Sure it matters, which is why you don't do a route with a traffic light in the middle of a planned interval. Your biking example is ridiculous. We're not talking about 2 hours between workouts. We're talking about 1-2 mins waiting for a stop light.

That is an exaggerated case of course, but I think it plays in the micro version just as well. Take a workout where you are doing cyclocross starts. Typically this workout involves generating 700+ watts from a standing start, waiting 2 minutes and doing it again. you might do 8-10 of these, where the AP in the start is 400+ watts for a minute, WAY over threshold. If I had auto pause on while I was recovering between starts, I record a workout of 10 minutes at 150% of FTP. In reality I had a workout with a duration of 30 minutes with an average output of 133 watts.

And yes I do this workout from time to time and it looks like this:

30 minutes sustained riding at 80% FTP followed by 10 standing starts with 2 minutes recovery between efforts.

which is 30 minutes at 192 watts and 30 minutes at 133 watts for an average of 162.5 watts. With auto pause this is a 40 minute workout with an average of 244 watts.

Someone want to convert those two to TSS?

I'm gonna guess about 70-80 TSS. You should do the workout with auto pause on and see what the difference is.

I still don't think that's the same as your computer auto pausing during a stop light though. TSS is calculated off normalized power, not average power, and it's a rolling 30-second average so short intervals with rest breaks and starting and stopping your computer has more potential to throw that off than a short break at a stop light every now and then.

If the concern is TSS, I don't see how pausing or not pausing at stop lights is going to throw off TSS enough to worry about. Most people don't even have an accurate FTP in the system, partly because they do tests to estimate FTP rather than doing a 40K TT and because people don't test often enough to be able to say their FTP is 100% correct all the time. That's going to throw off TSS more than a stoplight.

There's a difference between understanding what average power is, and what training is effective. Just because people stop and/or incorrectly calculate average power doesn't mean their training is ineffective.

Where it can go off the rails is when people misinterpret the numbers to indicate something about their capability, and use that misunderstanding about their capability when making key training/racing decisions.

If they are using good analysis software that correctly calculates such values, such misunderstandings should be reduced.

---

http://www.cyclecoach.com
http://www.aerocoach.com.au

Which completely doesn't matter a bit because in races you...coast. At least in bike races.

Interval sessions? I don't do those on stretches of road where I have to stop. Normal riding? I'm for sure not counting time when I'm not moving. Totally jacks up ride time and skews speed and power numbers. I want to know what I'm doing when I'm pedaling, not stopping.

This seems to be much ado about nothing. I will never understand wanting to have numbers for when I'm not actually moving. It's a bike ride. Makes zero sense to have numbers from times when I'm not riding my bike. .

What actual data do you find so necessary as to include information from times when you're not riding?

I can understand interval or workouts (again, I wouldn't ever do intervals when I knew I'd have to stop, and when simply reset if I did have to stop) but I can't understand full rides. You're inflating ride time at best and giving yourself an inaccurate picture of what you're actually doing riding.

Think about an actual race situation where you're forced to stop. Do you genuinely think that rest is going to have a positive effect? No race where I've ever had to stop was for the positive (free laps in crits, wheel changes in road races, etc). It was a highly anxious time after which I then had to put in a significantly harder effort to get back to the speed/position I was at. "Rest" actually hurt had negative consequences.

That's not in any way relevant to a training ride, though.

I don't get your position. You are riding at 20 mph. If you consistently ride at 20 mph, that is your speed. Why include time when you're not actually moving?

Do you think stopping and restarting and still maintaining a 20 mph average is actually easier than just riding a single, non-stop session at 20 mph? It's most certainly not! It's much, much more difficult.

Someone saying they averaged 20 mph while going in and out of town with redlights and stops and all that had to go significantly harder than someone who rolled out their front door and averaged 20 mph without ever slowing down or stopping.

But you'd suggest the first one was "cheating" and doing themselves a disservice? That's really strange.

My ftp isn't affected a bit by my computer numbers including stops or not. If you think it is, you have some very odd notions of physiology.

Your post reminds me of a guy I met about 15 y ago who told me he could average well over 400 W during criteriums. Yes, he was a cat. 1, but not all that big, so I knew instantly that the number was incorrect. Turns out he was a Mac user, and therefore never downloaded his SRM PCIV (which excluded zero values from averages) as at the time the SRM software only ran under Windows.

Bottom line: time not pedaling = time resting, which increases the power you can generate while actually pedaling. Excluding zero values therefore provides a distorted picture of what you did/what you can do/the physiological demands.

