Indoor vs. Outdoor Bike Performance

The following is a question from a coach. He and the athlete he coaches are somewhat frustrated by the difference inpower numbers they produce relative to heart rate when riding indoors as opposed to out on the road. This is a common complaint this time of year, one that I hear a lot.

--------------------------------------------------------------------------

QUESTION FROM THE COACH: I'm hoping to find an answer or an explanation to an athlete's FTP [functional threshold power] testing results.

Here's the 411.... the cyclist is a a masters racer... 44 years old, has been cycling for the last 6 years. Whenever he conducts a FTP test outside he is posting better 20 minute numbers vs a test on a trainer.  I looked at HR and that piece of the data is the same.  We have tried a multitude of things to improve his "trainer" FTP (to match his outside FTP), such as fans, using the trainer outside, music.... but there still seems to be a difference.  His prior coach actually had two FTP values for him depending on if he was doing a trainer workout or an outside workout.

Here are the numbers from a past test. I had him do a test on a trainer resulting in 258W and an average HR of 174.  He was frustrated at his low number so 10 minutes later he went outside and put down a 278W with 175 HR. 

--------------------------------------------------------------------------

MY ANSWER: This is a common question every winter. A lot of riders experience this. But I’ve also heard from some who have exactly the opposite problem—better numbers indoors than out. So it remains a mystery. There are a couple of possibilities that may explain it, but they are rather lame I’ll have to admit. (I'm assuming the same power meter and bike were used on both tests.

One is that when riding on the road there are brief episodes of “recovery” due to slight hills, cornering and changes in wind direction. These little changes don’t occur indoors on a trainer where power is usually much steadier. Some research shows that these little “mini-recoveries” allow the rider to go harder overall. In fact, I’m told that when going for an hour record on the track they sometimes train to be “unsteady” in power output order to take advantage of this little oddity.

The next is that when riding on the road the bike actually rocks from side to side slightly while the body remains relatively steady over it. On the trainer it’s the opposite – the bike is locked in place so the rider’s body moves slightly side to side while pedaling. This requires a somewhat different technique and muscle recruitment and so may elevate HR and effort “artificially.”

It sounds like you have already eliminated the third one – that heat build-up elevates HR when riding indoors more so than when on the road. Lots of fans help here.

I’d also be interested in seeing if the rider paced the two TTs the same way. In other words, did he negative or positive split the two trials in the same manner? Someone may be more adapted to riding on the road than the trainer and be much more unsure of how to pace themselves indoors. Or the other way around.

Could one of these explain a 7% difference? Probably not. Could some combination of them account for that? Possibly.

Bottom line is that I really don’t know.

Comments

[Chad]

For my n=1 experience, I did a 20 min test on the trainer for the first time ever a couple weeks ago, with an avg power of 263 watts. I was outside on my balcony and it was 28 deg F, so cooling was not an issue. One thing that I noticed was that I was forced to pedal at a lower cadence than I wanted to (around 94 rpm). I would prefer to be and I naturally settle in around 100 rpm for a hard effort, but if I shifted to an easier gear, it put me around 104 rpm which was too high for me. This definitely led to what I felt like was a sub-optimal test. I have not done a 20 min test outdoors since September, but my 20 min power then was 328 watts, and I'm positive I could do at least 295-300 watts outdoors around the time that I did the test on the trainer.

Another thing that I am curious about is if using a rear wheel based power meter such as a PowerTap affects measurement on the trainer. I have a PT and just bought a Quarq (I'm going to sell the PT). The other day, I did a 2.5 hr trainer ride with both the PT and Quarq on my bike, and the Quarq read around 10% higher on average (181 vs. 165 watts). To me, this seems a little high...I was expecting the Quarq to be maybe 5% high due to drivetrain losses. What if the PT measurement is somehow affected by the rear wheel being pressed up against the trainer cylinder? Later this week, I'll ride outdoors and see what the % difference is between the Quarq and PT...if the % difference is different for outdoors vs. indoors on the trainer, then that might explain something...

[Tristan]

Well, the engineer in me says this rider was somehow able to use 21kJ more energy outside than inside for the same perceived (and measured - HR more or less the same) effort.

I would put this 21kJ down to many small things all adding up:

1. He is gaining efficiency by using a higher torque and thus cadence and for whatever reason this is suiting his physiology - his body is able to use more energy this way rather than the higher cadence method.

2. Outdoors there may be more spikes and recoveries in the data. The other day I was doing a HOP workout (53x12 all out spikes every 3 mins for 30 sec) and beat my previous FTP by 10W. Somehow it was easier doing it this way than a steady build, which was my previous FTP test. I spiked more and rested more, I think my body likes this rather than 60 mins of turning the screws. I go with the micro-recovery theory.

3. Also, he was MOTIVATED to prove the indoor number wrong

4. It's outdoors! The cooling effect is much stronger. Heat dissipation is actually the single most ineffcient thing in our bodies. The body is actually putting out more energy than the energy that is recorded at the back wheel. The delta is lost due to heat build-up, stabilizing the core, working our lungs etc..

