Rolling km NGP
Suggest to put rolling km NGP data field and avg watt also good to know rolling ascent descent.
@lexterm77 Not clear on your suggestion? Avg watt is a data field available in custom sport modes. NGP is available on TrainingPeaks Running Pace S+ and is available on Climbing S+. Ascent/descent per h is a data field that can be added to any custom sport mode.
Average rolling 1km NGP is not on TP pace app, on that app NGP is a derivative(over a some short time interval it usually figures out in 20s that I started climbing steep trail and my NGP starts dropping fast)
My suggestion is similar to what power avg 3s 10s 30s data fields are already present but not as a function of time 3s,10s,30s but as a function of last traversed km.
Same for watts, ascent, descent. Basically data field has a one km short term memory( or last mile) and only for that distance in past you have average values of NGP, watt, ascent, descent(and HR). This is not same as pressing lap or having auto lap because what you are seeing is always you last km effort.
If the calculation is already done for HR rolling kilometre, it is very easy to integrate power ascent and descent averages of last kilometre as this is a great assessment of your last mile effort.
What is this useful for? It is useful for training specificity of ascending , descending in trail running. You can easily keep consistent NGP and watch if you have trained enough for descending, ascending or flat high cadence run.
Lets say your last mile rolling values are: 7min/km NGP 120bpm 150W 33m ascent when you are not trained for climbing, and as your muscle group for climbing improves you have a metric that would ALWAYS have constant distance, all other values are variable, so when you see after periods of training for ascent you will see something like NGP 7m/km 120bpm 150W 45m you improved your climbing specificity. Same goes for descent or flat training, you can measure improvements of specific skills and how they impact your metabolic output, speed, climbing and your terrain would (if sensors were perfect) be invariant.