Training Inside (structured) and Outside (unstructured) on Different Bikes/Power Meters

With the weather slowly getting nicer (hit 45 yesterday!) and a new bike coming this weekend I’m trying to get a plan for the spring/summer. Typically I just subscribe to a training program during the winter and cancel/pause it over the summer when I ride outside. I’m really liking TR and would like to keep my old bike on the trainer year around and continue to use it for the more structured rainy day ride.

However I’d like my outside rides to still “count” (I know it wont with AT (yet) but at least TSS (I think) would be nice to have). I don’t plan on doing training rides outside so they’d be unstructured. The bike would also have a different power meter (crank arm) vs the indoor one using power from the wheel on trainer so they’ll be “off”.

Suppose the third option is to just keep them separate and keep track of TSS in Strava. If AT (and AI FTP of course) ever comes for unstructured rides I’m wondering if the power meter difference will impact that.

Not sure if I’m missing something or overthinking but just wondering how you guys handle this as I’m sure I’m not the only one doing this.

Check the PM offset from each other by putting both on the trainer and record data from an erg session. Scale the power to match closely. Within wko or you can flag inside rides and either set separate ftps and/or power correction so the numbers match more closely

Put both power meters on the same bike. Dual record a workout on 2 different devices/computers and then compare the FIT files using a comparator like DC Rainmaker’s tool or Zwift Power’s comparison tool.

You can go second by second and compare the values. In all likelihood there will be some variances that are non-linear.

Any resources on the above? I’m not finding much out there.

But if I understand correctly…I put my new bike on the trainer with it’s power meter and do a ride. Get creative with some analysis using the above tools and figure out how much higher the power meter is than the trainer. Assuming it’s 5% I’d edit the trainer to read a 5% higher trainer and then it’s “equal”?

1 Like

Yup, basically. Some PMs let you go in and enter a manual scaling factor or offset. It may also just not be enough to really care. Like in the long run will it really matter if you look back and your 2 hour endurance rides say 95 instead of 100TSS? Probably not. I have 3 bikes with 3 different PMs (Road with a Stages L, Gravel with P2M and MTB with P2M) and I do structured stuff on all 3 and have never noticed or even attempted to measure a difference. They feel close enough so that’s good enough for me.

1 Like

So maybe I’m just overthinking this then (habit of mine) and it won’t be that big of a deal, especially until it starts being used for AT which my guess is will be smart enough to figure that difference out. I also stumbled across a thread that mentioned rides imported from Strava us their TSS so I could just have a different FTP on that and it’s set.

I do think it may be worth comparing them if you want to just to verify but I wouldn’t sweat a difference of <5%. Most PMs these days are pretty good and will most likely be pretty close.

I don’t think TR uses Strava’s TSS value. I’m fairly certain that TR will take the raw power data and recalculate TSS.

Luckily we’ve already solved the indoor/outdoor power meter problem in AI FTP Detection:

How to Use AI FTP Detection

1 Like

You would put the new bike on, or use your old bike and move the crankset over…or maybe just the crank arm if left sided.

This will give you 2 “power meters” on 1 bike, and then you can record on 2 gamin devices, for example, or a Garmin and trainerroad, or Zwift, or whatever. You set one device to record power meter #1, and another device to record power meter #2.

I forgot to mention, if the power from the wheel on trainer is virtual power, just forget all of this and use a power meter.

DCR Analyzer Tool Overview & Manual or, go to your profile->Analysis.

You could try to eyeball it, but the power averaging may not be the same, and you’ll get differing readouts, and end up chasing things.

The best is to do it after. The DCR tool, you can see in action with DCR or GPLama videos. Then you can take a close look at how the power if offset and if you can scale it closer.

Also, I think it is worth comparing. Power from a trainer will vary, by gearing. On my Neo 2T, the power gap can be 20+W to spot on, depending on which gear I’m using. So, I just forget all of this and use my power meter.


I assume when/if AT takes into account unstructured rides it will be the same?

If AI FTP is that smart, is TSS as well and I’m really worrying about nothing?

Good info and what I assumed was correct. Was trying to save myself the $50 a thru-axle adapter for my Kickr Snap costs and just not put the new bike on the trainer but I think my curiosity will get the best of me and I’ll end up doing it to get a baseline as suggested.