I don't get PowerMatch

And to add to that I thought Powermatch was supposed to control the trainer via my stages, so why is it if my stages reads 145w the trainer itself still reads 10w lower? I get that it’s not warmed up, I get that it’s a different power source, but shouldn’t Powermatch be compensating for exactly that when it does it’s match every 10 seconds, and leading both devices to sync up?

With regards to hitting your target power for the interval, I’d say just make sure you go into the interval with your desired cadence and maintain it. Don’t go chasing the power, because you’ll just end up wrestling with ERG.

Also, keep in mind this…just because your power looks pretty, doesn’t always mean it is. I’ve seen a bunch of folks post trainer rides that have power that follows the workout profile perfectly. It just doesn’t happen.

I’ve tried PowerMatch with a one-sided Stages (and a Tacx Neo 2) as well. I don’t think it worked well – the resistance in erg mode fluctuated enough that the demanded power output wandered +/-10% from the actual target. This wasn’t just noise. Periodically doing chunks of the interval at 10% harder made the interval significantly tougher. Disabling PowerMatch was an immediate improvement in the stability of the erg resistance.

Oddly, I noticed bad resistance behavior with Manual PowerMatch as well. The resistance change coming in to an interval was hard and fast. Even upping cadence to ~120 just before the interval started, I’d come to nearly a standstill and fight to get things going again at the start of just a sweet spot interval. Don’t see this problem with no PowerMatch.

For clarity, the Smoothing in TR only affects the numerical 3-digit displayed power value.

It does not affect the graphing in any way or at any time on the workout or finished ride data. It also has no impact on trainer response in ERG mode.

The only thing it does is adjust the time frame the current power data shows as a rolling average at the power value.


As others have said, if you see a given wattage from your stages, it’s likely the same number you’d see indoors or outdoors. This is the one that is most valuable to you for your riding? Forget what the trainer is telling you,because it doesn’t help you outside.

I also had a trainer v PM mismatch with my Kickr v stages (L only). I can say it was more than 10w and that some of it was down to the specs (accuracy) of each power source, whilst some of the the error was down to slight LR leg imbalance - stages was taking my slightly Stronger L leg power and doubling it.

1 Like

If this is the case, then my bad.

I had assumed it also helped to plot a slightly prettier picture on the graph… So that although in the background the power is up and down like a fiddler’s elbow, but on screen it looks nice and tidy.

Think your understanding is correct, my understanding is when you use power match your Smart trainer reported power isn’t used by anything. I believe Trainer Road takes your external power meter reading (from your stages in your case) and adjusts the resistance of your smart trainer until you hit the target power with your stages. Therefore a side effect of this for me would be that warming up of trainers is not a concern as the resistance changes required take care of that.

Thus as mentioned previously, never chase the power, concentrate on keeping a smooth steady cadence and everything should slot in. Any increase/decrease of cadence constantly will make the whole situation suffer. My Quarq with a H1 feels almost correct after a few seconds. Ten seconds and double doesn’t sound right.

Obviously different smart trainers take longer or shorter to do this and you become accustomed to your setup and can pre-empt changes in power and any actions you need to do with your cadence from that point onwards. I myself increase cadence 1-2 seconds for sprints/big changes upwards in power requirements. I then keep this cadence steady for the interval so it can settle…I also slow my cadence a second or two before a low power recovery interval so I don’t “over spin” (which results in lower power for several seconds) and when I feel it come in, then keep a steady cadence for it to settle.

10 seconds sounds a lot to wait for it to adjust, some intervals are over by then. Is the 10 seconds with you maintaining a steady cadence through out the interval power change? I did have this problem once, when Tainer Road had done some power matching tweaking and you could see this sine wave at the start of intervals slowly decrease as it searched to settle, they fixed it.

If you still have issues with steady power during a change and its taken a long time to settle I’d drop support an email, they can look at the data we can’t see and advise better.

1 Like

Powermatch adjusts the demand to the trainer so that the power measured at your power meter is what the workout wants you to do. So if your kickr is usually 10W lower, and TR wants you to do 200W, the stages should read 200W and the kickr 190W.

The problem is that most trainers and power meters don’t have a consistant linear offset like that. The kickr is not always 10W lower, it might be 10W lower at 100W, 5W at 200W, 0W at 250W, and read 20W higher at 400W. So even while not using powermatch might make it look like your hitting your intervals spot on, you might not actually be working at 125% FTP, because the power curve is not linear.

I don’t know if better trainers are better at that (you would assume so), so maybe there is less need for using power match now.

So what I’m going to do is try to switch power smoothing off (and change zwift to report instant power and not 3sec avg) to see how it affects that power number more specifically how it tracks. The problem I have now is that number always moves slow which maybe makes me think that I’m missing the interval, because if it’s running on a 5 second average then it’s basically 5 seconds behind, even if with Powermatch the erg resistance is quick, that number lags behind.

I just want to make sure I’m getting the most quality out of my workouts, so if using Powermatch leads to a bunch of hunting for target power, and spikes in power, I don’t think that’s beneficial to a good workout. If I have to do say McAdie and end up overshooting the overs by 10% for 5-10 seconds for each one, that’s a. Putting me outside of target power and b. Leading to extra fatigue for the following over unders. Overshooting power for let’s say 10 seconds, 4 times in a set, for 4 sets, it adds up

you might be overthinking it. I’m not aware of any science that has concluded picture perfect intervals are “best” for training results. Your intervals look fine to my eyes, after reading your “half assed and poor results” comment (LOL) was seriously expecting to see 250W instead of nearly hitting the 470W target. You were within 5% of an anaerobic target, I’d say if you want to maximize results on that type of workout you should turn off erg mode and really do a max effort.


I mean to be fair, with Powermatch I hit 458, 438, 430 in targets of 478. With H2 I hit 458 and 455. Maybe I’m overthinking it. Maybe I’m not riding right. Maybe a combination of the above. Also been studying for securities exams for work so my mind is a bit mush for comprehending anything also.

Really wish there was a quick 5-10 minute vid from the guys at TR, or even some third party user, that went over all this. I think it would be very helpful to a lot of folks

1 Like

You do seem to be very hung up on the details :joy:

Have you seen improvement during your TR career, or are you going backwards?

If you’ve improved, I’d say you’re doing something right.

Sprint 4 is the 430W interval, if you move the 30-second a bit you get this:
452W - 35:01-35:31
455W - 35:02-35:32
446W - 35:03-35:30

and I just read the goals for that workout (Crane +1):

Since each of these short, race-priming efforts targets increased muscle activation, make each of them matter by perfecting your ability to hammer yourself without breaking yourself.

You’ll know you’ve gone too hard when you can’t survive the post-sprint minute, and you might need to increase the Workout Intensity if you find yourself sailing through those final minutes.

So for that particular workout its not about doing a max 30-sec effort, its doing a hard 30-sec start followed by 1-minute of sweet spot. In the context of those goals, I’d leave erg mode on and use power match.


Based on this statement, are you attempting to adjust power on screen via your input on the pedals?

If so, you should not be doing that in ERG mode. This leads to my primary statements for ERG users…

  1. IGNORE POWER NUMBERS when using ERG mode.

  2. Stop watching / chasing a power number (even if you think you’re not adjusting to hit power, I bet most people watching power, are subconsciously changing their cadence because of power fluctuations).

  3. Focus only on your cadence and hold that as steady as you can at the cadence you want. Doing this will minimize the fluctuation you put into the system, that it in turn, tries to adjust.

Every moment you increase or decrease cadence, leads to a signal in the app and then to the trainer to increase or decrease resistance in order to try and hit the Target Power. Adding PowerMatch into the chain of calculations add inevitable delay and complexity, that can lead to more erratic trainer behavior and power control.

The rider is the key to all this (PowerMatch or not) because they are the primary input to the system via the cadence and the related power against the current resistance. Any variation will lead to ERG instability.

In short, make sure you are pedaling as smooth as possible and stop watching power in the workout. Let ERG via the trainer and app do their job while you do yours.


Went thru the same concern after buying a trainer and signing up for TrainerRoad. It is natural to want to optimize results, and wonder if PowerMatch is a compromise or not. Our bodies are a bit unpredictable, training that works wonders one season may not work as well the next season. If you read studies it is interesting to note that not everyone responds.

I’m not aware of any credible evidence that proves exact wattages for precise timeframes are better than what you achieve by doing the same work outside or using PowerMatch. Do the work, follow the plan, and it should work out. If it doesn’t you probably need to change things up a bit.

Good luck on the securities exams!

I think this was touched on in the other posts - but since you’re introducing a new component you see a bit more lag on intervals (both at the start and the end).

For instance in crane +1 your interval for Sprint 3 - for that official 30 seconds you hit 438/478 watts - but if I shift the start and stop back one second (18:01-18:31 instead of 18:00-18:30) you then hit an average power of 478 watts.

So really - even with your desire to exactly hit the numbers - you are still hitting the numbers - they just don’t align with the prescribed interval gaps exactly due to the delay

In other words - it seems like you’re fine


Thanks for the explanations guys, and yes I do think some of my OCD is playing into effect here. I just did Perkins -1 with Powermatch and then mirrored on Zwift using the Hammer. TR avg power was 177, Zwift was 174, for a 2 hour workout, so not far off, especially considering they don’t track close together the first 10-20 minutes, but then seem to equalize. I turned off power smoothing completely, left power match in auto, and just tried to hold a steady cadence. The POWER field did jump around a bit but most of the intervals I hit spot on, with a few being off 1-3w

I have two ride files for the same ride, one from TR using power match and my stages, one from Zwift that read the H2 power. Can I email these to someone to analyze them? I can’t find a way to upload the files to here.

Here’s a link to the ride from TR using Stages

And here’s a link to the Zwift file so this data is from the H2 only

1 Like

It won’t sync up. I sometimes record my Kickr with my Garmin to see where it’s power level is set to, and to see if there’s a consistent delta with my power meter. Long story short, there isn’t. It varies too much day by day, even hour by hour, to be usable…but that’s just my Kickr V1…a solid trainer with power match, just don’t use the built in power meter or the calculated power on the V1.

I just wish some company would build a “high end” direct drive trainer that requires an external power meter instead of all the built in stuff. Sync ANT+ power and cadence and rebroadcast on BLT. I would think that should cut the build cost significantly and remove one of the big annoyances some people have.

1 Like

Sure, but what are you trying to figure out?

Just eyeballing it, it looks like your trainer and your power meter are only different by about 2%. That’s on par with the accuracy of a typical power meter. If I were you, I would just use your smart trainer inside, no PowerMatch, FTP tested and set based on the power numbers from the trainer. Your numbers from outdoor rides will be a little off, but that error is at roughly the same level as the inaccuracy of the power meter, so no big deal.

Been using PM for a few weeks and I like it now, definitely makes me focus a bit more on being smooth, and I feel like I’ll get better consistency using a more reliable power source.