30 day summaries are misleading
-
@silentvoyager ok and I’m on iPhone. Strange that these two apps are so different. Thanks for the info.
-
@silentvoyager hmmm
Does this help?
Actually neither is right if you look at the total distance and the total time. For example the average speed should actually be 4.73 mph in this case, and the average pace is 12:40/mile. It appears the app simply averages numbers from all activities, which doesn’t really make sense for speed or pace. That would be correct only if all durations were the same. If I do a long 3 hour run at 6 mph, than do a 30 minute tempo run at 10 mph, is my average speed actually 8 mph?
This is not an average for the total distance and time as it would make no sense if I understand correctly. It add’s no value to say:
on the last 30 days I did 50km and had a duration of 10h so it’s 5km/h You want to get the pace of the individual activity and average that as you would do for HR -
@Dimitrios-Kanellopoulos, I understand that for each activity we want to see averages - that makes total sense. But averaging averages from multiple activities doesn’t make sense and may be misleading. The same applies to HR by the way. This averaging of averages kind of works if durations are about the same, but it will skew results if activities are substantially different in duration and intensity.
-
@Dimitrios-Kanellopoulos, to give you another argument, does it make sense to you that my average speed over last 30 days of trail running was 5 mph, and the average pace - 12:36/mile. Isn’t speed and pace essentially the same thing, just different (inverted) units of measurement.
-
@silentvoyager I agreed it can be misleading but on another argument I d say that summing up your HR and averaging it for X (30days can be too much) days could indicate how hard you are training. I would love to see that and it’s quite useful to see how it compared to another period no?
-
@silentvoyager said in 30 day summaries are misleading:
@Dimitrios-Kanellopoulos, to give you another argument, does it make sense to you that my average speed over last 30 days of trail running was 5 mph, and the average pace - 12:36/mile. Isn’t speed and pace essentially the same thing, just different (inverted) units of measurement.
Yes you are correct. I need to ask why this difference. I suspect some wounding is done as both are very close.
-
@Dimitrios-Kanellopoulos, yes the trend is a good thing to have, but the average is meaningless unless you do a weighed average. Better not have it at all.
-
@silentvoyager agreed, I am all in favor for a weightened avg.
BTW regarding the speed to pace etc I have the same incosistency but I am sure it’s due to rounding.
My AVG PACE for trails is 8.4269662921348 -> 8:25m/km
MY AVG SPEED Is 7.1 that sounds like not correct. But doing the maths manually I get a speed of 7.12 so for sure its just rounding as I see no speed values in those charts with a second decimal. Makes sense? -
@silentvoyager Do you have btw any suggestion? I d love to transfer user feedback to Suunto.
-
@Dimitrios-Kanellopoulos said in 30 day summaries are misleading:
BTW regarding the speed to pace etc I have the same incosistency but I am sure it’s due to rounding.
It isn’t the rounding. Averaging of speed is just mathematically wrong. Consider the following example. You run one km at 6 km/h (10 min/km), then you run another km at 10 km/h (6 min/km). Averaging pace gives you 8 min/km and that makes sense - you covered one km in 10 minutes, another in 6 minutes, so 8 minutes on average. Now, if you just try to average speed from two sepate km you’d get 8 km/h (the average of 6 and 10), which is 7:30/km when converted to pace. So we get the difference without any rounding. The difference may be even more striking if I pick other numbers.
The simplest solution is to just divide the total distance over the total time for speed or vice versa for the pace. That would be mathematically correct. Or at least always average the pace and convert it back to speed.
-
@silentvoyager all data are recorded in speed units. I am not talking about first averaging the speed and then converting.
The rounding is about the display value. Makes sense? On the chart 30 difference on the avg that is what I was talking about.
