Two riders log the same distance on a new chain. The first rides flat, dry roads in mild weather, spinning a comfortable cadence. The second spends their season in the mountains — long descents, punchy climbs, regular rain, and normalised power 15% above their baseline. The manufacturer says both chains should last the same distance. The science says otherwise. Only one of those riders is still within safe operating limits.
Distance-based tracking has always been the practical ceiling for maintenance tools. If the manufacturer states a component's lifespan, you get an alert when you reach it. It does not matter how that distance was ridden. This works well enough as a starting point. But component degradation is not linear, and it is not equal. The same consumable, ridden differently, wears on a fundamentally different curve.
Componentry now accounts for this.
Why the Same Distance Tells a Different Wear Story
Chain elongation is driven by the forces acting on the drivetrain, not the distance alone. At each link, the interface between pin and inner plate wears through a combination of mechanical load, abrasion from environmental contamination, and the quality of lubrication at the time of contact.
SILCA's technical analysis of chain friction identifies the inner plate-to-roller and inner plate-to-pin interfaces as the primary sites of energy loss during drivetrain operation — not the roller-to-cassette contact most riders assume. (SILCA: Chain Friction Explained)
When power output increases, the forces at these interfaces scale proportionally. A rider producing 300 watts through a climb is loading their drivetrain at a fundamentally different rate than a rider spinning 100 watts on a flat recovery ride. Multiply this across every ride in a season and the divergence between two riders covering the same total distance can be significant.
Weather compounds the effect. Water and grit act as abrasive compounds within the chain. Zero Friction Cycling's extensive real-world chain testing demonstrates that contaminated, wet conditions can reduce chain lifespan to less than half of what the same chain achieves in clean, dry conditions. (Zero Friction Cycling)
Terrain adds a further dimension. Descending at high load with rim brakes wears brake pads at a rate that flat-road estimates miss entirely. Extended climbing places consistent drivetrain stress that accelerates cassette wear. The gradient profile of your rides — not just their length — directly shapes how your components degrade.
The data to model this has always existed in your activity files. Elevation gain and loss, average power, normalised power, cadence, temperature, rainfall. What has been missing is the intelligence layer that transforms these raw metrics into a personalised wear projection.
What Componentry Now Calculates
The new predictive system introduces four capabilities, each building on the same underlying model.
Personalised Remaining Useful Life
Rather than projecting forward from the manufacturer's stated limit, Componentry now calculates a personalised estimate based on your actual riding behaviour. Every activity you sync contributes to a riding profile: your typical terrain, your average power output relative to your baseline, your weekly volume, and the weather conditions you ride in.
This profile generates an adjusted lifespan estimate and a projected replacement date. "At your current rate, replace by approximately 12 May" is a different and more useful statement than "you have reached 72% of the manufacturer's estimate."
Adjusted Manufacturer Estimates
Manufacturer specifications are built for an average rider in unspecified conditions. They are a useful starting point. They are not a reliable endpoint for a rider with a specific terrain profile, specific power output, and specific weather exposure.
Componentry now surfaces both numbers side by side. The manufacturer's stated limit and your personalised adjusted estimate, with a clear multiplier. If your riding style means you wear through a chain 30% sooner than the stated limit, you see that directly, and you see why.
Wear-Rate Factor Analysis
For each component, Componentry now produces a breakdown of what is driving wear at your personal rate. The contributing factors are drawn from your actual activity data, not generic assumptions.
An example output for a chain on a performance road bike might read:
- Climbing: 30% impact — Terrain score significantly above average for your ride profile
- Power output: 20% impact — Normalised power 15% above your own baseline
- Descending: 25% impact — High descent volume raises drivetrain stress
- Weather: 15% impact — 25% of rides logged in wet conditions
- Volume: 10% impact — Weekly riding volume above the population average
Each factor is tappable for further detail and contextual guidance. "Riding in wet conditions accelerates chain wear. Consider more frequent lubrication after wet rides" is advice calibrated to your situation, not generic maintenance advice.
Race-Day Readiness
The fourth capability is a forward simulation built for riders with a specific event on the calendar.
Select your race or event date, choose the bike, and Componentry projects the state of every component on that date based on your current riding patterns. The output is a readiness report:
- Components that will be in good condition: no action needed
- Components to monitor: consider replacing if the event is critical
- Components to replace before race day: specific recommended windows for when to order and fit
- Components projected to be overdue: will likely need replacement before the event date
The simulation accounts for the weeks between now and race day — your typical weekly volume, projected seasonal weather for your location, and an optional training load modifier if you know you will be building volume in the lead-up. As new activities come in, the projection updates automatically.
A rider with Unbound Gravel at the end of May can look at their bike today and get a specific answer: "Replace your rear tyre by 10 May. Your chain will be fine. Monitor your rear brake pads." That is a different kind of confidence than guessing based on rough usage estimates.
How the Model Works
The system computes three scores for every activity, derived from the data already in your activity files.
Terrain score is calculated from elevation gain relative to distance ridden. A flat ride scores low; a mountain stage scores five times higher or more. This single value captures the descending load (for brake pads and rims), the climbing stress (for drivetrains), and the technical nature of the terrain.
Intensity score is a normalised composite of power output, speed, and heart rate relative to your own baseline. Your easy endurance rides and your threshold intervals produce very different wear profiles for the same distance.
Weather severity combines temperature, precipitation, and humidity into a 0 to 1 score representing how hard conditions were on your components. This data is pulled from historical weather records matched to your activity's location and time. You do not need to log it manually.
These three scores combine into a ride stress score — a single number representing how hard a ride was on your components relative to a baseline flat, dry, easy effort. This score, aggregated across your recent activities, drives the personalised predictions.
The system applies progressive personalisation. New users start with population-level estimates informed by your emerging riding profile. Once you have accumulated enough activities on a specific bike, the model transitions to fully individualised predictions. A confidence indicator on each prediction — "Based on 127 rides" — makes the data volume explicit.
What This Looks Like in Practice
The predictive estimates appear in the component detail view alongside your existing wear bars. The adjusted lifespan bar shows your personalised estimate next to the manufacturer's. The predicted replacement date is displayed prominently. The factors card explains, in plain terms, what is driving your particular wear rate.
For your bike overview, the summary now shows the component closest to its predicted replacement date — your next action at a glance.
For riders preparing for an event, the race-day simulation is accessible directly from the bike detail page. Select your event date and get a complete readiness report. As your training progresses, the report updates.
Proactive Maintenance, Not Reactive Recovery
Component failures do not announce themselves in advance. A chain does not notify you when it crosses 0.5% elongation. Brake pads do not visibly change as they approach the wear limit. The audible cues — skipping, grinding, squealing — are symptoms that arrive after the damage is already underway.
The difference between a predicted replacement date and a mid-ride failure is not always a matter of how much further you can ride. It is a matter of knowing your limit before you reach it.
Componentry tracks the data your rides generate. The predictive layer converts that data into specific, actionable information: this component, this date, this reason. Not a guess based on distance alone. An estimate based on how you actually ride.
Recommended Reading
Understanding component wear mechanisms:
- SILCA: Chain Friction Explained — Technical breakdown of where friction originates in drivetrain systems
- Zero Friction Cycling: Chain Longevity Data — Independent wear testing at scale across real-world conditions
- Park Tool: When to Replace a Chain — Industry standard thresholds and the science behind them
Related Componentry posts: