Cycling race predictor · Vuelta a España 2026

Stage 17 — 2026-09-09

The same production stack priced under 3 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.

3 scenarios 149 riders priced models retrained d911de60f7 book snapshot markets_1734.json

← System overview, season boards & the full race views

The fork

3 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.

Pure speed

The final gallop — the sprinters' last clean road

sprint1.00

mountain0.00

hilly0.00

rolling0.00

break0.00

TT0.00

keyed: bunch sprint (board order by rider-type retrieval; prices unchanged)

The base case and official key. This reuses the campaign's pure-speed anchor set because stage 16 validated it one day ago on essentially the same question: Brennan ranked first and won from a 102-rider group. The terrain is even cleaner here — no meaningful climb, a long flat approach, and several sprint teams with no later opportunity.

Top 5 (blend): Brennan Matthew 52.2%; Meeus Jordi 9.0%; van Aert Wout 6.1%; De Schuyteneer Steffen 2.1%; Coquard Bryan 2.0%

Wind-strung sprint

Wind strings the bunch — still a sprint, but from a thinner front

sprint0.52

mountain0.01

hilly0.46

rolling0.01

break0.00

TT0.00

keyed: bunch sprint 0.7 / survivor gallop 0.2 / split sprint 0.1 (board order by rider-type retrieval; prices unchanged)

The weather stress test. Strong westerly wind is forecast and the final corner at 1.2 km can stretch a tired field, so this board asks whether the same favourites reorder if only a reduced front reaches the sprint. It remains a sprint branch: there is no climb or late launchpad to justify an escape-family key.

Top 5 (blend): Brennan Matthew 50.1%; van Aert Wout 13.0%; Meeus Jordi 10.9%; Coquard Bryan 2.6%; De Schuyteneer Steffen 2.1%

Shortened drag race

Heat-shortened drag race — fresher legs, maximum speed

sprint0.90

mountain0.00

hilly0.00

rolling0.10

break0.00

TT0.00

keyed: bunch sprint (board order by rider-type retrieval; prices unchanged)

A route-change sensitivity, not a claim that shortening has occurred. Forecast heat has prompted discussion of cutting the stage toward roughly 100 km. If that happens, the family is still bunch but accumulated fatigue falls and the board should tilt as far toward raw speed as the model permits.

Top 5 (blend): Brennan Matthew 52.8%; Meeus Jordi 7.7%; van Aert Wout 6.0%; Coquard Bryan 2.2%; De Schuyteneer Steffen 2.1%

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 Pure speed 84%, Wind-strung sprint 95%, Shortened drag race 83% — 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.

Rider	Team	Pure speed	Wind-strung sprint	Shortened drag race	Book odds	Book devig	Max edge
Brennan Matthew	Team Visma \| Lease a Bike	52.2%	50.1%	52.8%	1.50	45.0%	+7.9pp
van Aert Wout	Team Visma \| Lease a Bike	6.1%	13.0%	6.0%	12.00	5.6%	+7.4pp
Meeus Jordi	Red Bull-BORA-hansgrohe	9.0%	10.9%	7.7%	7.00	9.6%	+1.3pp
Coquard Bryan	Cofidis	2.0%	2.6%	2.2%	8.00	8.4%	-5.8pp
De Schuyteneer Steffen	Lotto-Intermarché	2.1%	2.1%	2.1%	81.00	0.8%	+1.3pp
Dainese Alberto	Soudal Quick-Step	1.2%	0.6%	0.9%	34.00	2.0%	-0.7pp
Hofstetter Hugo	NSN Cycling Team	0.9%	1.2%	0.9%	81.00	0.8%	+0.3pp
Romele Alessandro	XDS Astana Team	0.7%	1.1%	0.7%	41.00	1.6%	-0.5pp
Cort Magnus	Uno-X Mobility	0.5%	1.1%	0.5%	19.00	3.5%	-2.5pp

