Two battery numbers, both shown
Our own SoC-normalised range measurement sits beside the state-of-health figure Tesla reports for your VIN. When they disagree, you can see that they disagree.
Why one number is not enough
The usual way to judge degradation is to read the range shown at a full charge and compare it with the car's original figure. That number moves with the estimator's mood, with temperature, and with how you have been driving, so a single reading tells you very little and a chart of single readings tells you less.
Our measurement normalises for state of charge and uses only samples taken under conditions where the comparison holds. Tesla also publishes a state of health for your VIN through the Fleet API, computed by a method we do not control and cannot audit. Both are shown, with the history of both, and neither is presented as the truth about the other.
What a disagreement means
Sometimes the two numbers separate. That is information, not an error: they measure different things by different methods, and the shape of the gap over months is more useful than either figure on a given afternoon. The page says which is which, and does not average them into one comfortable percentage.
We publish no fleet degradation curve until we have a sample worth publishing, and when we do it will carry its n. The beta fleet is our own cars.
Under the surface, on Pro
Stats for geeks adds what only a streaming subscriber can see at all: brick voltage spread, module temperatures, isolation resistance, pack current and voltage. Battery health itself, and Tesla's own state of health, are on the free plan.
Free forever for one car. No card.