I haven’t tried.
Add to Saturday…Select from the options a Group Ride..set 4hrs..set the RPE you usually ride at.
Similarly with Tuesday, Wednesday, Thursday when you tend to do 1.5hr rides, select either Solo or Group as describes it best.
The more information you put into you calendar, the better TR will be able to recommend workouts and manage fatigue.
Also, you can then see whether these rides, or a portion of them, benefit or detract from your FTP.
1 Like
@eddie any chance to take a look at this?
How far out is the prediction?
My first few days of predictions when it was 28 days away jumped up and down a bunch. Once I got a week or two in it settled to exactly what it’s going to be in 2 days and hasn’t changed more than +/- 2 watts.
It was 16 days out or so.
I get that it changes (and should). I just don‘t get the massive change after rating a workout easy, that had a 91% probability of being easy.
1 Like
Assuming that is in workout text? That is VERY old legacy text (like from Chad). Remember the goal of the workout is to do the work generally as prescribed. If you are off a little and a 0.1 below the athlete level, that’s noise. If you stop for 15 minutes in the middle of a workout, changes the workout. I keep the workout instructions on for fun…so clearly Chads voice, but I ignore 90% of it. But it passes the time!
3 Likes
@DavidYYC let me look into this and get back to you!
My question is what happens when I skip the cool down after the main sets? I usually do about 1 minute and then jump off becuase I ride at 5 a.m. and need to get to work. Does skipping the cool down and ending at 56 min instead of 1 hr lead it to believe I shortened the workout in a way that affects the prediction? I’d assume yes but was curious if others had ideas.
Sorry to be blunt but you did shorten the workout. You aren’t missing any important long term gains though. I wouldn’t focus too much on short fluctuations on FTP prediction. Focus on building a plan, doing quality workouts, and staying consistent. We never had this feature before. We just trained and looked at long term gains (hopefully!). Don’t stress after every workout.
4 Likes
Hey @AussieRider 
!
I think this is just a great example of how TrainerRoad now recommends workouts. Not every step needs to be higher than the last. Sometimes a lower Workout Level is meant to keep your training consistent and productive so you can see higher gains and not overdo it.
You are also adding quite a bit of unstructured training + doing some sort of training pretty much every day, so the system is taking all of this into account and adjusting your efforts to keep a good balance of training vs. recovery to see fitness gains.
You have a Recovery Week coming up, and then off to Hawaii for a race? I would be mindful of staying fresh and not adding too much prior to it. Also just overall trying to respect Recovery Weeks as they are meant for the body to absorb all the work you’ve been doing and be ready for the next block.
I also see that you added a training annotation that it’s clearing your training day for Tuesday 17th. Is that intentional?
1 Like
I got this sorted yesterday Caro, But thanks anyway!.
1 Like
Are you getting a Cut Short survey after cutting the cool-down short? If so, can you give me an example? I see you are getting full credit on rides you’ve ended the cool down early.
The only ride I am seeing that you didn’t get full credit on everything is this one, but you do have an extended pause here between intervals sets (however, you did get full credit on Threshold which is the intended zone for this workout):
2 Likes
Yay! I see Nick got to you. That’s great!
Hi Caro—
To my knowledge, I’ve gotten credit. I was more curious about if cutting the cool down short had a meaningful impact because I might try to finish the full hour if I did. If it’s irrelevant, I’ll keep doing what I do because it helps me get out of the house faster. Does it make a difference if I finish out those last minutes?
Thanks for looking at this! You all are doing awesome!
Lisa
I am part of the beta group with the update on the 6th. Up until yesterday things were pretty much going as expected. Love the new program and have found all the workouts to be spot on. I completed 3 weeks of a self scheduled VO2 Block. For the first 3 weeks the prediction varied between 320-328 watts. I planned a recovery week and picked up Covid. It was mild so I continued to ride and shortened quite a few of the workouts…Basically emphasized the recovery. This Tuesday I was going to jump into a threshold block BUT decided to give a couple extra days of easy until Thursday. The Prediction stayed around 319/320.
