"real life" power numbers are pretty low for many riders. A lot less than the 200+ watts that often are mentioned in online posts.
My usual riding partner usually shows a little higher estimated watts on strava. I have a power meter, so my numbers are more accurate. Even so, there's different methods of smoothing the raw data, so I might be off by 10% or so. I'm around 170 pounds, he's 175 or so. We are both retired, and ride 4000 to 6000 miles a year.
Recently, on the same ride:
Me: 33.1 miles, 2103 feet, 14.1 mph. "avg power:
119w" "weighted avg power:
138w". (I started the recording a couple tenths of a mile into the ride. My Garmin should be fairly accurate on elevation gain.)
His estimates: 33.4 miles, 1920 feet, 13.6 mph. "average watts:
121w"
"Weighted average power" is comparable to
"normalized power" It does require a power meter to include the short, hard efforts into the calculation, instead of just estimating based on speed and slope.
Strava could be fairly accurate on a long, steady climb, where most of the effort is against gravity instead of wind resistance. On the flats, it won't know about wind direction, or drafting, or even rough or smooth road surfaces. And I see my watts drop significantly when I get in the drops.