Models / calibrated-classifier

Pace Mode

Which pace mode a driver is running — push, hold, manage, warm-up or delta — per lap.

calibrated-classifieranalytics.strategy

How it works

Each lap's pace, with the tyre's degradation trend taken out, is assigned to one of five named modes. The boundaries between modes are fitted per compound from the season's own pace distribution rather than fixed cut-offs, so a "push" on a hard tyre and a "push" on a soft are judged against their own compound. The first representative green laps of a stint are labelled WARM_UP. Laps run under a Safety Car, VSC or red flag are labelled CAUTION, because no driving mode can be read from them.

When to use it

  • Tell the story of a stint on screen: when a driver switches from managing tyres to pushing, it usually signals a stop or an attack coming.
  • Annotate a timing tower or lap chart with what each driver is doing, not just how fast they are going.
  • Separate a slow lap that is deliberate tyre management from one that is a genuine lack of pace.

Inputs

  • degradation-detrended pace
  • lap-to-lap pace variability
  • tyre compound
  • stint lap
  • track status

Output

drivers[].pace.modePUSH | HOLD | MANAGE | WARM_UP | DELTA for this lap, or CAUTION when the lap ran under SC, VSC or red flag.

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": "1",
      "tla": "VER",
      "pace": {
        "mode": "PUSH"
      }
    }
  ]
}

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

PUSH means the lap was run faster than the driver's degradation-adjusted baseline for that compound; MANAGE means slower than it; HOLD sits in between. DELTA marks laps whose pace is unusually variable rather than consistently fast or slow. A single lap can be noisy, so look at runs of the same mode across several laps. CAUTION is not a driving style — it says the lap cannot be read.

Limitations

  • Pit in- and out-laps and caution laps are not scored as driving modes. If the latest lap could not be scored, the last representative mode is carried.
  • In wet running most laps read as DELTA, so the label carries less information.
  • Published in Race and Sprint sessions only. If the scoring service is unavailable the field is omitted rather than guessed.

Calibration

Mode boundaries are re-fitted per compound from each season's own pace distribution rather than set by hand. See the calibration reference for the shared pipeline.

Related models

  • Pace Normalization (LTOE) — Lap Time Over Expected — the normalized pace delta, per lap.
  • Tyre Cliff — P(tyre performance cliff within the next few laps), per driver per lap, plus live tyre health.
  • Overtake Probability — P(a clean on-track pass on the next lap), per attacking driver.

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