much appreciated! that thread was great!
OK, here's a shot at something useful for a Strava Chrome plugin:
I can compare the above calculation to a scale that equates this Sunday's Mountain's of Misery race training load calculated by Strava to to a 20 mile run effort at marathon pace estimated to be a Z3 effort, and a 10k swim race effort at an rating of 10 out of 10. Can then compare shorter, more intense efforts used as scaling variables to see how that affects things. Could then create a 2nd order factor that represents three subfactors of short, medium and long max intensities, and use that single factor as a scaling metric. I could then collect friend's perceived equivalents to continue to validate this scale. The plugin could also randomly ask users to rate workouts 1-10 RPE and the plugin would auto re-validate (eCFA) and improve the scaling structure - compare it to users' area-under-Fitness-curves to formulate a sliding scaling structure correlated to user's dy/dx Fitness.
Users could voluntarily share more dynamic data, such as several more perception scales from the user, e.g. quality of sleep, hours of sleep, hours of nap; quality of protein, fat and perception of under/over-eating scales, etc. which can be included in the calculation, with on-going validation as explained above.
wovebike.com | Wove on instagram
OK, here's a shot at something useful for a Strava Chrome plugin:
- Swimming:
- When recording swim workout in Strava, user asked to rate workout on scale from 1 to 10 (RPE)
- User prompted to rate the effect humidity & heat had on the workout on scale from 1 to 5 - weight given to workout.
- Plugin calculates time after last workout - weights are given to workouts that follow another workout within 8 hours time even if on separate days - logarithmic growth model at 15 minute time intervals
- Standardize Training Load variable on 1 to 10 scale
- Model fitting - model ASU '90's 26 week training plan that has quite a bit of evidence-based reliability to refine the above model.
- Cycling:
- Plugin pulls "Training Load" variable from Strava for cycling
- User prompted to rate the effect humidity & heat had on the workout on scale from 1 to 5 - weight given to workout.
- Plugin calculates time after last workout - weights are given to workouts that follow another workout within 8 hours time even if on separate days - logarithmic growth model at 15 minute time intervals
- Standardize Training Load variable on 1 to 10 scale
- Running:
- From the GAP-based Run Pace Distribution page on Strava, multiply a weight for each training zone listed in Strava by the seconds spent in that zone.
- Z1-Z6: 0.65, 0.75, 0.84, 0.9, 0.95, 1.0
- User prompted to rate the effect humidity & heat had on the workout on scale from 1 to 5 - weight given to workout.
- Plugin calculates time after last workout - weights are given to workouts that follow another workout within 8 hours time even if on separate days - logarithmic growth model at 15 minute time intervals
- Standardize Training Load variable on 1 to 10 scale
- Model fitting - model 26 week 5k, 10k and marathon plans out of Lydiard's Run to the Top to refine the weights given to training zones with the evidence-based assumption that a the result Form should be "perfect", in terms of this model. This would weight the value of longer, slower runs; recovery runs; and training at tempo and threshold efforts. Note: I would take into account personal emails Lydiard sent me in the past where he specified that his MWF 60min runs included a sustained AT/threshold effort, and T, R 90min runs and Sat 120-150min runs included sustained AeT/tempo efforts.
- Sum Training Load(s) for each day
- Fitness = previous day's Fitness + (current day's Training Load- previous day's Fitness)/(previous 42 day's avg Fitness [seeded until 42 days reached)
- Fitness value displayed on chart would be dy/dx Fitness - this would take into consideration consistency, volume and intensity over the past 42 days.
- Fatigue = previous day's Fatigue + (previous day's Training Load - previous day's Fatigue)/(previous 7 day's avg Fatigue [seeded until 7 days reached])
- Form =
Fitness - ( Fatigue) - Over-training = Fatigue/Fitness. a value of 1.5 standardized to zero. At or above zero indicates risk of over-training and/or injury.
- Allow manual entry for future Training Loads by sport or by cumulative sports for 2 or 3 by day(s), week(s), month(s)
- copy/paste drag option on calendar to copy Training Loads to multiple and/or non-consecutive days, weeks, months
I can compare the above calculation to a scale that equates this Sunday's Mountain's of Misery race training load calculated by Strava to to a 20 mile run effort at marathon pace estimated to be a Z3 effort, and a 10k swim race effort at an rating of 10 out of 10. Can then compare shorter, more intense efforts used as scaling variables to see how that affects things. Could then create a 2nd order factor that represents three subfactors of short, medium and long max intensities, and use that single factor as a scaling metric. I could then collect friend's perceived equivalents to continue to validate this scale. The plugin could also randomly ask users to rate workouts 1-10 RPE and the plugin would auto re-validate (eCFA) and improve the scaling structure - compare it to users' area-under-Fitness-curves to formulate a sliding scaling structure correlated to user's dy/dx Fitness.
Users could voluntarily share more dynamic data, such as several more perception scales from the user, e.g. quality of sleep, hours of sleep, hours of nap; quality of protein, fat and perception of under/over-eating scales, etc. which can be included in the calculation, with on-going validation as explained above.
wovebike.com | Wove on instagram
Last edited by:
milesthedog: May 25, 16 7:37