something to consider (minorly) wrote:
AquaDyne wrote:
The simplest explanation of this would be that data is rounded to integer RPM from some calculation that produces fractions, and is consistently rounded in a way such that every third number rounds down (vs. up). So 59.5 (round up), 60.5 (round up), 61.5 (round down), 62.5 (round up) could produce the pattern you see.
As an aside, although you provide a reasonable explanation of what is going on with SW's data set, do you have any idea why they would choose to round their numbers that way, instead of just always rounding up (or down)?
No, but that's also probably an unlikely explanation. It's just a hint toward the fact that rounding is clearly involved somehow in the calculation.
Consider stride data rounded to the nearest tenth of a stride per second (e.g., 1.4, 1.5, 1.6). Now, maintain a running average of the last 3 readings with the caveat that you'll always have 2 of the same number and one of the number above or below it (e.g., 1.4/1.4/1.5 or 1.4/1.5/1.5 or 1.5/1.5/1.6). Average those 3 numbers, multiply by 60 (to minutes), divide by 2 (convert cadence to rpm) and round to nearest integer. The results are 40, 41, 43, 44, 46, 47, 49, 50, etc. These match the "holes" in the histogram.
There's still an explanation needed for why the 3 readings are never identical (1.5/1.5/1.5) or consecutive (1.4/1.5/1.6) but it seems reasonable to think that there's an earlier calculation step involving roundoff that leaves this as an artifact.
Bottom line, I'm not suspicious.