Predictions
Learn how Componentry predicts component wear and race-day readiness based on how you actually ride.
Manufacturer specs assume an average rider. Componentry's personalised predictions look at how you actually ride — your terrain, power, weather exposure, and riding style — to estimate when each component on your bike will run out. They show up across the app once you've connected a bike computer and logged enough rides.
What Predictions Give You
Predictions turn flat, linear wear bars into an estimate grounded in your reality:
- Personalised replacement dates — see when each component is projected to run out at your current riding rate, not a generic mileage spec.
- Adjusted lifespan — the distance the component is likely to last on your bike, compared to the manufacturer's number.
- Wear drivers — a plain-English breakdown of what's causing wear on each component, so you can budget, adjust, or just know what to expect.
- Race-Day Readiness — pick a future event date and see every component's projected state on that day, with a clear action list for what to replace beforehand.
[Screenshot placeholder: Component detail showing personalised prediction alongside standard wear bar]
How Predictions Work
Predictions read the data already flowing into Componentry from your rides — distance, elevation, power, cadence, speed, weather at your location — and project it forward. The estimate is specific to the bike the component is on: your gravel bike's terrain profile doesn't distort your road bike's chain estimate.
A few principles are worth knowing up front:
- Predictions sharpen over time. The first estimates are directional; as more rides come in, confidence rises and the numbers tighten.
- Manufacturer specs stay visible. Personalised estimates are shown alongside manufacturer lifespans, not instead of them. You always see both.
- Predictions refresh automatically. Every time a new ride syncs and every time you log a component replacement, relevant predictions update.
- You don't wait for them. Predictions are computed in the background and cached, so component pages load instantly.
[Screenshot placeholder: Bike detail showing several components each with a predicted replacement date]
What Predictions Cover
Predictions are available today for the components where wear is primarily ride-driven:
- Chains
- Cassettes
- Brake pads (disc and rim)
- Brake rotors and calipers
- Tyres and tubes
- Cables
- Chainrings
- Derailleurs
- Cranks, spiders, and pedals
- Wheel rims and discs
- Suspension components
- Fluids
Some components — bar tape, grips, saddles, and other mostly-cosmetic or time-dominated items — don't currently have personalised predictions. You'll see a note on those component types explaining why.
Where You'll See Predictions
Predictions surface in a few places across the app:
- Component detail — a dedicated Personalised Prediction section with distance remaining, predicted replacement date, adjusted lifespan, and the "What's driving wear" breakdown.
- Bike detail — each component shows its projected replacement date alongside standard wear bars.
- Race-Day Readiness — a dedicated page per bike for planning around a future event.
- Advanced Insights card — on the Apps page, showing your connection state, ride count toward the next tier, and which providers power predictions.
[Screenshot placeholder: Advanced Insights card showing progress toward the next prediction tier]
What You Need to Unlock Predictions
Personalised predictions require detailed ride data that only comes from FIT files generated by a connected bike computer — Wahoo, Garmin, or Hammerhead. Strava syncing alone tracks your activities but doesn't expose the granular power, cadence, and elevation data predictions need.
Predictions unlock progressively as you log rides with a supported bike computer. The next sections cover exactly how that works, along with detail on each prediction feature.
Privacy
Your ride data stays in your account. Predictions are computed from your own activities and never shared.
The following sections walk through unlocking predictions, reading personalised wear estimates, interpreting wear factors, and using Race-Day Readiness to plan around an upcoming event.