Following-on from my suspicions as to the accuracy of my car’s reporting of my travel statistics, here’s another mystery. The length of my journey from home to work, as recorded by my odometer this morning, was 24.7 km. The length as recorded by Google Maps is 25.2 km. So, my odometer **underreads**. Or it did this morning, anyway. But the car overreads my speed. If it gets the speed from revolutions-of-the-wheels data, one might expect that an overreading speedometer implies that the speed is being calculated based on a larger tyre circumference than actually exists. That makes sense with my current tyres, which are getting close to needing replacement. (Yes, they are still legal).

But then, assuming the odometer also goes off wheel revolutions, I’d expect that to overread too. But it showed less than GoogleMaps.

Now, also, I note that the ‘average speed’ indicator agreed with the odomometer and clock. The 24.7 km (as the odometer read) was done in 32 minutes which means an average speed of 24.7 x 32/60 = 46.3 km/h. The ‘average speed’ indicator read 46 km/h. At least this is consistent. But, not so fast…We can (and should) do a bit better with this assessment. Given I measured the 32 minutes off my car clock, which reads only to a minute accuracy, it could actually have been 31 – 33 minutes. At 31 minutes we’d get an average speed of 47.8 km/h, or at 33 a speed of 44.9 km/h.

Now, if we instead take GoogleMaps’s 25.2 km, we get average speeds of 45.8 – 48.8 km/h. That also includes the 46 km/h speed indicated by the computer. So, the average speed reading is consistent with both GoogleMaps AND the odometer measures of distance. I could do better by measuring my journey time in seconds, but that involves fumbling for a mobile phone while starting and stopping.

So, the speed information doesn’t help me work out this puzzle (why the odometer underreads but the speedometer overreads). Is the speedometer deliberately set to overread, then?