Yesterday i did an easy Zwift pacer ride. After the ride prediction was 319. I then did 3 short commutes 1 to drop my kids at school and then to work and back…about another 1/2 hour of riding. The AI prediction then dropped to 303 after the commutes. Also on the commutes a “why did you end your workout early” message was coming up. Never seen that before. I then re-scheduled the Threshold workout I had pinned to something easier. Prediction went back up to 315. This seems about right. Nailed the workout and it dropped again to 312.
To me, just seemed bizarre on the last 2 days before the prediction comes to fruition it went from 319–>303–>315–>312 in the span of 24 hours!!!
2 Likes
Hi Caro, yes it’s intentional. Travelling on that day. TR recommends I do six structured workouts a week. I still do 5-6. I just prefer to not have all six structured (so I can do an easy 90 min ride and not be worried about traffic lights or take a day off if I feel like it). This was a recommendation at some point by @Nate_Pearson, stick to a LV plan and add your easy rides around it.
as per the recommendation of someone else on this thread I’ve started to add repeating rides for a sat social group ride and a solo easy mid week ride to let AI no I’ll be doing a min of 4 rides per week.
The threshold ride I scheduled for today (4.1) TR AI had a 3.4 scheduled, came with a red do not recommend warning. Stats suggested a high failure rate. It was hard but no where near “very hard” I would say it was a 7/10 effort and that’s with me reducing the 6 min recovery intervals to 4.
I cut 15mins off the end of Dollar Law the other day and the PL police didn’t bat an eyelid.
Knocked a couple of points off the endurance progression score but I didn’t get a failure survey nore did it negatively effect my predicted FTP in a few days.
1 Like
About half way through my first prediction window, and I still struggle to see the value in what feels like a random number generator at times.
I rode a sweet spot workout a couple of days ago, and felt great. I was in two minds about marking it down as easy or moderate. I eventually went with easy. Out of curiosity, I’ve now gone back and changed it to moderate, and the predicted FTP dropped by 9W. There were no subsequent changes to planned workouts. That’s just silly.
I really hope that the prediction feature gets better over time. It’s a very promising feature, but doesn’t seem to be worthwhile for me at the moment.
Does this situation make sense to anyone else? (It doesn’t to me and I was in the beta, I’ve read hundreds and hundreds of posts and I’ve listened to the two Ai podcast episodes)…
TR staff: I’d love to hear from you on this. My TR username is: dbf
On Jan. 9 I got an AI FTP Detection at 219. My predicted FTP since then for Feb. 6 (tomorrow) was mostly bouncing around 227/226 as I proceeded through my Ai prescribed workouts and completed all of them. Then earlier this week I got sick for a few days and skipped one VO2 workout on Monday and one easy endurance workout on Tuesday. Yesterday I got back on the bike and on the plan. After completing yesterday’s ride my FTP prediction is all the way down to 221. Over the course of four weeks I hit all the workouts, missing only one key workout, and the FTP prediction goes down that much right at the end?
But here’s where it gets even weirder. If I go into the calendar and tell it to give me an additional Ai recommended Threshold workout today (normally a rest day for me) it assigns Gayley, predicts it will be Hard, and tomorrow’s predicted FTP shoots back up to 225. Meanwhile tomorrow’s planned workout, Mount Baldy -2, doesn’t change (despite what would be back-to-back threshold workouts).
I understand that the predicted FTP is more volatile earlier in a time period but should get more locked down near the end. I have the opposite happening. Just one workout missed or added is drastically impacting the prediction with only one day left.
Finally, if I delete Gayley from today, leaving it as a rest day, but then click into tomorrow’s planned Mount Baldy -2 and go looking for alternates, Gayley is there but is “Not Recommended”. How could a workout the AI is suggesting I do today and would be Hard, also be Not Recommended as an alternate tomorrow???
You should tag @eddie @Caro.Gomez-Villafane and/or @SeanHurley. Just posting here it’s easy for your post to get missed.
And remember, AiFTP is in beta.
2 Likes