Evidence behind 'efficiency' of indoor training

I see a lot of references both through Trainerroad and other cycling websites about how there is greater efficiency with indoor training.

Statements can range from very broad statements like ‘it’s more bang for your buck’ or very specific things like ‘every hour on the trainer is worth 1.5 hours of outdoor riding’.

I was wondering if this is backed up by any sort of data? From my own experiences it is certainly true to some extent based on my own progress - but is this something someone has looked into from a scientific perspective at any stage?

I imagine that the best data to look at would be cadence/pedalling time data and averaged/normalised power comparisons between indoor and outdoor rides - but I can’t seem to find these comparisons anywhere.


Most people I know ride outside with their headunit on Auto-Pause, so I suspect the data isn’t being collected. Test it yourself - ride outdoors a few times with the Auto-Pause setting turned off and see how much time you spend at 0W. Time spent close to 0W is something you just don’t get with a trainer.


I’d argue that every person recording data on their rides is offering a “scientific perspective” and hence this position is backed up by plenty of data.

As @DuncanM23 said, it’s obvious that a lot of outdoor riding is going to include breaks in effort such as freewheeling and braking that you just don’t need to do on a turbo trainer.

The reason indoor training can be so effective is that you can do almost the exact amount of work at any given time. I know, for me, doing this outdoors in my area just wouldn’t be possible. There’s too much traffic, too many traffic lights and no perfectly consistent flat or climb

It’s not autopause you need to change, its “record zeros”. Autopause is GPS based so that would still record zeros as long as you’re moving.


Good question! I don’t have an answer, but I have a related question…

I have a power meter and HR strap. I do an outdoor ride with stops at road junctions, downhill sections, slowing down tight bends etc etc. At the end of it I have a record of power output and HR, and TrainerRoad gives me a TSS. Is this equivalent to an indoor TSS? Obviously the workout may not be as specific (might not be a sweetspot workout for example), but if I record a TSS of 100 in 2 1/2 hour ride outside, is this roughly the same fitness benefit as a 90 minute TR workout that gives me a TSS of 100?

Coach Jonathan likes to say “not all TSS is created equal” and inclined to agree with him.

The TSS I accumulate from a long steady ride feels very different to short V02 max intervals for example.

TSS is fine for basic tracking of work over long periods of time, but I wouldn’t let it dictate things too much


TSS is just IF^2 x 100 x time, IF is NP/FTP

Its the same inside and outdoors. However, outside its easy to accumulate lots of TSS by riding a long time at low intensity. You can even have a couple of cafe stops too.

To me, TSS is most useful when doing a lot of long outdoor rides. It doesn’t work well when doing a lot of racing (especially not cx and crits). One race can leave you with sore legs for days, and still only have about 70 TSS, because its short.


Yep there is data. I guess what I’m looking for is if there is any group that have retrospectively analysed this on a large scale. Until we see that all it sounds like to me is “my aunty Betty used garlic to cure her warts and it worked great, therefore garlic cures warts always”. It doesn’t hold much weight, and sits at the bottom rung (if any) on a hierarchy of evidence.

E.g. “from analysing the data of 5000 indoor and outdoor “zone 2 target” rides of trained athletes, the average outdoor ride spent an average 50 - 70% (or whatever) of the overall ride time in the desired zone”

I can’t find anything like that. All I get are anecdotes like the ones you’ve both used - “it’s obvious”, “it is so effective because”, “try it and you’ll see”.

Don’t get me wrong - I’m on board and love my indoor training.

I feel like Trainerroad data would be a perfect dataset to retrospectively analyse for this purpose.

1 Like

Regarding data - it depends what you want to compare. I’m pretty certain that if you did the exact same training sessions inside and outdoors, you’d get the same training effect. However usually it is difficult to do a highly structured session outside, so realistically you’d probably train more accurately inside. Plus compliance is likely higher, because weather and traffic don’t come into it.

The issue is that most people probably compare structured training inside to ‘just riding’ outside. If both are done for the same amount of time, structured training will have you do more work inside.