Question: Did he do both FTP efforts back to back? It's not uncommon for the second effort to be better because you have already warmed up, primed the lactate buffers etc. Thus the RPE is less for moreater output...or the output is greater for the same RPE

[Ponch]

In my experience, riding indoors with rollers has a lower PE (and higher resultant power) than riding indoors with a trainer.

It may be because riding rollers more closely matches the balance and muscle use of riding outside. (the rear skewer is not fixed).

[Duanegran]

Looking at this another way, does it really matter? Some people are able to post the same numbers indoors (or up hill, or on flats, etc) and others post different numbers. From a training standpoint can one say that they are equivalent? If you ride 20w lower in doors but once you go outside you still ride +20w then maybe you have two numbers instead of one.

[Gerry Gucher]

Hi, I am a fun cyclist doing occasional hobby triathlons and ride around 3000km/year. I also have quite a bit different FTP, carefully determined by the exact same protocol(monod-scherrer 5min/20min with 15min rest in between) using the same Powertap.
Outdoor flat - 229W, Outdoor uphill (6%) 240W, Indoor 219W (using single fan, open window).
I would think there is no single reason but an accumulation of little differences, i.e. a single fan can never cool as much as outdoor headwind, rolling momentum, psychology, slight difference in cadence etc. etc.
I even have a 7% different FTP between flat and uphill (uphill being higher). I can never hold the same top power cadence uphill (~80) vs. flat (~98).
For me it's pretty clear, depending on personal abilities a watt is not just a watt. I simply use different power zones indoor/outdoor/flat/hill when I do a systematic power based training.

[Brendan]

My ftp numbers were higher on a kurt kinetic rock and roll vs the static road pro. It seems to have something to do with the bike moving and all the force not going through the body hence fatigue and rpe being impacted. Interestingly the ftp on the r&r is identical to my road number whereas the static machine was 10% lower with rpe higher. I guess these things can be quite atlete specific.

[Randall Maddox]

I did not have time to read every comment, but I imagine the difference has to do with INERTIA. Studied physics 43 years ago but I remember 'that if a body is in motion, it has a tendency to stay in motion.' (or something like that?). You have inertia working for you on the road and do not a fixed device. Well maybe the wheel has little? It is turning.

It also may have to do with the difference in what you are overcoming. Air drag on the road and wheel drag on the trainer?

I feel stronger about the inertia idea. We might consult someone who got past Physics 101? I am an Economics major, as you probably remember. So I would be better determining the TIME/Value of indoor training.

[Mark]

Lots of comment on why, but very little on what to do about it. Here's my thought...
At 259W the zones are:
Z1 0 142.45
Z2 145 194.25
Z3 197 233.1
Z4 236 271.95
Z5 275 310.8
Z6 313 388.5
Z7 388.5 +

At 278W the zones are:
Z1 0 152.9
Z2 156 208.5
Z3 211 250.2
Z4 253 291.9
Z5 295 333.6
Z6 336 417
Z7 417 +

If the intent is ensure the athlete is training in the proper zone.. indoor or outdoor, have him train in the overlapping portion of two sets of zones... specifically:
Z1 0 142
Z2 155 194
Z3 211 233
Z4 252 272
Z5 294 311
Z6 336 389
Z7 417 +
Wouldn't this ensure he gets the aerobic or anaerobic benefit of his power-based training regardless of whether he's training indoor or outdoor? I get there's gaps between the zones, but that's easily manageable in actual training, and wouldn't this be better than trying to remember a different set of zones depending on whether your training was occuring indoor or outdoor?
(and yes, it's winter again so the discussion on this topic is popping up again!)
Mark

[Bruce]

I have been curious about the discrepencies of training indoors and outdoors. Yesterday I did a little test. After a 45 minute warm-up I rode up a 7% grade at Sub-LT power and it felt comfortable. I then rode on a flat at the same power and it felt much more challenging. Then I compared data from yesterday to indoor rides of similar power and found that in all situations, the torque was about the same for the wattage. As a matter of fact, most measurements were similar. The only difference was RPE. I have always thought I am more suited to climbing, I find it easier than pushing on the flats, but now I am coming to the conclusion that the biggest variable, perhaps the only significant variable, is psychological. You hate TT, they will be challenging, hate hills and they grow steeper and longer. Hate indoor training, it will show.

[CadenceCoaching]

There are a couple of things I have found about the trainer that cause issues for me. First of all is that it is unnatural. The bike does not natural track in a straight line. The bike torques ever so slightly underneath the rider as they crank on the pedals. More efficient cyclists have less of an issue with this as they have learned to be more smooth. This could account for some higher power numbers. For lower power numbers there is the mental issue, I output less power than I have seen athletes whose power numbers are lower in headwinds, on false flats and on the trainer simply cause they feel like they should be moving faster. It defeats you mentally to hammer and move nowhere.

[Ezclimb]

For what it's worth, I'm curious as to the type of Power meter used. Is it possible that if the sensors in a crank (for instance) are affected by change in temperature from indoors and outdoors and the rider doesn't calibrate properly, this would have an effect on the reading?