The final gallop — the sprinters' last clean road — top 30, all heads

#	Rider	Blend	Cascade	LambdaRank	Podium	Top 15	Book odds
1	Brennan Matthew	52.2%	40.1%	65.3%	65.4%	85.8%	1.50
2	Meeus Jordi	9.0%	9.6%	10.6%	36.3%	79.3%	7.00
3	van Aert Wout	6.1%	5.7%	7.4%	27.4%	66.6%	12.00
4	De Schuyteneer Steffen	2.1%	1.9%	2.6%	8.1%	62.9%	81.00
5	Coquard Bryan	2.0%	3.3%	2.0%	9.2%	81.8%	8.00
6	Dainese Alberto	1.2%	1.4%	1.5%	8.6%	50.3%	34.00
7	Hofstetter Hugo	0.9%	1.2%	1.1%	8.3%	79.9%	81.00
8	Brustenga Marc	0.9%	0.9%	1.0%	6.1%	48.2%	151.00
9	Aular Orluis	0.8%	0.6%	1.1%	5.7%	67.7%	81.00
10	Nys Thibau	0.8%	2.8%	0.5%	9.0%	35.1%	81.00
11	Laurance Axel	0.8%	2.4%	0.5%	7.2%	48.1%	101.00
12	Romele Alessandro	0.7%	1.6%	0.6%	7.9%	60.5%	41.00
13	Tronchon Bastien	0.6%	1.5%	0.5%	6.9%	61.5%	67.00
14	Braet Vito	0.6%	1.0%	0.6%	8.1%	62.2%	41.00
15	Govekar Matevz	0.6%	1.3%	0.5%	7.3%	50.3%	151.00
16	Miquel Delgado Pau	0.6%	1.3%	0.5%	4.6%	50.8%	151.00
17	Macias Cesar	0.5%	0.9%	0.5%	7.2%	65.9%	201.00
18	Cort Magnus	0.5%	1.4%	0.3%	7.5%	55.3%	19.00
19	Renard Alexis	0.3%	1.0%	0.2%	4.4%	46.2%	151.00
20	Albanese Vincenzo	0.3%	0.2%	0.4%	3.2%	53.5%	126.00
21	Canal Carlos	0.2%	0.2%	0.3%	2.4%	20.0%	401.00
22	Oliveira Ivo	0.2%	0.2%	0.2%	2.5%	25.7%	151.00
23	Camprubí Marcel	0.1%	0.1%	0.1%	2.3%	38.8%	301.00
24	Mulubrhan Henok	0.1%	0.1%	0.1%	1.2%	17.5%	251.00
25	Hayter Ethan	0.1%	0.3%	0.0%	1.5%	15.7%	301.00
26	Debruyne Ramses	0.1%	0.1%	0.1%	1.0%	11.8%	201.00
27	Garofoli Gianmarco	0.1%	0.1%	0.1%	0.9%	9.5%	501.00
28	Brenner Marco	0.1%	0.2%	0.1%	1.4%	9.0%	501.00
29	Vader Milan	0.1%	0.1%	0.1%	0.7%	18.8%	601.00
30	Fedorov Yevgeniy	0.1%	0.2%	0.0%	0.6%	21.0%	301.00

Wind strings the bunch — still a sprint, but from a thinner front — top 30, all heads

#	Rider	Blend	Cascade	LambdaRank	Podium	Top 15	Book odds
1	Brennan Matthew	50.1%	37.4%	55.9%	63.1%	84.2%	1.50
2	van Aert Wout	13.0%	10.4%	14.4%	42.9%	72.6%	12.00
3	Meeus Jordi	10.9%	11.1%	11.5%	36.6%	75.2%	7.00
4	Coquard Bryan	2.6%	4.8%	2.2%	12.3%	77.7%	8.00
5	De Schuyteneer Steffen	2.1%	2.1%	2.3%	7.8%	60.7%	81.00
6	Hofstetter Hugo	1.2%	1.6%	1.1%	8.1%	71.6%	81.00
7	Romele Alessandro	1.1%	2.9%	0.7%	11.1%	56.9%	41.00
8	Cort Magnus	1.1%	2.2%	0.8%	15.9%	57.4%	19.00
9	Laurance Axel	1.0%	2.3%	0.7%	8.7%	49.2%	101.00
10	Aular Orluis	1.0%	0.6%	1.1%	6.8%	61.5%	81.00
11	Albanese Vincenzo	1.0%	1.5%	0.9%	8.9%	62.2%	126.00
12	Nys Thibau	0.9%	2.5%	0.6%	8.8%	47.3%	81.00
13	Govekar Matevz	0.9%	1.8%	0.7%	8.7%	44.8%	151.00
14	Braet Vito	0.8%	1.4%	0.7%	9.9%	59.0%	41.00
15	Tronchon Bastien	0.8%	1.9%	0.5%	7.6%	54.5%	67.00
16	Dainese Alberto	0.6%	1.3%	0.5%	6.8%	46.8%	34.00
17	Renard Alexis	0.6%	1.5%	0.4%	4.9%	48.3%	151.00
18	Miquel Delgado Pau	0.5%	0.9%	0.5%	4.1%	41.3%	151.00
19	Canal Carlos	0.4%	0.3%	0.5%	3.4%	21.9%	401.00
20	Debruyne Ramses	0.4%	0.2%	0.4%	4.8%	21.1%	201.00
21	Hayter Ethan	0.3%	1.1%	0.1%	2.3%	12.2%	301.00
22	Brustenga Marc	0.2%	0.2%	0.2%	4.8%	51.3%	151.00
23	Mulubrhan Henok	0.2%	0.2%	0.2%	1.8%	20.6%	251.00
24	Fisher-Black Finn	0.2%	0.3%	0.2%	2.6%	13.9%	251.00
25	Camprubí Marcel	0.2%	0.2%	0.2%	2.9%	29.2%	301.00
26	Brenner Marco	0.2%	0.3%	0.1%	2.6%	15.2%	501.00
27	Macias Cesar	0.2%	0.1%	0.2%	2.4%	52.3%	201.00
28	Oliveira Ivo	0.1%	0.2%	0.1%	2.7%	21.3%	151.00
29	Cobo Ivan	0.1%	0.2%	0.1%	1.6%	12.6%	501.00
30	Garofoli Gianmarco	0.1%	0.1%	0.1%	1.1%	16.2%	501.00

