If you take out the descending, what is your NP? With 2,100 ft of climbing, and I’m guessing that amount of descending, your NP is probably being brought down by low / zero power during the descents.
Another thing that could be looked at is cadence. Depending on your trainer setup you may be using the same cadence all the time whereas outside cadence is constantly changing and putting a different level of stress on the legs/lungs. I don’t have hills around where I live, but LOTS of swirly wind so even on flat routes going in the same direction I’m often left shifting gears a lot bouncing between a “comfortable” cadence (sometimes I use this as a training tool just doing it as a “fixed gear” ride and working through those dips and peaks in cadence as they come). This could be a double whammy for you in this case if the cadence is significantly different while climbing than efforts on the trainer since position on the bike is going to be different and muscle utilization would be different from the change is cadence as well.
That’s a good point about the descending throwing things off–there is more opportunity for coasting or even just quick stretches where the power drops to zero when riding outside whereas on the trainer my legs pretty much don’t stop moving the entire time which keeps the overall power for the ride up.
What does your power profile look like? Although you said you FTP via the ramp test was 316, most people in reality cannot hold that power for an hour.
Also, where you well rested for the effort? The ramp test typically comes after a recovery week, so did you perform this effort after a recovery week, or had you been throttling it leading into this effort?
Lastly, if you live in a warm climate the heat could absolutely knock you down a fair bit if you are not acclimated.
Does your analysis of the rides include comparing variability indices? I’m guessing your trainer time will show a lower VI than the outdoor ride – meaning less consistent power outdoors (as has been mentioned as a distinct possibility in this thread) than indoors. If you have those numbers compare them.
Not all efforts are the same. You’d really have to analyze your output across a wide band of power to understand the details there.
I’ve been on group rides with 280 np, and solo rides with 280np and let me say they are not the same thing at all. 
I find near threshold (or over) intervals considerably harder than steady state riding at the same power, probably mostly mental, but compare a tt rider vs a climber stylewise. I also know that I hate headwinds and can become demoralized in the wind.
The other thing is that I don’t (like many) put out higher power on a climb, typically my numbers match very closely, and I somewhat put that down to the amount of steady state higher speed riding I do and my generally fairly consistent cadence on both.
Lastly I’ve been wondering lately about the efficiency of my drivetrain and the trainer itself and how the resistance style (I have a kickr core) might have an effect on the power stroke, especially in erg mode, since all of my riding is indoors nowadays. Previously with my power meter pedals I noticed that body position on the bike had some impact on the instant power readings and I wondered if I was subconsciously optimizing for that body position to get bigger numbers. oh the tangled web.
Hi all, thank you so much for the thought provoking replies (exactly what I was hoping for!). Sorry I’m so late in responding, had meetings all morning. 
I’ll try to reply to the important questions here.
@mcneese.chad I ride a Wahoo Kickr in erg mode. Powertap P1 pedals (gen 1). These outdoors rides that I’m referencing are on the same bike/powermeter as indoors (I’m not including MTB/CX rides in this discussion). Usually keep it in the small ring, 39/16 or 17.
Power and elevation profile are attached here. Finish is 635 feet higher than the start. General flow is: big climb, medium descent, 6.5 miles of flat, big climb, short descent, shorter very steep climb to finish. As you can see, it appears that my output is not very steady. NP on the middle flat part was ~250 W.
@tommyexmoor don’t use an HRM, would be good right about now.
@TwentyOneSpokes yes, my TTE at FTP is probably no more than 30 mins, but much much longer at 90-95%.
@old_but_not_dead_yet Testing FTP with ramp test. This year I have hit
288–>296–>299–>300–>305–>–>310–>316
@HMG Yeah, it’s hot out, but it’s pretty damn hot in, too. 
I’ve also been riding small chainring exclusively indoors. Partly because it’s quieter, partly because I’m a CX racer which is kind of a ‘small ring application’.
@JoeX Yeah, you may have it, not enough time on the roads.
