There is a serious discrepancy between the old and new forecast. On Sep 16, I started AI forecast plan starting at TP 257 W and FP (10 min) of 296 W. It predicted to increase TP to 271 W and FP to 317 W. From the beginning, there was a one week break planned, which happened to be last week. The UI asked me to adapt the forecast this morning when I came back which I did, and it still predicts the FP to increase to 313W. So a bit lower originally, but all is good.
Now I was curious about the new release. I was playing with the new UI and using the same settings, FP power increase beyond 8 W (i.e. to final number of 287 W) says that I likely won’t be able to meet the goal. At the same time, it says that this will result in my final TP of 246W.
In summary, the new UI/forecast says that over the next 2 months of training, my TP and FP will be lower by 9 W and 11 W, respectively, than when I started the program on Sep 16. Something seems seriously wrong here. Screenshots attached.
Which tab do you select to show the top image you posted?
What happens if you change from Mixed to Polar then see how far you can push Power Increase slider to the right and remain achievable?
271/317 isn’t the final target. It’s just he end of the dials. If you move the slider, you’ll see where it ends up.
Also you have 2 months to double your High Training Load which is identified as a yellow warning and might be a bit of a stretch to reach. You can now create a plan and still save it, even if the goal isn’t reachable by the forecaster.
@ridgerider2 the top image is from “Program” and then selecting the “Start” button in the top right area. Switching to polar allows me to go to 12W increase
@xertedbrain Just to be clear, I am not saying whether the old or the new forecast is correct, just that they are inconsistent. I moved the slider and 271/317 indeed seem to be the final target, see screenshot below. Where can I see the yellow warning? When I was originally planning the training in September, it gave me no warning and that is the part that bothers me.
Edit: I think the following best highlights that something is strange. My TP was 250W a week ago, over the week off it dropped to 240W (with small decay setting too) and the new UI forecast tells me that the best I can do in the next two months is to increase it by 6 W. That makes very little sense.
The yellow warning refers to the color of the Achievability text. Examples:
In your case the yellow warning refers to the more than doubling of High strain required.
I don’t think you can easily compare a forecast issued weeks ago with one generated now since conditions have changed. The process is constantly evolving, and the prediction assumes you complete the plan as forecast which is highly unlikely (never need to Adapt Forecast).
I prefer XATA options with a blank calendar but when I tested Goals during the Beta I experienced both a decline situation when I wasn’t keeping up with the plan and on a second test, I exceeded the target Goal prior to Target Date by upping my hours.
On my first test I felt I was getting stretched and tapered back. On the second test I adjusted Recovery Demands and responded well enough to add more hours.
In this respect your goal is a moving target depending on how closely you follow the plan and react to it.
If you feel your signature is understated at the moment, try one of the Fitness Test workouts in Slope mode on a HIT day and see how you do. Then add any remaining training recommended for the day.
I see your points and I understand that the forecast is a dynamic process. But we are talking about 39 W (original) vs 13 W (current) discrepancy in FP. This is 300% difference. It feels like the current prediction only sees my current training load (after the 1 week break, which tanked it to ~66) and does take into account a longer training history (I was previously consistently at ~82-88).
Then there is the elephant in the room - a prediction that in 2 months my TP will be lower than when I started the plan 2 weeks ago.
I have been with Xert for 1 year now. But to be honest, these recent numbers feel like random numbers that make me question the utility of the entire forecast. I wish I could understand what is going on here.
We’d have to dig into your data more to understand why this occurred. It possible the new algorithm isn’t as liberal about assigning training as the previous since we’ve made some improvements to how it determines what training is possible based on each user’s history and settings.
There are never any “random” numbers, unless there is a computing error. That can happen. Often though it’s more of a matter of digging more deeply into why the AI algorithm chose what it did and often there is an explanation. 2 weeks off can make your fitness tank really quickly and it takes far longer than 2 weeks to build it back so we’d have to look at your data to be able to answer your questions.
Have you send this into support yet so we can have a look? Can’t post your private data here publicly to properly answer your concerns.
