0.6%
9Barbuto RiccardoTeam Technipes #inEmiliaRomagna CaffΓ¨ Borbone–
10Berasategi XabierEuskaltel-Euskadi3.0%

Scenario scoring β€” probability given to the actual winner

ScenarioBlend win pBook devig
Hindsight (post-hoc)0.88%–