Lets say you go do a 2 hour hilly ride on your own. Your avg power for the 2 hours is 300 watts (not including zero's). Your effort level was moderately hard.

Now, hypothetically speaking, go do another 2 hour ride (maybe a week later) that is completely flat with zero stopping so the zero's or no zero's are irrelevant here. I can guarantee you that you won't be anywhere close to holding the whole 2 hours at 300 watts. I'd bet you would average maybe around 200 watts for the 2 hours yet you gave the same amount of effort compared to the hilly ride.

The point of including zero's is ensure all your rides are measured and reported from the same viewpoint. Without including zero's, you cannot compare ride to ride (unless you ride the same exact route every single time).

---

blog

As Andy said, time spent producing no (or low) power impacts the power you can produce. I consider that useful information.

Aside from that basic principle, in some events it's also important to understand how much non-work you can get away with. It takes craft and skill to produce the least amount of power so that you can put the hammer down when it really matters.

But as I pointed out earlier, incorrectly calculating average power doesn't mean one's training is poor, it just means average power has been incorrectly calculated, or that you are calculating something else.

---

http://www.cyclecoach.com
http://www.aerocoach.com.au

You're responding to the wrong person.

I'm not talking about coasting. I'm talking about not moving, ie, at a stoplight. Your example has nothing to do with autopause.

If you could point me to my post where I ever mentioned not averaging zeros, I'd appreciate it.

I'm talking about stopping, as in not moving. There seems to be a communication error on your end.

Yes, I'm calculating the average power I produced when my bike was actually moving forward.

To be sure, that is different.

It's also different to calculating an average of all non-zero power.

---

http://www.cyclecoach.com
http://www.aerocoach.com.au

Yes, we said the same thing, I think you misunderstood my point to be the opposite of what you said. It is much harder to average 20mph while stopping, but if someone auto pauses their ride, and then says they averaged 20, I'm saying they cheated, not the other way around.

Whenever I do a TT type ride, my normalized power is lower than my average power. Crit/RR are always much higher normalized. And, yes, I'm including zeros.

40k example.

They cheated because they worked harder? Personally my rides with a lot of stop/starting tend to have a lower average speed (with autopause on) than longer uninterrupted rides.

Unless you're stopping for 5 minutes at a time I doubt the recovery is going to offset the repeated accelerations plus the faster cruising speed necessary to offset the time spent accelerating.

---

---------------------------------------------------------
All we have to decide is what to do with the time that is given to us. ~Gandalf

Here is the disconnect. The average number you are posting is irrelevant. Why would you care what your average power is over intervals including zeros? If you're doing the intervals what matters is the interval power not the interval plus rest power. Its the same thing as why stops lights are irrelevant. I'm not looking at my average power with a 4 min break for a stoplight. I'm looking at my average power, on the part of the ride that doesn't include stoplights which is most of it. Also, based on your analysis your ride was 60 mins, where you really only rode 40 mins. I would much rather know the latter number not the former since I want to know actual ride time, which is way more important to me.

Whether auto pause for relatively short periods has an effect on the validity of average power numbers for assessing your training is surely dependent upon the intensity of the ride? On a base ride at 65% FTP for 5 hours, I don't think the training effect would be significantly influenced by a few stops for a couple of minutes. However on a tempo ride at 90% FTP, that few minutes break would allow you to achieve the 90% level with much less effort and hence less training effect.

I was thinking of the example of the run/walk strategy for the IM marathon, where you are going at race pace, but the run segments are at a higher speed than you would achieve by a straight run with no stops. If you took the average of your running speed, then this would not be a useful figure for planning a IM marathon.

Not because they worked harder, because they got to 20 by taking out rest periods. Their garmin says 20 when they really only went 18.

Whatever software you're using isn't calculating normalized ("weighted") and/or average power correctly.

As a result, any TSS ("training load") values are also going to be incorrect.

No, I didn't.

Rest is rest, regardless of whether you're on or off the bike.

Nope.

I'm saying someone averaging 20, WITH autostop, who has to stop and slow down and accelerate, is working SIGNIFICANTLY harder than someone who averages 20 but never has to stop at a stop sign or stoplight and just maintains steady power/speed.

So what I'm taking from this is you think the first guy is "masking" their ride and actually has it easier because he's not including the stops. But that's absolutely not the case because slowing down and accelerating multiple times and still maintaining 20 is obviously harder to do.

And if that's not what you're saying, then I apologize.

Yeah, you did. Which is why you gave a completely irrelevant example about excluding zeros. The exact opposite of what I was discussing. .

No point in arguing about it. Your example is proof-perfect.