Defining failure within AT

Hello All, I just finished the below workout. Before AT I would have classified this as a fail. I was therefore surprised that I was served up the the PASS survey. I know I could have failed myself by selecting “All Out” as a survey answer but the fact that the system interpreted my result as a pass got me thinking…

I know the TR team has said on several occasions that the odd 10 second back pedal doesn’t constitute failure, but this workout shows (to me) a clear pattern of struggle with breaks coming with increasing frequency as the workout approaches its end.

Am I too hard on myself or is AT a bit of a softy? Any ideas on how much worse I would need to have performed to get this classified as a failure? Would I have had to skip an entire interval?

For starters, what did you actually rate the workout?

To your main question, this could be a borderline case, but I can see why it might still put it in the “win” column (depending on rating at least). You did a LOT of work there, and got what may well be the desired training stimulus of the intended workout.

I have been guilty of missing the forest for the trees, and fixating on low numbers or “cheats” like that. But we need to step back, consider the goal, and what we got from the effort. It’s possible you are good even with the shortcomings.


What was the difficulty level? (productive/stretch/breakthrough)

1 Like

Your question is part of why I asked mine. I’m wondering if I’m scaling failure correctly in my head. I would have rated it “all out” but my understanding is that would have “overruled” AT’s assessment that I had passed, so I rated it “very hard” instead.

I think your forest vs. trees analogy is the right one. I’m currious how many more trees I could have cut down before it said the forest was gone :wink:

1 Like

Productive. This workout is a 7.7 for Vo2. I was a 7.6 going in to the workout.

1 Like

I would define that as a struggle and not a failure too (you finished) but I can’t take any credit the AI in AT :wink:


Keeping in mind that my sheet is NOT officially endorsed by TR, but it is based upon much discussion and includes direct statements from TR, I would ask you to review this and consider your choice:

From your statements, I would think 5 - All Out would have been appropriate. This matters because it would offer the “Struggle” followup, and give you a chance to consider and answer ‘why’. That helps TR/AT frame the effort and plan going forward.

With the PL/WOL nearly matching, level of fighting you showed, and what I see as a 5 indicates that likely something was not ideal. As mentioned, even a 5 AO is NOT a “failure”. You hit a limit. What matters next is to understand why.

Considering that this was “Productive” and only a small increase from your existing PL, would you consider this a workout you were prepped and ready for or were you off for any reason (Sleep, stress, food, etc.)?

1 Like

On that scale, 5 is probably the better answer. I was psyched, prepped, ready etc. just missing a bit of power to make it a clean finish. I’ve given “all out” as an answer before and, in that occasion, AT massively adapted down the subsequent workouts. I just changed my survey answer to 5 and interestingly (for now) AT left the rest of my calendar as is.

Yeah, changing a workout rating after the initial one doesn’t always trigger a new round of adaptations. It’s possible another trigger, even waiting until tomorrow can lead to new adaptations.

That delay is one reason it’s worth it to even take a short moment and not rush your survey selection. Even on the ones I am fairly certain, I refer to my chart and double-check since I know that changing later does not always trigger what we expect. Best to get it right the first time generally speaking.

Good to know, much appreciated. Thanks!

1 Like

I think you should rate it all out not because of AT but because it was an all out effort for you


I think you are right!

Well I disagree! One thing to consider about TR position on ‘back pedal breaks’ like this is whether or not they are able to distinguish between these and technical disruptions in data flow. I suspect they can’t. So the least disruptive option is just to call all things that look like this a pass.

So I agree with you: this workout was harder than you could handle that day. I’m not going to hang a negative label on it but it is reasonable that you should give your next series of workouts a think. Or, you might just be the type of athlete that will plow right through workout after workout like this. I’d worry that such a practice might lead to dangerous overreach.

I think you’d be correct to rate this workout as all out.

Good point on the data gap safeguard. Thanks to the advice given here, I changed my answer to “all out” with “intensity” as my explanation. Now I just hope I don’t get too badly adapted down - I was looking forward to my next Vo2 workout on Saturday :wink:

  • Disagree, at least in some cases.

  • Pure data loss is one thing, data = 0 is another. I can see the difference right in the graph of my complete workout each and every time I get a true data drop vs me stopping pedaling for any reason.

  • Add in the fact that some devices will report cadence (even when reversed) which is a sign of a BP despite 0w.

  • No idea what exactly they are or are not watching for, but I think it’s not necessary to assume they are ‘blind’ to the reality of the situation. Again, case dependent, and some tell a very complete story.

1 Like

@Syd_Nakter you have the hand on the tiller of your own Training Ship…AT is just the navigator. If AT suggests a workout you know is much easier than what you can handle, well, that’s what they super-cool ‘Alternates’ button is for. (TR has you covered, see?)

I guess the main point is there was a workout on your plan that was so hard you had to stop pedalling. Not right at the end but somewhere in the middle & then several more times over the last half of the workout. If you keep doing workouts like that there is a risk that it will have a negative impact on your long term progress. Heck…medium term progress. I don’t think you can progress like that for long.

1 Like

fwiw I’ve had “fail” surveys on a “Very Hard” score, not just “All Out”. I would’ve expected a fail survey on that myself, and have - I have TR set to autopause, I wonder does it track pauses more than back pedal?

Yup, and if in fact he was actually “ready” for this and still struggled, it’s possible a downward adaptation is actually a GOOD thing. We know it sucks to hit those backsteps, but sometimes they are real and very necessary. Gotta keep perspective here too.

1 Like
  • Yes, it absolutely does track pauses now. They mentioned it at least twice on the cast.

It is possible for TR to see that and react accordingly. Pauses in low recoveries are handled differently than ones in the middle of work intervals. Count and location within are all checked.

@Macy the plot thickens :wink: very interesting @mcneese.chad @Brennus no worries I have the hand on the tiller and I’m not going to hurt myself, appreciate the concern though. Thanks. This place rocks!

1 Like