@AlphaDogCycling Not totally sure about NP just on the climbs/flats, but isn’t that the point of using NP vs. POW? Shouldn’t NP take care of that (to some extent)?
@wyku Cadence indoors is usually in the 80-90 range. Yesterday was 83, so a little low but nothing crazy.
@framness For sure less consistent power outside!!
Thanks to all, keep the suggestions coming!
5 Likes
This could probably all be summarized by riding outdoors more often, like you’ve mentioned. I just wanted to share that I’ve had the same results. After an entire winter of exclusively indoor training, I was surprised by how soft I felt outdoors. I really just don’t think it’s the same. And I’m talking road miles, obviously as soon as you go off road, it’s even more striking.
Think of the added cooling benefits you get outdoors, that always helps me nail outdoor workouts more effectively. And apologies if all of this has been mentioned, just adding my anecdotes and don’t have time to read the entire thread.
3 Likes
- Good, but you left out the gearing you use while in ERG. Can you add that too?
1 Like
If you look at the way that NP is calculated (see below), it is readily apparent that coasting / lower power will decrease NP compared to riding at a more constant power.
The other observation is that your power profile looks nothing like I would assume it would if you are trying to TT this route / ride the entire route at your FTP. You spike your power at the bottom of the first climb, and then your power fluctuates pretty widely. I would assume your power would look more like this on the climbs - targeting a power number you think you can hold, and staying pretty close to that number:
If you have a hard time doing this outside, then that is definitely a skill to work on - being able to sit on a power number, and keep it steady. I’m pretty good at this on climbs (up to about 40 - 50 minutes), but much worse on flats.
NP Calculation Methodology:
- Calculate a 30-second rolling average of the power data
- Raise these values to the fourth power
- Average the resulting values
- Take the fourth root of the result
6 Likes
You know, yeah, it’s exactly profiles like that example that I can never hit. I know this sounds ridiculous, but honestly that makes me question my power meter (I know, I know, it’s a poor artist who blames his tools…)
Here is the 6.5 mile flat section from the middle of that road ride (Powertap P1):
Looks crazy ragged to how I ‘feel’ like I was riding.
Here is a long-ish climb I do a lot on the MTB, from a recent ride (4iiii single sided crank PM):
And finally we’ve got a 10 minute hard interval I put in on a relatively flat stretch of gravel on my CX bike (quarq PM) (note there are a couple of trail disruptions on this route, you can see the effects on the profile)
Maybe I’m crazy, but the road looks the most spiky, which really shouldn’t be the case. Like I said, I start to question the PM, but it’s the same PM I use indoors, where it always tracks the workout as it should
Have you ever Froome climbed? That is, climbing basically looking at your computer showing something like 7 sec smoothed power, with the goal of keeping it at a specific power number? If not, try doing at least a 10 minute climb at a “comfortable” pace (e.g, 90 - 95% FTP) with the goal of being as consistent on the power as possible. And then see how smooth that ends up being afterwards. Being able to climb / ride at a really consistent power number is a skill.
2 Likes
Let me guess: wonky power meter, the failings of which are obscured indoors due to how the data are being collected and displayed.
+1
Yeah…maybe? I’d love to believe that, but it seems unlikely that the wonkyness wouldn’t manifest in some visible way indoors.
Very likely that I’m just crap at TTs. My numerous triathlon failures can attest to that…
Looks artificially smooth, like you are recording Kickr power. Do you have a direct drive (wheel off) Kickr? Are you using PowerMatch, or is it disabled?
4 Likes
Yep, direct drive kickr. Just checked to confirm, power source listed as ‘Saris power meter’.
It’s not that smooth, some of those intervals are a little jagged. Could you post one that looks different?
Similar workout as the one you posted (this is Truuli -2) and the setup is:
- 2017 Kickr direct-drive
- Stages gen3 dual-sided power meter
- TrainerRoad app setup to use PowerMatch
and another one with Stages gen2 left-only, this time its Truuli the same workout as you posted:
Wow. That’s eye opening, I’m going to have to investigate. Can I ask, does the ride page say something ‘Stages power meter’ under power source?