Wear two power meters on the same ride? Run a Garmin and a Wahoo simultaneously? Drop both FIT files in. The tool auto-aligns them, shows where each sensor disagrees, runs a power-duration curve for each, and gives you a plain-English diagnosis (offset, drift, dropouts, smoothing differences). All processing is in your browser, files never leave your device.
Compare up to three FIT files from the same activity to diagnose sensor disagreement, offsets, drift, dropouts, smoothing.
Device A is the reference baseline. Label your devices for clearer charts.
Drag horizontally on any chart to zoom, all charts zoom together. Double-click to reset.
Auto-alignment uses cross-correlation on the strongest shared signal. Adjust manually if needed.
Automated assessment of disagreement patterns between devices.
Overlaid GPS traces. Misalignment here usually indicates GPS quality differences.
Positive = other device reads higher than reference.
Mean max power across standard durations. Shows how device choice affects peak power claims.
From the same ride or run. Different head units recording the same activity. Device A is the reference. Files are parsed entirely in your browser, nothing is uploaded. Works with FIT from Garmin, Wahoo, Hammerhead, Stages, Quarq, Zwift, Rouvy, anything that exports FIT.
Even if both head units started recording at slightly different times, the tool finds the offset that maximises agreement on power (or HR, or speed, whichever is shared and strongest). You can override it manually with the offset sliders.
Some files have power but no GPS (smart trainer). Some have HR + power but no cadence. The Detected Sources panel shows what each file holds, useful when you’re comparing a head unit log against a Zwift export.
Categorised as OK / low / moderate / high disagreement. Examples: “Device B reads 4.2% lower than A across the whole ride” (calibration offset). “Power agreement degrades after minute 90” (drift, possible temperature compensation issue). “HR strap shows 12 dropouts” (electrode contact). “Cadence missing on Device C”.
Drag horizontally on any chart to zoom into a specific interval, all charts (power, HR, cadence, speed, elevation) zoom to the same window. Double-click resets. Smoothing options: raw, 5s, 10s, 30s averages.
If your Garmin says you hit 1100 W peak and your Stages says 870 W, the PDC chart shows the gap across every duration from 1s to the full ride. This is what your CP estimate is built on, device choice can shift it by 5–15 watts.
Most athletes never look at sensor disagreement until a number doesn’t make sense, an FTP test that doesn’t match feel, a race-day power that’s 30 W lower than training, a HR that pegs at 200 bpm for two minutes mid-climb. The fix is rarely “the meter is wrong”, it’s usually offset, drift, or a recording gap. Three real uses:
What this tool can’t do: tell you which device is “correct.” That requires a known reference (lab dyno, calibrated HR ECG). What it can do is show you when two devices agree (one of them is probably right), when they disagree consistently (calibration), and when disagreement varies with conditions (drift, environmental).
Drop a single FIT file from a steady-power flat lap, get your aerodynamic drag, same FIT-parser, different calculation.
Open → Critical power (CP)Once you trust your numbers, get your sustainable power threshold from short + long efforts.
Open → Power to finish (course-based)Use confirmed power numbers to predict race time on 132+ real triathlon courses.
Open →Sensor disagreement is rarely random. The pattern tells you what’s actually happening, bad calibration, dying battery, position change, temperature drift, software firmware mismatch. Send your two FIT files plus the comparison output and we’ll tell you what to fix and what to trust.
Bug, missing metric, or feedback? Email info@yousuli.co.
