How much better than his rating a sprinter performs when this rider delivers him, as last man, in sprint display points. It is not a rating of the rider: the points land on the sprinter. Rider ratings →
Loading…
How to read this board
- What the number is. The sprint display points his sprinter gains on a flat bunch finish with this rider as last man. It is never added to the rider's own rating.
- How credit is given. In each flat bunch finish where a team's best sprinter arrives with the group, the credit follows whether the sprinter beat expectation, and goes to the three fastest teammates with him. “Last man” means the fastest of them by sprint rating, not necessarily whoever rode last, and “credited most with” is a fact about the model, not a claim about anyone's designated lead-out rider.
- A climbing filter. A teammate among the field's leading climbers is left out of the train unless he is also one of its leading sprinters. Climbers are not lead-out riders, and dropping one promotes the real lead-out rider into the train.
- Why a lead-out specialist can sit lower than expected. Only the three fastest teammates are credited, at fixed weights. The model's own fit says those weights undersell the second and third riders, and give nothing to the fourth to sixth when they are worth about a fifth of the last man, so riders whose work ends further from the line rank lower than a fan would put them.
- It decays every day. A value loses part of itself every day without new credit. The last-credited column says how long a rider has been sliding.
- A floor, not a ceiling. The figure shown uses the form-controlled rate, which is deliberately conservative; hover a value for the full effect.
- Career λ is a different measure. It is fitted once a month on every bunch sprint at once: a rider's effect on his teammates' finishing odds, net of the team's strength, the size of the squad and his own ability. It is shrunk toward zero, measured from an average helper rather than a helper with no record, and it does not predict later seasons. It sits beside the career high, never on its axis.
- Men and women are measured separately. Each board uses its own fitted rate and ranks riders within its own peloton, so the two scales do not read across.
- The method and the evidence behind it: Rating the lead-out man.