There is plenty of research out there that has measured the efficacy of training programs and sessions. It is well proven that certain types of stress lead to certain types of adaptation. I don’t know why you need scientists to then prove that following these programs and sessions to the letter (ie indoors) is better than struggling to replicate the exact demands (ie outside).

No doubt trainer road data would be really useful to see and I know Nate has spoken previously about how cool it would be to share the information. But there isn’t then a sister-set of data for outside workouts (not yet at least) that you could make a fair comparison with that proves that one is better than the other

Just to add, it’s not even just the efficiency of the workout itself (structure, no freewheeling, braking, etc), it’s all the extra time spent on stuff that you need outside but not inside. Getting your bike out and putting it back, pumping the tires up, having to clean and lube it more often, putting on sun cream, sun glasses, helmet, packing your bike/pockets with keys, nutrition, money, phone, etc. Lots of minor stuff but does all.add up, and most of us are time constrained.

I’m incredibly lucky to live near flat, quiet roads where I can do structured outdoor riding with few to no interruptions, so I reckon in my case indoor riding might only be 10 or 20% more efficient. But I suspect that on average going outdoors does add 5-10 minutes of extra non-training time, which doesn’t matter at all if I’m heading out for a few hours, but is significant if I only have a 1 hour window to train. Of course, that also assumes you have a dedicated room and trainer bike so everything is ready to go, if you have to put your bike on and off the trainer every time it’s probably a wash.

I"m in the camp of it’s not more efficient per se, but it is more targeted towards a specific training zone. Trying to hold x watts for 20 minutes outside is almost impossible unless you live somewhere where it is pancake flat or have 20 minute climbs that have consistent grades. Most climbs I know have slightly varying grades, so power will also fluctuate.

Riding indoors though you aren’t clmbing, or working towards trying to hold a steady power over varied terrain. I do live somewhere rural, so I can do most rides with < 10% coasting/stopped. Zero power during descents, may also be building on a skill so isn’t always wasted time.

  • Anecdote: Garlic clears up my warts.
  • Anecdote: I get stronger faster when I train inside.
  • Data: I spent 30% of this outdoor training session freewheeling.

Is your question Is outdoor training less efficient than indoor training, or How important is maintaining consistent power/cadence in a training session? The consistency you can achieve outside is so dependent on your route and the environment that I’m not sure it’d even be possible to generalize over 5000 rides and come out with actionable conclusions. A study would need to control the workout, the route, the time of day and the weather, and the demographic and fitness levels of the riders, and the results would be too hyperspecific to be useful.

The TR workout instructions and conversations on the podcast all talk about how a 10-second backspin break here and there doesn’t impact the goals of the session, so there must be some threshold below 100% at which point consistency stops mattering. If you can hit that breaking point of consistency outside, then you could replace your indoor rides with outdoor. The research that backs up the 10-second backspin break (assuming it exists) is probably what you’re after.

Pedaling is pedaling so in that sense there is not any difference between pedaling on the trainer or outside.

The increase efficiency of an indoor trainer comes from the fact that you are not coasting and doing other non-productive riding that inevitably occurs on almost every outdoor ride. You can see this just looking at the data from any outdoor vs indoor ride. That is where the efficiency comes in. For example, I frequently do the Wynn workout outside. That is a one hour workout on the trainer. It takes me an hour and a half to do that same work out outside because I have to ride to and from the spot where I can do the intervals. I get The same workout in doing it outside, it just takes 50% longer to accomplish it .

If you do have a wide-open road with no stops, it is entirely possible to match rhe efficiency of a trainer workout outside. It just doesn’t happen that way all that often.


Surely, there is no such thing as the efficiency of outdoor versus indoor riding. It all depends on circumstances – frequency of stop signs / intersections; extent of downhills; wind direction and strength; and the like.

