Cycling race predictor · CRO Race

Stage 6 — 2026-09-27

The same production stack priced under 2 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.

2 scenarios 100 riders priced models retrained 75c6e7af51 book snapshot no capture

← System overview, season boards & the full race views

The fork

2 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.

Main read

The main read — the day as the key says it goes

sprint0.48

mountain0.04

hilly0.39

rolling0.02

break0.07

TT0.00

keyed: bunch sprint 0.6 / survivor gallop 0.2 / split sprint 0.1 (written by the automated reader, not by a person) (board order by rider-type retrieval; prices unchanged)

The main board. It inherits the config-root day key and is priced on the reference set that key selects.

Top 5 (blend): Magnier Paul 57.8%; Teutenberg Tim Torn 4.9%; Kogut Oded 4.1%; Blikra Erlend 3.6%; Donati Davide 1.8%

Alt: res_3

The other branch — a reduced bunch sprint

sprint0.01

mountain0.12

hilly0.02

rolling0.45

break0.41

TT0.00

keyed: survivor gallop (written by the automated reader, not by a person) (board order by rider-type retrieval; prices unchanged)

The alternate branch, chosen by the driver because the day key carries weight in a second family (a sprint off a hard finale). It is priced as if that shape were certain, so the two boards bracket the read.

Top 5 (blend): Magnier Paul 33.0%; Tiberi Antonio 7.6%; Teutenberg Tim Torn 7.5%; Van Wilder Ilan 6.4%; Boichis Adrien 6.4%

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 Main read 87%, Alt: res_3 100% — 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	Main read	Alt: res_3	Book odds	Book devig	Max edge
Magnier Paul	Soudal Quick-Step	57.8%	33.0%	n/a	–	–
Tiberi Antonio	Bahrain Victorious	0.3%	7.6%	n/a	–	–
Teutenberg Tim Torn	Lidl-Trek	4.9%	7.5%	n/a	–	–
Van Wilder Ilan	Soudal Quick-Step	0.1%	6.4%	n/a	–	–
Boichis Adrien	Red Bull-BORA-hansgrohe	1.4%	6.4%	n/a	–	–
Zambanini Edoardo	Bahrain Victorious	1.4%	6.1%	n/a	–	–
Kogut Oded	NSN Cycling Team	4.1%	2.2%	n/a	–	–
Blikra Erlend	Uno-X Mobility	3.6%	0.7%	n/a	–	–
Wellens Tim	UAE Team Emirates-XRG	0.1%	3.4%	n/a	–	–
Politt Nils	UAE Team Emirates-XRG	0.1%	3.3%	n/a	–	–
Kulset Johannes	Uno-X Mobility	0.1%	2.0%	n/a	–	–
Donati Davide	Red Bull-BORA-hansgrohe	1.8%	0.4%	n/a	–	–
Martinez Daniel	Red Bull-BORA-hansgrohe	0.0%	1.7%	n/a	–	–
Teuns Dylan	Cofidis	0.1%	1.6%	n/a	–	–
Belletta Dario	Team Polti VisitMalta	0.7%	1.6%	n/a	–	–
Zanoncello Enrico	Bardiani-CSF 7 Saber	1.5%	0.6%	n/a	–	–
Oliveira Rui	UAE Team Emirates-XRG	1.3%	1.3%	n/a	–	–
Bevort Carl-Frederik	Uno-X Mobility	0.9%	1.2%	n/a	–	–

The main read — the day as the key says it goes — top 30, all heads

#	Rider	Blend	Cascade	LambdaRank	Podium	Top 15	Book odds
1	Magnier Paul	57.8%	50.2%	68.1%	68.9%	80.8%	n/a
2	Teutenberg Tim Torn	4.9%	5.8%	5.3%	20.0%	66.1%	n/a
3	Kogut Oded	4.1%	4.8%	4.5%	16.8%	46.2%	n/a
4	Blikra Erlend	3.6%	6.0%	3.6%	17.8%	41.6%	n/a
5	Donati Davide	1.8%	3.1%	1.7%	9.4%	42.4%	n/a
6	Zanoncello Enrico	1.5%	1.0%	1.9%	7.8%	39.5%	n/a
7	Zambanini Edoardo	1.4%	0.9%	1.7%	4.8%	41.1%	n/a
8	Boichis Adrien	1.4%	1.6%	1.5%	5.4%	38.3%	n/a
9	Oliveira Rui	1.3%	1.4%	1.5%	8.1%	38.9%	n/a
10	Darder Sergi	1.0%	1.1%	1.1%	7.3%	58.3%	n/a
11	Bevort Carl-Frederik	0.9%	1.4%	0.9%	5.5%	43.1%	n/a
12	Erzen Zak	0.8%	0.9%	0.8%	4.4%	39.4%	n/a
13	Viviani Attilio	0.7%	0.6%	0.8%	3.5%	23.9%	n/a
14	Belletta Dario	0.7%	0.5%	0.8%	2.3%	42.7%	n/a
15	Buratti Nicolò	0.4%	0.9%	0.4%	3.6%	32.7%	n/a
16	Theuns Edward	0.3%	0.2%	0.4%	2.5%	23.2%	n/a
17	Gazzoli Michele	0.3%	0.7%	0.3%	3.9%	26.3%	n/a
18	Skerl Daniel	0.3%	0.7%	0.2%	3.3%	19.0%	n/a
19	McDonald Brody	0.3%	0.8%	0.2%	3.3%	37.3%	n/a
20	Martín Gotzon	0.3%	0.1%	0.3%	1.4%	26.7%	n/a
21	Tiberi Antonio	0.3%	0.1%	0.3%	1.5%	17.8%	n/a
22	Pedersen Henrik	0.2%	0.1%	0.3%	1.2%	23.7%	n/a
23	Maestri Mirco	0.2%	0.1%	0.2%	0.6%	14.2%	n/a
24	De Longhi Lorenzo	0.2%	0.2%	0.2%	1.4%	14.0%	n/a
25	Pesenti Thomas	0.1%	0.1%	0.2%	1.2%	20.3%	n/a
26	Barceló Fernando	0.1%	0.1%	0.1%	0.9%	13.9%	n/a
27	Izquierdo Clement	0.1%	0.3%	0.1%	1.3%	16.9%	n/a
28	Kulset Johannes	0.1%	0.1%	0.1%	0.6%	12.8%	n/a
29	van Gestel Dries	0.1%	0.1%	0.1%	0.6%	7.7%	n/a
30	Politt Nils	0.1%	0.1%	0.1%	0.4%	7.4%	n/a

