Check stats for Greg's 15x3min/3min workout

[sanderroosendaal]

Downloads.zip

[sanderroosendaal]

Second interval - the CSV looks like this:

w 1 144 176.09 787.5 112.64
w 1 145 178.44 798 112.61
r 1 146 182.6 809 113.05
r 1 147 185.29 819.8 120.49
r 1 148 187.76 827.9 151.92
r 1 149 191.48 839.1 166.27

So pace 1:52.6 when starting last stroke at 2:59.9 (according to my algo). 798m done. The first rest stroke starts at 809m.

Extrapolation at that pace gives an additional 0.4m, so
total work distance should be 798+0.4 = 798m. PM5 gives 805m. :-/

Using ElapsedTime, 178.4 seconds at final W stroke, we would get 798m+7m = 805m. So ... the question is why rowingdata counts the elapsed time as 1:59.9 instead of 1:58.4.

[sanderroosendaal]

OK, the calculation is done right. Rowingdata adds 1.4 seconds worth (6.2m). However, it subtracts the 3.8m " Horizontal (meters)" value of the first stroke. Probably the cum_dist calculation ...

[sanderroosendaal]

Solving this is possible, but at the risk of messing up results for Just Row (when you add intervals after the row).
The ' Horizontal (meters)' value resets to a low value at the start of each interval. This low value represents the meters that you glide between the start of the interval and the start of the first interval stroke.

However, this data is only available on workouts programmed as intervals - not when you insert intervals after the fact or try to split set intervals into smaller chunks.

The code could of course try to detect what is the case (df.iloc[startintervalindex-1,' Horizontal (meters)']>df.iloc[startintervalindex,' Horizontal (meters)'] and then add the value of the first stroke ...