Correct, inserted workouts with various TSS to change CTL and test the system. Turns out I can perform at FTP for many hours on end .
The normal function is indeed only an approximation but surprisingly accurate. Was puzzled by that as well. Not sure Join even uses an analytical function, and saw some hysteresis in the score when cycling through inputs. Don’t think RPE affects the score, but (deviation from expected) is probably used as input to adjust future workouts.
Looked into fatigue resistance a little (through intervals.icu). Interesting but don’t think that is easily applicable in a score. It would require very specific workouts, result in few data points and have limited relevance for most users. Personally I never do or need an all out effort at the end of a long ride, cardiac drift is a much more relevant indicator for me (has its own issues though).
I have 3 x 1.5 hr during the week and I maxed out the weekends at 2x6 to see what it would give me.
It started around 8hr per week and grows to about 13.5 hr 24 weeks from now.
My weekend rides for now are not longer than 3 hours. Maybe there is no real benefit to 4 hour rides when you are not yet enough advanced or when the goal you set doesn’t call for that.
Just like with the regular power profile, you know which sections are more relevant because you did an all-out effort for that duration, and which parts of the curve can be ignored. For the fatigue resistance curve to be meaningful you do need all-out efforts after an hour or two of riding, these data points are harder to come by indeed.
M - 4:30
T - off
W - 5:00
Th - 2:00
Fr - 4:00
Sat - 1:00
Sun - Off
It peaks our at 13 hours. That may be to do with the proximity to race day and not leaving time for "base*. Also, all the rides have some sort of effort in them so far. It’s a different philosophy to what I’m used to.
Definitely the sliders thing is the coolest feature.
I’m trying it out at the moment.Very early days but the one thing that strikes me (in a positive way) is they give the impression that it’s OK to go easy. I’m using the Xert workout player and you can import the workouts into that using the Zwift format very easily. I’m not sure how to run them in TR though. EDIT: I must have done something right because the workout has just appeared in my TR calendar!
@carytb Yes. That is what I really really like. Other solutions (including Xert) I always felt pressured into doing a ton which never showed as enough and if I didn’t I felt guilty. With JOIN I’m not feeling that at all.
So slightly higher intensity then what the 1.5 hr workout would have been. Since I am always at the high end of the range the workout went 70%->75%->80%->85%.
Does not look very intense, but I must say it was just enough to leave me with a very very light tingling sensation in the legs. Tomorrow is rest day and looking forward to VO2max session on Friday, which is also including some endurance by putting the 30-30 blocks in the middle of a constant 70% block:
So JOIN did something for me that I’ve been waiting a long time for: I marked a planned workout day as sick (RSV, it was doing bad things with my usually well-controlled asthma), and the app actually cancelled the next three days of workouts as rest days, then marked the next few workouts as endurance only.
I don’t remember the details, but basically here’s what it did (all are 90 minute workouts). Originally planned: Z2 with 40s sprints x 8, intensive endurance, rest day, 60m Z2 + 30m threshold, tempo After marking a sick day: rest day, rest day, rest day, intensive endurance, Z2 w/ some tempo
Very cool, as it not only cancelled the day’s work, but gave me a few full rest days then eased me back into things.
Interestingly Join seems determined to have me do a VO2 workout. It looks like they want athletes doing one every week during “build”.
I haven’t really had a chance/opted into any workouts yet as I’ve been trying to build a second base
before I’m limited to shorter rides over summer.
I like a whole heap about what Join does, not sure it’s as good as TR yet though. Heaps of the features resemble some of the features Nate has hinted at an awful lot.
.
