Isn’t this the exact purpose of Normalized Power? The ability to estimate the steady-state-equivalent of an inconsistent effort so it can be compared to a true steady-state effort?
If you’re going to omit the 0w readings then why not also omit the 1w readings? And then why not omit all readings under 10w? Or 100w?
Your nonzero avg power if you coast through the corners will look very different than if you managed to soft pedal exactly 1w through all the same corners, but in fact the two efforts would be nearly identical physiologically. NP handles this appropriately but nonzero avg doesn’t.
Whether NP is a perfect formula or not is up for debate, but I don’t think anyone seriously doubts that it is designed specifically to account for the kind of discrepancies you’re trying to address by removing zeroes or the fact that it’s a far superior method of doing so.
It sounds to me when you talk of it being harder to hold the same average on a crit course than a velodrome that the fundamental flaw is in creating an equivalence between technical ‘hardness’ and physiological ‘hardness’. Yes it is more difficult technically to hold the same average with sharp corners, but your body can’t tell the difference between coasting for a corner and coasting for rest, just as it can’t tell the difference between 100w soft pedal for recovery vs 100w soft pedal because you’re stuck behind a tractor on a narrow road. Technical difficulty does not equal physiological hardness (except maybe in MTB but that’s a different conversation).
In fact, it actually IS as easy, physiologically, to average x watts on a course with sharp corners as it is on a velodrome. It’s just technically far more difficult so you’re likely to crash. That makes no difference to your muscles or aerobic engine though.
So it’s not a “road bike” to “MTB” on the same trainer swap, but a totally SB20 smart bike swap to XC bike on Kickr trainer change. That is a MASSIVE variable since the “trainers” are totally different to begin with:
The SB20 has a much larger and heavier flywheel with a fixed drive belt ratio.
The Kickr has a decent flywheel, but it’s 12/16 lbs weight (depending on your model) is light compared to the SB20 that is around 50 lbs IIRC.
Couple in what is likely a lower drive ratio from the MTB on the Kickr (guessing you are running a 30/32t chainring to somewhere in the middle of the cassette (for a straight-ish chain line?) and the actual “pedal feel” is likely VERY different.
Then add in the likely fact that your power data measurement won’t match, unless you transferred power meter pedals from the SB20 to the MTB as your power data source. There is a strong chance that the power data between the SB20 and the Kickr DON’T match.
And finally we return to your comment about bike fit. That is likely different unless you setup the SB20 to mimic your MTB. We are adaptable and within reason, can put out similar power as long as the setups aren’t drastically different. But we have no idea on your setups without lots more info. May be significant or negligible depending on the deltas between your positions on each setup.
With all that, you have more variables than you can shake a stick at. Any single one of those could lead to a real difference in feel, but the mix blows up to a more complex comparison. Not much to say here other than the fact that you feel a difference makes perfect sense to me.
At best, you need to consider doing a new FTP test since any results you have from the SB20 could be very different from the Kickr, along with all the other variables I covered. It’s likely a mistake to use an FTP from the SB20 on the Kickr and hope they are “close enough to work” without a lot more research and testing.
You could hold 250 for half the ride and 150 for the other half, too. Still average 200.
Every time you go a little under you then have to go a little over to chase the avg back up, rather than just making the correction and getting back to the power you’re intending to maintain.
I think the question I have is why do you think that holding exactly 200 watts while pedaling is useful, and inherently better than just doing your endurance ride with zeros included and riding around 200 watts?
To maintain an average wattage over any length of a ride, you are constantly trying to bring your average back up or down. After you stop pedaling you are going to accelerate to get back to 200. It’s a yo-yo.
If a coach prescribes a power target, for better or worse, the intention is for the athlete to ride that power as much as possible during the ride or intervals, not to end with that target as their average.
The latter is a simple idea we amateurs may have for a challenge.
It does sound like @Deej approach is a yes-and approach, though.
It’s not for me, but they’re not replacing data, they’re just adding one more data point. It’d be too tempting for me to keep that number by occasionally coasting. You’d have to be very disciplined or have the real average power or NP right next to it—something to indicate how often you’ve dropped power and rested.
It sounds like they have a specific urban commute where this has been helpful for moderating power. I would rather look at the zone distribution chart for this data. Maybe they look at this plus that. And the other.
Any average is sort of useless unless the variability is super low and you didn’t stop much.
Sure, it obviously gives him a bit of information that he wants it to. If it works for him, that’s fine.
I’ve certainly experienced how much more effort it takes to move that “average” needle at minute 120 than it does at minute 10. Most people just use some 3-10sec smoothing, right?
This actually seems to me an entirely reasonable suggestion. Taken to its logical conclusion, you’d omit all the readings, and can then stop worrying about it all.
This is the default workout screen on my Garmin 840, just after the last stop light, about 8+ minutes into the 100-minute (1:40:00) interval:
At this point my average power is 119W and NP is 163W.
From the real-time power graph I can quickly and easily see that I’m on target except for when I had to stop for traffic lights. If something feels off, I can quickly glance down and see if I’m over/under target.
Here is an animated GIF of my bike computer every 5 minutes:
Yeah, I was bored that day and thought an animated gif might look cool, so I set a mental alarm clock inside my head and took screenshots about every 5 minutes.
Last screenshot at the end of the interval:
Average power 193W, NP 200W.
Bike computers these days have other tools to help with keeping power steady when not stopped, tools like the real-time power graph or normalized power or time-in-zone.
- breathing and heart rate stayed chill - indeed HR is basically all endurance and below.