The other branch — a reduced bunch sprint — top 30, all heads

#	Rider	Blend	Cascade	LambdaRank	Podium	Top 15	Book odds
1	Magnier Paul	33.0%	41.0%	30.9%	56.8%	69.7%	n/a
2	Tiberi Antonio	7.6%	6.7%	7.8%	25.6%	76.0%	n/a
3	Teutenberg Tim Torn	7.5%	7.6%	7.5%	20.6%	57.6%	n/a
4	Van Wilder Ilan	6.4%	4.8%	6.8%	13.8%	48.1%	n/a
5	Boichis Adrien	6.4%	6.3%	6.4%	21.1%	64.2%	n/a
6	Zambanini Edoardo	6.1%	2.6%	6.9%	11.0%	57.3%	n/a
7	Wellens Tim	3.4%	4.7%	3.1%	15.3%	68.5%	n/a
8	Politt Nils	3.3%	2.2%	3.6%	7.7%	36.4%	n/a
9	Kogut Oded	2.2%	1.6%	2.3%	5.9%	17.7%	n/a
10	Kulset Johannes	2.0%	1.9%	2.0%	6.6%	51.2%	n/a
11	Martinez Daniel	1.7%	2.2%	1.5%	9.5%	40.8%	n/a
12	Teuns Dylan	1.6%	1.5%	1.7%	5.1%	51.8%	n/a
13	Belletta Dario	1.6%	0.6%	1.8%	2.6%	31.9%	n/a
14	Oliveira Rui	1.3%	1.6%	1.2%	8.2%	32.5%	n/a
15	Bevort Carl-Frederik	1.2%	1.6%	1.1%	5.3%	33.6%	n/a
16	Blikra Erlend	0.7%	0.8%	0.7%	4.2%	14.4%	n/a
17	Theuns Edward	0.7%	0.7%	0.7%	4.4%	28.3%	n/a
18	Izquierdo Clement	0.7%	0.9%	0.6%	3.2%	26.4%	n/a
19	Zanoncello Enrico	0.6%	0.6%	0.7%	4.4%	22.5%	n/a
20	Pedersen Henrik	0.6%	1.0%	0.5%	4.0%	34.4%	n/a
21	Balmer Alexandre	0.6%	0.8%	0.6%	2.7%	17.4%	n/a
22	Tolio Alex	0.5%	0.2%	0.6%	1.3%	31.1%	n/a
23	Pesenti Thomas	0.5%	0.2%	0.6%	2.0%	24.1%	n/a
24	Martín Gotzon	0.5%	0.2%	0.5%	1.6%	21.0%	n/a
25	Cras Steff	0.5%	0.3%	0.5%	2.2%	26.8%	n/a
26	Donati Davide	0.4%	1.3%	0.2%	4.1%	18.1%	n/a
27	Darder Sergi	0.4%	0.4%	0.4%	3.0%	23.6%	n/a
28	Nicolau Joel	0.4%	0.2%	0.4%	1.3%	32.6%	n/a
29	Barceló Fernando	0.4%	0.2%	0.4%	2.1%	11.3%	n/a
30	Agirre Jon	0.4%	0.2%	0.4%	1.0%	24.5%	n/a

After the stage

Results — won by Donati Davide

Pos	Rider	Team	Main read win p	Alt: res_3 win p
1	Donati Davide	Red Bull-BORA-hansgrohe	1.8%	0.4%
2	De Longhi Lorenzo	Cofidis	0.2%	0.0%
3	Magnier Paul	Soudal Quick-Step	57.8%	33.0%
4	Teutenberg Tim Torn	Lidl-Trek	4.9%	7.5%
5	Blikra Erlend	Uno-X Mobility	3.6%	0.7%
6	Bevort Carl-Frederik	Uno-X Mobility	0.9%	1.2%
7	Zanoncello Enrico	Bardiani-CSF 7 Saber	1.5%	0.6%
8	McDonald Brody	Modern Adventure Pro Cycling	0.3%	0.0%
9	Petric Bastian	Pogi Team Gusto Ljubljana	0.0%	0.0%
10	Darder Sergi	Caja Rural-Seguros RGA	1.0%	0.4%

Scenario scoring — probability given to the actual winner

Scenario	Blend win p	Book devig
Main read	1.79%	–
Alt: res_3	0.43%	–