Crash Intelligence · Tour de France 2026 · v2.4.1
Before you enter
PeloRisk surfaces AI-generated crash risk scores for professional cycling stages. This tool is built for race safety officials and credentialed personnel. Please read and agree to the terms below before accessing the platform.

1. Decision Support Only. PeloRisk is a probabilistic decision-support tool. Risk scores are outputs of an AI model trained on telemetry, terrain, and historical incident data. They are designed to inform — not replace — human judgment by qualified race directors, commissaires, and safety officers.

2. The Model Can Be Wrong. No AI model can fully predict crash events. PeloRisk may produce false positives (high scores on incident-free stages) and false negatives (low scores on stages where crashes occur). Users must not rely solely on this tool for any safety-critical decision.

3. No Automatic Actions. No score output from PeloRisk automatically triggers race neutralization, medical dispatch, or any operational intervention. All actions remain the sole responsibility of authorized race personnel under applicable UCI regulations.

4. New Climb Uncertainty. Stages featuring climbs with no prior Tour de France history carry elevated model uncertainty. These outputs should be treated with additional caution.

5. No Affiliation. PeloRisk is not affiliated with, endorsed by, or operated under the authority of ASO, the UCI, or any official Tour de France organization.

6. No Liability. The creators of PeloRisk accept no liability for decisions made based on the outputs of this tool. Use of this platform constitutes full acceptance of these terms.

Scroll the terms panel to read in full before agreeing.
AI-Powered Race Safety · v2.4.1

The signal
before the
chaos.

PeloRisk ingests live telemetry — peloton density, speed differentials, gradient, weather, and historical incident data — and surfaces a crash risk score to race directors in real time. Not a verdict. A signal.

21
Stages Analyzed
3,333km
Total Route
54Km
Elevation Gain
6
Critical-Risk Stages

What the model reads
📡
Positional Data
GPS-derived peloton density, inter-rider gaps, and pack fragmentation at 2Hz refresh rate.
Speed Delta
Velocity differentials across the pack — sudden deceleration waves are crash precursors.
🏔
Terrain Profile
Live gradient, road surface type, corner radius, and descent speed potential from route files.
🌬
Weather Feeds
Crosswind index, precipitation probability, and visibility — all correlated with historical incident data.
📊
Historical Crashes
30-year incident database by road segment, stage type, km-to-go, and race fatigue index.
🧠
Race Dynamics
Sprint lead-out tension, GC attack probability, and team formation patterns at key km markers.
Stage-by-stage risk ratings
Critical (80–100)
High (60–79)
Moderate (40–59)
Low (0–39)
↓ Click any stage for details
Tour de France 2026 — Barcelona → Paris
Jul 4 – Jul 26 · 21 Stages · Model v2.4.1
# Date Stage Type km Risk Score Key Flags
Human in the loop. Always.

The Model's Job

PeloRisk surfaces probability, zone, and reason — not verdicts. A score of 87 on Stage 6's Tourmalet descent means the confluence of 180 riders, a 9% gradient, and forecast crosswinds puts this window in the top decile of historical crash risk for this road segment.

The model never triggers an action. It makes the human smarter.

The Human's Job

Race directors, medical cars, and commissaires receive the signal. They decide whether to accelerate neutralization, reposition vehicles, or broadcast warnings. False positives cost race time. False negatives cost lives. That asymmetry is why the model flags and the human acts.

No False Confidence

If telemetry is sparse — a GPS dropout, a late-joining team — PeloRisk surfaces an "insufficient data" state rather than extrapolate. A wrong score is worse than no score in a 60km/h peloton.

Continuous Calibration

Every incident is logged with its pre-crash telemetry snapshot. The model retrains on each Tour edition, tightening precision on new climb profiles, updated road surfaces, and evolving team tactics. The 2026 model includes 6 new climbs added to the route this edition.

⚠ Disclaimer

PeloRisk is a decision-support tool, not a safety authority. Risk scores are probabilistic outputs derived from telemetry, terrain, and historical data — they are designed to inform race officials, not replace their judgment. No score, however high, automatically triggers race neutralization or any operational action. All decisions remain with credentialed race directors and commissaires under UCI rules.

This tool is not affiliated with ASO, the UCI, or any official Tour de France organization. Stage ratings shown here are model outputs for demonstration purposes.

⚠ Model Limitations

This model can and does make mistakes. Crash risk is inherently unpredictable — a single mechanical failure, an unexpected weather shift, or an unpredictable rider action can cause an incident that no model would have flagged. PeloRisk can also produce false positives (high scores on stages that finish incident-free) and false negatives (low scores on stages where crashes occur).

Scores for stages on new climbs carry additional uncertainty — the model has no historical incident data for those road segments and relies entirely on terrain inference. These outputs should be treated with extra caution.