Huge discrepancy 4iii precision 3 power vs Wahoo Kickr 2018 Edition KICKR

First post here. I’ve been getting back into using my indoor trainer for some time efficiency reasons and have noticed a massive discrepancy between the power trainerroad has me at vs what my power meter has.

I recently did a 1.5 hour sweespot workout where the prescription was 307 watts for 5x9 for (FTP is 323 and that was found using the same wahoo kickr and a trainerroad ramp test). I noticed the warmup 310, which was 3 mins, felt significantly harder than it should have been. I figured it was fatigue so i lowered difficulty by 7% so each interval would be 285 watts. I did the first two and my HR was closing in on threshold levels. I decided to setup my garmin and 4iii to run an indoor workout at the same time and see what the differences were. My avg for the next 9 min interval on my 4iii was 323 watts, an almost 40 watt discrepancy.

Now I am normally a skeptic when it comes to my fitness, but I started doing some math on some very common strava segments that I have done repeated times outside with this powermeter. I recently did an effort on this segment which is 1.55 miles long and avgs a 5% grade in 6:50 seconds, which comes out to 13.6 mph avg and a 358 watt avg. I am 188 lbs, 17 lb bike, and give or take 8 lbs of gear including water, shoes, repair kits etc. I used this calculator Cycling Wattage Calculator, which again I understand is a rough estimate, but the math seems fairly sound and it comes out to 348.5 watt average to maintain that speed (with a -5 tailwind, I don’t think it was this much, but wanted to give benefit of the doubt). I did this for about 5 other attempts on this segment and some other much longer climbs i’ve done and compared using that calculator, and all of them came out to +/- 2% to what the 4iii recorded.

Anyone have any idea how/why the kickr and trainerroad are so far off?

Edit:

1. Kickr Spindown / calibration on trainerroad was completed after a 10 min off workout warmup
2. 4iii sensor was calibrated prior to measuring, as well as recalibrated between intervals.

Aside from the interval analysis, you may want to do a fuller comparison with one of the tools shared here:

Then some basic stuff to cover:

• Is your 4iiii power meter properly calibrated (zero offset)?
• Is your 4iiii power meter a single-sided or dual-sided? If single, realize that is a notable variable when compared to the Kickr which is “Total Power” vs “Single-Sided Doubled”.
• Is your Kickr properly calibrated (spindown)? I think this should auto calibrate, but it may be necessary to take the extra steps in a case like this.
• Recognize the difference in locations measured via crank arm and post-hub (trainer). This introduces drivetrain efficiency into the variables.

I’d personally avoid using the calculators you used as there are likely variables at play (wind, rolling resistance, etc.) that make split comparisons rather flawed.

Your best bet is to do a set of data collection with the bike & power meter on the trainer (all proper calibration performed) and capture ride files for detailed comparison.

1 Like

1. Yes
2. single sided. Understood, but a 13% difference in reading between those two numbers seems outside of the range of normal, imo.
3. Yes, done using the calibration inside trainer road
4. Any math on the efficiency differences here?

As for the calculator, i don’t disagree entirely, but it does seem to match very closely to the power i have from my 4iiii across a number of segments in different locations on different days. If you dig into the math, variables they introduce into their formula, and assumptions they introduce, then I think there are some reasonable conclusions to be deduced here considering the numbers it outputs are so close to the numbers the 4iiii produces.

1. The question is necessary even if it is not a single smoking gun for your issue here. Single-sided vs total power here may not be the difference, but it’s impossible to ignore it either. We have seen enough deltas in L/R reporting to at least include it as a factor.

1. I don’t have a ready resource without some Googel-Fu, but it seems to range from low single-digit percentage to larger for dirty a/o worn drivetrains. Again, not likely the single issue here, but it has to be added to the list as a known factor.

On the calculator, it is fine to use for an interesting data point, but nothing will beats a 1:1 comparison between your two devices at the very same moment. That is the place real time & effort needs to be spent.

