๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰ Introducing AI FTP Detection ๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰

Thatโ€™s my standard method when I skip the ramp test (i.e. all the time). Itโ€™s even better with AT, because it may bump up the progression for that workout category as a result.

Looks like Iโ€™m wrong and the wording is just fine. Sorry for the confusion. Outside imported rides with power data count for the 12 workout minimum.

Thanks. Helps if you read the instructions carefully!

Can I have you edit any/all posts above to correct any inaccurate info?

Iโ€™d like to minimize the potential for readers catching erroneous info and possibly missing the corrected info later. Thanks in advance. :smiley:

This issue is being reviewed by our team. Thanks for reporting!

Is it only outdoor rides?

Maybe this explains the big inaccuracy in my AI FTP explained in my previous post.

Agreed. My next ramp test isnโ€™t due until march 1st. Based on my recent workouts, Iโ€™m pretty confident that my current FTP is set quite good. But still, itโ€™ll be darned hard not to take a sneek peek beforehand :slight_smile:

Coming from a 50 year old amateur who doesnโ€™t like ramp tests at all: this new feature has made my day. Boy, Iโ€™ve been smiling all day now. Thanks TR!

I was thinking of using the completed wo for the same day for 2 weeks earlier, the planned wo for the same day the next week or TrainNow depending how I am feeling :upside_down_face:

Same here! Glad I can skip a FTP test and not self adjust :heart_eyes_cat:

The model looks at any and all rides with power data to make a prediction.

My outside rides with power are using a 4iiii single sided power meter that reads about 4-7% higher then my trainer. Right now all my rides are inside, but how is that going to effect the AI detection, especially if i am doing a hybrid of in and outdoor rides?

I use a different power meter for my outdoor rides than for my indoor rides. How will FTP Prediction handle any differences between the my two power meters?

This is one of the benefits of using a model as opposed to using an estimate. We donโ€™t simply estimate your FTP based on outdoor power data (which is often reads higher than indoors). Instead, we include your outdoor power data in addition to the various data points we already have to predict what FTP youโ€™re likely able to achieve as a result of the Ramp Test.

I wasnโ€™t hugely fussed about having this feature as I always thought I had a reasonable feel for where I was. Quick play this morning and I love it, well done TR

The only item I miss though is when I look on my calendarโ€ฆI dont see the ramp test that was scheduled nor that AI FTP detection was used. I see the replacement ride I didnt do. I do like to see an indication on the calendar that I assessed my FTP.

add a ramp test to the calendar as another workout on the same day, and simply dont do it. You see workouts you didnt do in your calendar as well

Yeah, as mentioned, you can add another Ramp there as a flag, but Iโ€™d skip that.

  • As a hack, I have been adding an annotation to my calendar anytime I make a manual change to my FTP or perform a test. Itโ€™s just pure text with a few short notes and a โ€œflagโ€ that I can track.

  • I do think it would be great if TR also included something automatically on the calendar too, maybe linked to the TS page.

TR gives you the FTP history in your account, and you can also see the FTP change points on the Training Stress page as well.

Another very happy customer. I didnโ€™t loathe the ramp test but did feel a bit of a letdown when after 6 weeks, I would test the same or even 1 watt lower. Crazy? I agree. Last ramp test was one where I didnโ€™t increase at all, but I felt that I tested badly, so I bumped up my FTP 5 watts 172 from tested 167 result) and had a tough 4 weeks of build, but all-in-all, felt that it was a good choice. Ramp test scheduled for this week and I used the FTP detection and it has me at 172. Coincidence? I donโ€™t care. If I had tested and come in at less than 172 again, I wouldnโ€™t have decreased it as I know I was improving. To have the FTP validated gave me wings on my workout today. The placebo effect is grand! Thanks TR. Loved this new iteration!

I was still in the โ€œprocessingโ€ queue this morning so I tried something new;

  1. deleted the ramp test from yesterday that I created
  2. deleted todayโ€™s workout
  3. added a ramp test today
  4. clicked the estimate button

Then, after a few minutes, I got the estimation:

I got mine to work! I just wanted to see what it was predicting at the moment and itโ€™s saying 301 > 311 which is about what I was expecting/hoping to see by the end of SSBLV2. Doing this towards the end of the block is super motivational to me to stay strong for the last week and a half so I can bump up the number a tiny bit. Stoked for this feature and canโ€™t wait to see how it unfolds!

This combined with AT seems perfect for my life currently. With a baby at home my training was all over the place last season. Iโ€™d have hit or miss ramp tests depending on my levels of sleep and my fitness yo-yoed constantly due to the fatigue building up. Canโ€™t wait for the build block using this feature with AT!

Anyone else want to use the predicted FTP as a carrot during a ramp test to see if you can beat it? Thats how I beat my estimate (would proabably have tested lower if I didnโ€™t see that prediction)