The Measurement Science of Power Meters: Strain Gauge Placement, Zero Drift, and the Real Meaning of ±1% Accuracy

Power Meters Do Not Measure the Same Thing

Power meters on the market are categorized by measurement location into crank-based, pedal-based, hub-based, spider-based, and others. Their common principle is to use strain gauges to measure deformation, calculate torque, and then multiply by angular velocity to obtain power. However, different measurement locations lead to different sources of error and consistency—this is the fundamental reason why “numbers don’t match up after switching power meters.”

Characteristics of Different Measurement Locations

What ±1% Accuracy Actually Means

A stated “±1% accuracy” does not mean every second’s reading is accurate to within 1%; rather, it is the overall error bound under specified conditions. In practice, this means: you cannot use the absolute values of one power meter to finely compare against another; when comparing power across devices, a difference of 2 to 3% is likely just instrument variation, not a change in fitness. Training decisions should be based on trends from the same device, not absolute values across devices.

Temperature Drift and Zero Offset

Strain gauges are sensitive to temperature. Riding from a warm indoor environment to cold outdoor conditions will cause the zero point to drift, leading to a systematic offset in power readings. This is why zero offset calibration should be performed before departure and after significant temperature changes. Ignoring this step may mean your entire training session’s power data carries an invisible bias.

Discipline for Using a Power Meter Well

• Consistently use the same device to track trends, avoiding cross-device absolute value comparisons
• Perform zero offset calibration before every ride and after drastic temperature changes
• Be aware of the error introduced by left-right imbalance when using single-sided estimation devices
• When interpreting progress, first confirm whether the gap exceeds the instrument’s error band

Why This Matters to All the Previously Discussed Training Topics

FTP, W’, fatigue decay rate, pacing strategy—all of these are built on power data. If you cannot even distinguish “how much of a gap counts as a real change,” then all the sophisticated training models discussed earlier will be contaminated by instrument noise. Power meter literacy is the foundation of all other data-driven training.

Power meters give us an illusion: that numbers are the truth. But behind every watt reading lies strain gauge deformation, temperature drift, and measurement location bias. Those who truly know how to use a power meter are not the ones staring at absolute numbers, but those who understand how large their device’s error is, and therefore know which “improvements” are real and which are just noise lying to them.

Related Reading

• Power Meter Selection and Calibration: The Impact of Accuracy Error on Training Data
• Power Meter Selection and Calibration: Differences Between Static and Dynamic Calibration and Recommended Frequency
• The Complete Guide to Cycling Power Meters: Accuracy, Installation Methods, and Usage Recommendations by Brand
• Power Meter Buying Guide: Accuracy and Convenience of Hub-Based vs. Pedal-Based vs. Crank-Based