The one thing that I still find frustrating with a dumb trainer is that I routinely “fail” short intervals (5 seconds, for example). For example, I recently did Ebbets and “failed” the sprints during the last interval. It turns out though, if I just move the interval by one second, I didn’t fail them.
With interval search, I can now just search within the workout (let’s say by the minimum power for the lowest sprint interval and 5 seconds duration) and I get a list of all the 5 second-sprints. Then I can compare to the intervals.
What would be super cool would be if TR could do that work automatically by applying a certain fuzziness to the intervals. So if in the workout summary it lists Sprint 12 as 350/375 or whatever, it could just look 1-2 seconds to the right and see if that changes the power upwards. Does that make sense?
That totally makes sense and in my opinion, this sounds like a great idea. It’s tough to nail down short intervals like this on a dumb trainer, even though you’re still doing the required work you’re off by a few seconds. I’ve passed your suggestion on to our product management team!
I tend to treat intervals as if they start ~5-7 seconds earlier than they do to get around this.
Coincidentally, there is no such lag when you let off the power (evil cyclery god’s I’m looking at you). So all sprint intervals tend to get an extra ~5 sec of duration tacked on.
Obviously this is tricky with shorter intervals, especially ~30 sec and under, where extending the intervals ~5 sec is actually a significant amount of extra training load.
Hi @larry would it be possible to add to the search an interval range. For now you can search time spent above x Watts, but would be nice to search it in range mode as well.
@larry also, on the time frame, although it says time longer than X when you search it limits the interval for the specific time. Is it possible to expand this as well?
@larry@Bryce
This feature request would be very handy. Some additional idea:
I know you’re not too much into HR (at least as far as live data is concerned) but could you apply this also for the HR average of the intervals in the interval list? So looking up to 30 seconds (for very short intervals you might take less seconds) to the right and showing the max average HR for the interval duration?
HR is deferred and in the graph you always see those HR waves deferred to the power spikesof the intervals.
With an interval of 2 min my average HR of 0:00 - 2:00 is not as interesting to me as the average HR of 0:15 - 2:15 for example (or maybe just the last part like 1:00 - 2:15)…
If the code for this feature request for power is in place, the same code/principle could be used for HR.
@Nate_Pearson Am I making sense here? Which part of the HR graph do you look at for intervals? Wouldn’t you like to see the matching deferred average HR for an interval in the interval list?
Hey @Rizzi! Thanks for your feedback! You’re right, we’re not big on HR and I wouldn’t be giving you a fair response if I told you that we’d look into doing this. For the moment, there’s no code in place to support this feature request on the power side, but after speaking to our app product manager, it sounds like we have something in mind that will take care of this and is on the road map. No such luck on the HR side, at least how things stand right now.
If a work interval is N seconds long, TR should report our power for that work interval as our max power over any N-second interval that starts within a couple seconds of the right time (at least if we are finishing the full interval).
Sometimes with Erg mode, we can start and end intervals a couple seconds late or early. With short work intervals, this leads TR to report that we’ve significantly undershot the target power when we haven’t (we just did the work interval slightly later/earlier than prescribed). See for example this thread: I don't get PowerMatch - #33 by bbarrera
I’m hoping someone can help me out here or if not, open a feature request.
Setup - Bkool Smart Pro on ERG Mode with ANT+ connected to the laptop.
After doing Pierce I fancied a dive into the data to compare various intervals within the workout, I did so and noticed a lag in every interval. I’m using Sprint 27 below just for reference but it’s apparent in all of them.
You’ll notice that the “actual” and “power” lines are offset almost the same amount opposite sides of the target power. I don’t know what’s going on with the “actual” line but i’m nailing the interval target on my “power” line it’s just out with TRs constraints for the interval and is therefore chopping off some of the effort.
I know i’m getting the work done I just think it’ll be a bit of pain looking back at the overview of historical intervals when it says i’m 10 watts below the target. It would be good to be able to program in the lag so that it auto-corrects after the workout is complete or be able to drag the target power blocks for best fit of the “power” lines.
Hopefully this makes sense to someone, it ended up a bit of a ramble.