But it is quite easy to examine particular rides and to compare their efficiency with a given TR workout. For example, yesterday, I was scheduled to do Red Hill + 8: 1:30 hours, 111 TSS, comprising 5 * 6-minute intervals at 115% of FTP, plus warm up, rests and cool down. I tried to replicate it outside [wahoo, so following learned instructions rather than being pushed onto garmin]. This was a rural route, not too many stop signs / intersections. with slight elevation changes, but a wind. Result?

[A] It took 2:00:59 hrs to get 113 TSS. Efficiency outside ~ 75 per cent.
[B} Rest intervals were generally longer than prescribed [waiting for a rise / flat section / turn into the wind], so heart rate in zones 4 or 5 [> 85 per cent of HR max] only 00:25:25 hours, compared to ~ 00:30:00 hours indoors. Efficiency outside ~ 85 per cent.
[C] On one interval, I mis-remembered the road, and there was a stop sign in the middle of the interval. On another, I mis-remembered again, and there was a downhill / with the wind section where it was hard to maintain the power target. Two intervals messed up, out of five. Efficiency of interval completion outside ~ 60 per cent.

And so on. All you have to do really is to examine a couple of your rides in this manner to see what is really going on here – loss of efficiency, depending on your measure. Based on experiences like this, I usually reckon that an outside ride will take 30 - 50 per cent longer than an equivalent inside ride. But on some rides, with lots of stop lights, it could be far greater.

The point is that it is easy to actually identify efficiency losses outside – though the riding might be nicer [which become more and more important, the longer and longer the workout is]. You don’t need a study of 1000 riders to identify this with data: just look at some examples. And remember that each TR workout: outside ride comparison is unique.


I do think that TSS is to keep up with how much you can handle. So if your body can handle 1000TSS for a week - then if tss is 400 or 200 doesnt matter - it gains your fitness - however if you end of the week are hitting over that 1000TSS, then you are overreaching ect. So its a number to get you to not overtrain ect.

Look. Indoor riding is more effective. Not many who can go on a ride outside and be home after an hour, and have been doing all the work you can do on the trainer. Its the only thing that seperate those too.

So the reason indoor cycling is more efficient, is because you can do a whole lot when you are there - and the pedaling is 100% - its ever outside. So indoor riding is definitely more time efficient than outside, but not harder or better.

1 Like

Also, most people have to commute to a route that is suitable for interval training. It takes me 15-30 minutes one-way to do that — which just adds filler to training time. Plus, even on quiet roads it may not be possible to focus as much on your power numbers, especially once you get deep into the red.

If efficiency is about avoiding the waste of time and effort in training then Michael has given us example data, most people with a power meter have similar, but this is self evident - it doesn’t need a study to determine if indoor is more efficient than outdoor.

However, effectiveness is a different question, and calls into question how you train indoors. This “study of studies” for example;

There are plenty more research papers, but combine the evidence of training responses with the self evident efficiency of indoor training and you have an objective view on indoor training.

Of course it’s not all one way, core temperature is a challenge indoors - we may need a study to prove core temp is important, as below

But we don’t need a study to know that we get hot training indoors.

Lots of negativity towards outdoor training only. There are loads of people out there who don’t use a trainer and are competitive athletes. I’m not saying one is better than the other so please don’t bite my head off.

In regards to the nonsensicle debate about trainer efficiency, it really depends on the individual workout.
Are you able to workout your whole body on a trainer? What’s the efficiency of a trainer vs using all those extra muscles to actually ride outdoors? Could you only do trainer workouts?
I’m pretty sure I have a load of rides on file that show me doing more work outdoors than on the trainer. They also line up with big bumps in performance after a week or so.
I don’t think I’ve ever heard the TR crew suggest a trainer only based training plan. I also don’t remember hearing claims of a constant ratio of trainer to outdoor riding benefit. More of a min/max ratio.

Most of the time, I can sneak in a really solid interval workout within 1:30 hrs in total. Some days I have to can them as I have a hard day on the tools planned and won’t be able to work. I can also do this to myself on the local trail, but I’m much less likely to do so day after day. Definately not at 6am.

Plenty of data out there.

1 Like