This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Cumulative average from split times doesn't match overall average

I have a 220 device that i use for both running and cycling. i was calculating a cumulative average based on the split times i downloaded from Garmin connect. i would think the cumulative average should match the overall average. Am i missing something?
  • Depends how you have calculated your cumulative average - you can't average an average - but if you are saying that you add the distances for the splits and add the times for the splits and then divide one by the other and get a different average pace/speed then that is strange but may be down to moving time vs total elapsed time. Would be useful to see the activity to understand better.
  • thanks

    thanks...i guess i didn't realize you couldn't average an average , matches now
  • No worries its a common mistake - seems like it ought to work but it doesn't :)
  • Former Member
    0 Former Member over 9 years ago
    My idea behind the Speed/Average "mismatch"

    Depends how you have calculated your cumulative average - you can't average an average - but if you are saying that you add the distances for the splits and add the times for the splits and then divide one by the other and get a different average pace/speed then that is strange but may be down to moving time vs total elapsed time. Would be useful to see the activity to understand better.

    Have a look at this average chart:


    I never really payed attention to the graphics on Garmin connect. Since I noticed my actual speeds shown on my watch while running where faster then the average, I decided to take a look at the GPS (XML based) file in Excel. I calculated several things, of which the average speed (and pace) on the current speeds. It might not surprise that the calculated average was significantly faster than the average shown on the chart or in the splits.

    Looking deeper into it, the total distance (as well as the 1K laps) seemed pretty accurate and since I do trust the stopwatch, the average (distance / time) was correct. I concluded it should have had something to do with the accuracy of the individual measurements done by GPS. Since the watch has a high frequency of checking your GPS coordinates, a mismatch of a couple meters will result in a significant mismatch of speed.

    For example, when you run a pace of 5:00 minutes per kilometer (1000 meters), you go 12 kilometers per hour. This is 3,3333meter a second. When your watch records 4 meters every second it measures, your pace is less then 4:00 minutes per kilometer (14,4 km/h).

    So my verdict. The actual GPS speed is not reliable at "low speeds". Since I used the actual speed to control/monitor my pace, I decided to switch over to the lap pace "counter" on my watch. This gives you an current average since the start of your current lap.

    I hope this helps in understanding the speed vs average "mismatch".