UKH

In lieu of Strava changing their rules (see other thread on milking the pandemic), I signed up to premium. I've been running a lot in 2020 - much more so since the lockdown began. I got a new watch in February which records heart rate, the data of which Strava has been able to pump into the fitness and freshness app. 

As the year has gone on my fitness has increased massively. Starting at 0 and building to over 100. With it, my fatigue has risen to over 100 as well. My form is negative. 

Is the idea to aim to go for a tempo/race at the point where all the lines cross? This would be Sunday (3 rest days) when my fitness will have dipped slightly, my fatigue will have dropped below my fitness (for the first time in about 4 months) and my form will have changes from -34 to +8. If fatigue is super high, I'm assuming injuries are more likely? (Edit to add: I only currently take one rest day a week but some of the runs are easy)

Funnily, I'm not really aiming for anything - just started running again to stay fit and focus the mind in the mornings. For lockdown, it was my excuse to get out the flat and get some fresh air/sunshine.

Anyone got any links to decent articles on this? Really appreciate the heads up. 

I’m not sure the algorithms for these things, presumably based on heart rate, are particularly good. The fitness and recovery metrics built in to my Garmin watch are certainly pretty laughable. I would take with a huge pinch of salt.

If you believe anything aim for a low negative form unless heading for a race.
 

Unfortunately I suspect it is not very accurate even if you have taken particular care to set up you HR and running speed zones. Garmin’s version is very flakey.  Training Peaks version makes the most sense to me. I have run very well with poor scores and vice versa based on Strava. 

Can't really comment on the strava metric - but with running injuries are normally correlated to sharp increases in training load (i.e. more miles or faster pace).  I suspect the fatigue metric is more a representation of how well rested you are and hence how likely you are to perform well on a heavy session / race.

Yeah I think these things are so rough, maybe over time they'll get more accurate.

I just updated my watch and didnt bother with a more expensive one that can give me recovery and VO2 max estimates for exactly that reason. I'd rather just race hard and use a VDot Daniel's Calculator which is more accurate and useful for me as I use Daniels stuff a lot.

I don’t use that on Strava but Training Peaks has something similar - with that you should aim to race at +5 to +10. You basically build your fitness up which raises fatigue, then taper off so while you might lose a little fitness you gain in freshness. 

If it helps the strava and trainingpeaks fitness/freshness calculations AFAIK use the same equation or at least very similar, just calculate the individual session scores differently. It's very simple to code up in excel and have a play around with. Garmin uses a different system with recovery metrics, vo2 max estimation etc licensed from firstbeat.

With the strava/trainingpeaks one you need a bit of experience to tell what works best for you performance-wise. Some people like the balance to be as positive as possible, I like to be just slightly in the plus, around +5 on trainingpeaks.

Note that it tends to under estimate the effects of sessions of short intervals in my experience.