Heat-shortened drag race — fresher legs, maximum speed — top 30, all heads

#	Rider	Blend	Cascade	LambdaRank	Podium	Top 15	Book odds
1	Brennan Matthew	52.8%	38.9%	67.0%	65.1%	86.1%	1.50
2	Meeus Jordi	7.7%	9.4%	8.8%	36.3%	79.1%	7.00
3	van Aert Wout	6.0%	5.6%	7.3%	27.3%	65.7%	12.00
4	Coquard Bryan	2.2%	3.5%	2.3%	9.9%	78.9%	8.00
5	De Schuyteneer Steffen	2.1%	1.8%	2.6%	7.9%	64.9%	81.00
6	Dainese Alberto	0.9%	1.4%	1.0%	8.7%	50.1%	34.00
7	Hofstetter Hugo	0.9%	1.3%	1.0%	8.6%	77.6%	81.00
8	Laurance Axel	0.9%	2.6%	0.6%	8.1%	48.4%	101.00
9	Aular Orluis	0.9%	0.6%	1.1%	6.0%	64.7%	81.00
10	Nys Thibau	0.8%	2.4%	0.5%	8.6%	48.3%	81.00
11	Romele Alessandro	0.7%	1.8%	0.5%	7.3%	57.9%	41.00
12	Govekar Matevz	0.7%	1.6%	0.6%	8.5%	52.4%	151.00
13	Miquel Delgado Pau	0.6%	1.5%	0.6%	5.1%	52.7%	151.00
14	Tronchon Bastien	0.6%	1.5%	0.5%	6.9%	60.3%	67.00
15	Braet Vito	0.6%	1.0%	0.6%	8.3%	61.2%	41.00
16	Cort Magnus	0.5%	1.4%	0.4%	8.2%	54.7%	19.00
17	Macias Cesar	0.5%	0.9%	0.5%	6.8%	66.4%	201.00
18	Brustenga Marc	0.4%	0.2%	0.6%	4.9%	73.0%	151.00
19	Albanese Vincenzo	0.4%	0.3%	0.5%	3.4%	50.7%	126.00
20	Renard Alexis	0.3%	1.1%	0.2%	4.3%	46.5%	151.00
21	Canal Carlos	0.2%	0.2%	0.3%	2.4%	17.3%	401.00
22	Oliveira Ivo	0.2%	0.3%	0.2%	2.7%	23.0%	151.00
23	Camprubí Marcel	0.1%	0.2%	0.2%	2.3%	36.0%	301.00
24	Mulubrhan Henok	0.1%	0.1%	0.1%	1.4%	16.6%	251.00
25	Debruyne Ramses	0.1%	0.1%	0.1%	1.4%	13.8%	201.00
26	Brenner Marco	0.1%	0.2%	0.1%	1.6%	9.5%	501.00
27	Hayter Ethan	0.1%	0.3%	0.1%	1.5%	15.6%	301.00
28	Garofoli Gianmarco	0.1%	0.1%	0.1%	0.9%	8.8%	501.00
29	Küng Stefan	0.1%	0.2%	0.1%	0.6%	7.9%	301.00
30	Vader Milan	0.1%	0.1%	0.1%	0.7%	17.7%	601.00

After the stage

Results — won by Brennan Matthew

Pos	Rider	Team	Pure speed win p	Wind-strung sprint win p	Shortened drag race win p
1	Brennan Matthew	Team Visma \| Lease a Bike	52.2%	50.1%	52.8%
2	Meeus Jordi	Red Bull-BORA-hansgrohe	9.0%	10.9%	7.7%
3	Cort Magnus	Uno-X Mobility	0.5%	1.1%	0.5%
4	Romele Alessandro	XDS Astana Team	0.7%	1.1%	0.7%
5	De Schuyteneer Steffen	Lotto-Intermarché	2.1%	2.1%	2.1%
6	Laurance Axel	Netcompany INEOS Cycling Team	0.8%	1.0%	0.9%
7	Miquel Delgado Pau	Bahrain Victorious	0.6%	0.5%	0.6%
8	Macias Cesar	Burgos Burpellet BH	0.5%	0.2%	0.5%
9	Hofstetter Hugo	NSN Cycling Team	0.9%	1.2%	0.9%
10	Coquard Bryan	Cofidis	2.0%	2.6%	2.2%

Scenario scoring — probability given to the actual winner

Scenario	Blend win p	Book devig
Pure speed	52.16%	44.96%
Wind-strung sprint	50.11%	44.96%
Shortened drag race	52.82%	44.96%