Models / gradient-boosted-ensemble

Pit Stop Count

P(one / two / three+ total stops) for the race, per driver — re-estimated every lap.

gradient-boosted-ensembleanalytics.strategy

How it works

Race-level strategy shape, conditioned on the race so far. Pre-race context — circuit degradation profile, pit-lane loss, grid slot, race distance, starting compound and rain signal — is combined with as-of-lap state: stops already taken, current tyre age and compound, track status, recent safety-car laps and laps remaining. It is recomputed once per completed lap. In the opening laps it reduces to the pre-race estimate, which is also published on analytics.race-preview.

When to use it

Over/under pit-stop markets and one-stop versus two-stop strategy questions.
Pre-race previews: the race-preview feed carries the grid-stage distribution before lights out.
Strategy screens that show how a driver's plan changes as the race unfolds.

Inputs

circuit degradation profile
pit-lane delta
grid position
race distance
rain probability
starting compound
stops so far
tyre age & compound
safety-car history
laps remaining

Output

drivers[].pit.plan.pByStopsProbabilities keyed "1" / "2" / "3+" for the final total number of stops.

drivers[].pit.plan.expectedStopsExpected final number of stops.

drivers[].pit.plan.mostLikelyStopsThe most likely total; three or more is reported as 3.

pitStopCount (analytics.race-preview)Pre-race pOneStop, pTwoStop, pThreePlusStop and expectedStops, per driver.

Example payload

An excerpt of one analytics.strategy delivery, trimmed to the fields this model produces. Values are illustrative. The full message shape is on the feed reference.

json

{
  "feed": "analytics.strategy",
  "lap": 35,
  "drivers": [
    {
      "number": "4",
      "tla": "NOR",
      "pit": {
        "plan": {
          "mostLikelyStops": 1,
          "expectedStops": 1.32,
          "pByStops": {
            "1": 0.7,
            "2": 0.28,
            "3+": 0.02
          }
        }
      }
    }
  ]
}

Use it

Subscribe to analytics.strategy with a webhook and read this model's fields from each delivery:

bash

curl -X POST https://api.racehooks.io/v1/webhooks \
  -H "Authorization: Bearer $RACEHOOKS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"webhookUrl":"https://your-app.com/f1","feedId":"analytics.strategy"}'

To try it without waiting for a race weekend, replay the recorded Canadian Grand Prix (2026-canada_r) from the quickstart.

How to read it

pByStops counts total stops for the whole race, including stops already made — a driver who has stopped once and shows a high "2" is expected to stop once more. Values move after events that change the strategy picture, such as a safety car or an early stop. Before the race, use the race-preview version, which assumes a medium-compound start.

Limitations

It is updated once per completed lap, not mid-lap.
The pre-race distribution on race-preview assumes a medium-compound start and is optional per driver.
Three or more stops are grouped into one bucket.

FAQ

Does pByStops count stops already made?

Yes. It is the probability of the final total number of stops for the race, conditioned on what has happened so far.

Is there a pre-race pit stop count?

Yes. analytics.race-preview carries pOneStop, pTwoStop, pThreePlusStop and expectedStops per driver, assuming a medium-compound start.

Related models

Pit Stop Prediction — Probability of a green-flag pit stop on the next lap; stops under safety car, VSC or red flag are not counted.
Pit Window Survival — How much longer each driver's current stint is likely to run.
Tire Compound — The next compound a pitting driver takes, as probabilities.

Get the data

This model is delivered on analytics.strategy. Subscribe by webhook or stream it over SSE. The quickstart walks through registering an endpoint and replaying a past session against it.

← All models