If you generate some coupled ride test data, I’d then follow up with each component maker so they can help you trouble shoot this. There isn’t any easier way to deal with this that I know, especially if one (or both?) of them is giving errant data.

On the discrepancy you are seeing: easily ~1/2 could be your inherent left - right power balance. Not a perfect example but my Quarq regularly shows me at 52 - 48% left - right power balance (yes, I know this is an estimate) which would be either 4% high or low to “true” power depending upon which side was doubled.

Drive train losses is 1 to 2% for a clean, waxed chain to up to 10% or more for a gunked up chain.

1 Like

I have a gross left/right imbalance, and I can see differences that large when using left only device.

I’ve had rides where my left/right is as much as 44/56 which means that doubling the 56 side would inflate the numbers 12% which would be 38-39 watts different on that interval.

I’d find a known good powermeter, test against that, and get a left/right balance checked.

This could let you know what device if any is faulty, and let you know if this is a matter of an imbalance that needs a scale factor.

Keep in mind, imbalances can change over the power range and based on muscles being used.

When I am aero and using glutes, low intensity, massive difference.

When I am out of the saddle, pushing hard, numbers look identical.

Good luck troubleshooting

Thanks for the input, chain is religiously cleaned and greased weekly if not multiple times a week. I am ok with a 1-2% difference. 13% makes it hard to follow a training plan.

Interesting for sure. I am not sure I have the same imbalance. For the volume I put in, if I had this severe of an imbalance, I feel like I would have more injuries and fatigue in one side. Plus, the fatigue in on leg would be more significant during my 90min-120min sessions on the trainer, which doesn’t occur. I do agree the only true way to know is to get a dual sided powermeter.

Lots of assumptions there lol.

I am decently high volume, and currently race in the 1/2 field and am a handful of points from getting my cat 1 and don’t really get injured on the bike, and fatigue has never felt any ddifferent between the legs.

As I said only one way to test and know, I went ignorant for years, I had no clue til I was troubleshooting a similar issue, then got a power meter with left/right and it confirmed my suspicions.

1 Like

Most definitely, and certainly wasn’t trying to assume anything negative. All interesting points, going to have to find some ways to test this out efficiently. Good luck with that cat 1 promotion!

Last simple idea: have you verified that the 4iii is setup with the correct crank length?

If so, then my only suggestion (unless you can borrow power meter pedals from someone) would be to mimic DCRainmaker’s / GPLama’s power meter test protocols and dual record your 4iii and Wahoo Kickr and then use something like DCRainmaker’s Analyzer tool to compare the outputs side by side. Given you have a single sided power meter, I would compare:

• Easy endurance pace at your normal / self selected cadence
• Sweet Spot / Threshold intervals at your normal / self selected cadence
• Sweet Spot / Threshold intervals at higher than your normal / self selected cadence
• Sweet Spot / Threshold intervals at your at lower than your normal / self selected cadence

What the above will help spot is if your single sided 4iii and wahoo kickr are consistently off by more than 1 - 3%, or only under certain circumstances. If the divergences gets larger under certain circumstances (e.g., higher than normal / lower than normal cadence) then part of what you are seeing is do to the 4iii doubling power from one side where you (like everyone) aren’t 100% symmetric in your pedaling.

2 Likes

Nothing negative assume, just trying to help alleviate some possible headache long-term from my own experience with similar issues

It’s great to come up with ideas, but even better to verify

And thanks, hoping for this to be the year

It’s not ideal but I have the same problem and similar huge discrepancy between my 4iiii/kickr indoor power and my quarq that I use on my MTB where I do most of my riding and training.

Part of my discrepancy I actually put down to my position on the road bike which I use indoors. I feel far less powerful in that position, my quads don’t feel activated the same way I do on my MTB.

Anyway as much as it stank I use two different FTPS, one indoor one outdoor. I didn’t really use tss as a major guide so it didn’t muck up too much on training peaks. Can’t guarantee that for my coach though

Can the scaling function on the 4iiii app for right/left imbalance be used to more closely match your trainer as @dfquigley suggests?