I’ve tried searching the forum but must not be using the right terms… I rode outside yesterday for the first time with my new Garmin Rally 100 power pedals. Since it is the 100 only the left has the meter. My numbers were ALL OVER the place trying to do an endurance ride at 140W. I rode with 3s avg power being displayed as recommended by TR. I’d be riding along with the same cadence with no noticeable changes in the road/wind and my numbers would jump from 100 to 190 to 90 to 150. I noticed I seem to have the bad habit of pushing my foot towards the front/top of my shoe instead of down towards the pedal (I hadn’t realized this…) so maybe that is part of the problem? I live in a hilly area and this wasn’t even an issue with variable terrain - it was second to second on level roads. Are the articles about this? Personal tips? I’ve seen posts of people who have flattened their power curves over the year so I know some of this must be technique - I just don’t know what technique 
In the mean time is their value in doing a 10s avg just so my anxiety isn’t as high while I ride Thanks!
You just have to learn RPE. Once you nail down RPE, the power number will follow. Also, don’t read too much into the short run numbers. 145 average is just as good as 140 which is just as good as 135. You’re not a machine on ERG mode
This is how power works. It’s also why I generally prefer to do structured intervals on the trainer. Also, when looking at ride data, use Normalized Power and not Avg Power.
I wasn’t doing intervals - it was an endurance ride with sustained power ~130-140. I was so shocked at the moment to moment variation. I’ve only ever had power data inside (on erg mode) so this was all new to me. And I just felt like I was all over the place. And puzzled by the changes despite no change in RPE/cadence/speed in those moments where my watts were varying by 50W moment to moment.
Have you ridden on your trainer in resistance mode? If not, maybe give that a go and see how your power varies without ERG.
Just pedal. Second-by-second variations in power are 1) normal, and 2) physiologically meaningless.
The only reason to worry about smoothing things out is so avoid annoying any riding partners by constantly surging, then slowing down (the way Marty Jemison does).
You are doing nothing wrong.
This is completely normal. Look at longer averages (10 seconds) and try to adjust your effort slowly. On flat ground that means you will only need to change your cadence a little and very slowly.
For endurance rides, you can also pace by heart rate, which changes more slowly than power.
10s average might be the lowest useful data point to look at on your head unit.
Personally, I ride endurance by heart rate and feel. It’s as simple as pressure on the pedals, no coasting, and keeping your power under threshold and near the zone you are targeting (like Z2 or low Z3 (tempo). It’s ok that it varies with the terrain.
I have a data screen where I look at things like HR, average lap power, 10s power, time, etc. I do power intervals by pushing the lap button and riding to the average power of the lap while keeping an eye on the 3 sec power.
On that screen I have ave lap power, 3 second power, and power. I’ve been thinking of changing it to 10 second power and 3 second power. Power is kind of useless. And mostly I look at the average.
Just looking at this, I’d guess that there are moments that aren’t endurance, especially the first quarter or so, because it looks like the average is too high. But as others have suggested, look at longer intervals. If you don’t already, you could look at the workout in intervals.icu or something similar. That takes that graph with more variation but you also get a 30 second average. I like that more for outside endurance workouts because it gives me more of idea or what’s really going on.
And you’ll also get smoother with time and better at keeping in zone, though as others have pointed out, you don’t need to be perfect. I know from seeing other people’s workouts on this forum that other people are better than me, but I’m close enough on my endurance rides.
I was the same when I first got a power meter. I then obsessed over staying within ~10 watts all the time, even for the threshold+ intervals. Even feathering breaks to keep the resistance in corners, when slowing down, for quick dips. My average power for each interval would be within a few watts.
Then I would freak out on competitive group rides when my power was all over the place and especially during surges; it ultimately made me worse off.
Now, I’ve been without a power meter on the bike for a year or so and I’ve rebalanced by learning RPE much better. When I get a power meter again, I’ll let the numbers fluctuate more.
My snarky reply to your first post would be “tell me you use Erg mode without telling me you use Erg mode.”
But not here to be snarky. Just to echo others. That is a perfectly good endurance ride. That is exactly what real riding looks like. With some practice, you can smooth that out some, but I wouldn’t focus too much on that. Just ride.
Probably a separate topic from your question…I am starting to believe, just based on questions on this forum, that Erg mode has some serious downsides. Now I am an old(er) grumpy hater who started with a stopwatch and RPE, so I could just be biased.
I had to change my data screen on group rides to something simple like speed/time or the map.
My group ride is 350-500 watts up every damn hill. Frequently seeing 150%+ of my FTP was psyching me out.
It’s a skill, you’ll get better at it. I also struggle the first few rides outside after a winter inside. My tip would be to not focus to much on the power, especially 3s. Turn it to 10s. Get a good idea what 140w feels like and then focus on how that feels on the legs. Then, just try to keep that feeling. Using your power data just as a reference.
It’s not erg mode, it’s believing that there is something magical about training “in zone”.
Looks pretty good actually. Power outdoors is just not consistant like on the trainer. Especially single-sided measurements, because weight shifts affect it. Now that you have the power pedals, put that bike on the trainer and record power from the pedals, not the trainer. Then try things like having a drink or reaching over to pick something up, with either hand, and watch your power change. Also when you’re riding outdoors, you subconciously alter your power a little - you might unweigh the bike slightly for a pothole, or push a bit harder for some broken tarmac. Or you ease off with one foot for a corner, etc.
Also if you want to know the trick in how to make it look better…display the power graph with some smoothing…
Certainly agree, that is the bigger issue. Erg mode and some preaching behind the benefits helps prop up this belief.
This is what I eventually did. A lot of damage had already been done, though. Which is a shame, because I think I was in tune with my ‘biological bank account’ before I got a power meter. I think I’d do fine now.
I no longer do that group ride but in retrospect, I think the power meter was telling me that I should have trained those 30-60 second blasts at 150%.
As others have said. It looks like a normally variable ride.
There are quite a few things you can learn from this;
How much terrain influences your efforts
How much easier it is to focus on particular output levels when you’re indoors
How to manage your outdoor efforts to be more efficient/effective
How much heart rate only training masks actual effort
It’s quite interesting, all in all.