A lot - most - of the wattage values you see posted are usually estimates, and often quite poor ones depending on the particular bias of the post/article/etc. And then, of course, there’s the actual accuracy of the powermeter as well, calibration procedure, offset, etc, etc. Most of the numbers posted in this thread are probably wrong. Wiggins didn’t use a powermeter. Neither did Lionel. Campanaerts did, but I don’t know that anyone has actually seen the file. FWIW, I heard that he did 450w from someone who’s actually pretty involved in the industry, which is almost certainly incorrect, but I wonder where that 330w number actually comes from, since that also is almost certainly incorrect, though much more likely to be closer to the truth than 450 given Campanaert’s size and the fact that it was at altitude. Dowsett posted himself that he did somewhere between 350-360 when he set the record the first time, which seems correct.
Dan Bigham is probably the best actual reference point, because he’s so meticulous. I’d say if you want to get a sense, look at Bigham’s (unofficial) hour attempt at Derby, where he did 357 to go 52.631 - https://www.instagram.com/p/B9ADHoHBci3/?hl=en
Derby isn’t known to be a particularly fast or slow track, and I don’t believe they attempted to optimize for air pressure. It was more about modeling and testing of their model; they predicted to within a few meters how far he’d go. And Dan is probably about as close to perfectly optimized in terms of equipment and position as you can get, though I’d bet he still has some more gains in terms of skinsuit (I don’t know if that was in the latest Velteq suit) and - possibly - shoe cover-socks. I think they’d pretty much finalized the cockpit in terms of monoriser and base bar at this point, so likely the remaining gains - which may be significant - are in the fabric/clothing department. Helmet/bike/cockpit/wheels/tires/position/drivetrain all seem pretty much final at this point.
The technical optimizations are pretty astonishing. After spending time with the Huub guys, Ashton went from 4:26/7 (2018) to 4:16 (2019) in Carson, shattering the track record of 4:25 (which was from World Champs). Some of that was fitness; but I’d doubt much. Almost all of it would have been technical optimization. Which also goes a long way to explaining why distance and power so poorly correlate. Zirbel was a monster in terms of output, but arguably rode the least optimized setup for an hour I can remember, though Aguas is so fast it makes up for a lot of sins.
The technical optimization of the hour - and of track cycling in general - both makes it more and less interesting, I think. The chase of a lot of these records comes down to who has the resources - either personally or externally or both - to go after them. A legitimate national record attempt (for most countries) can be made on 5ish w/kg and a legitimate overall record attempt on 5.5ish wkg, which is nothing really remarkable in and of itself. 5 or 5.5w/kg for an hour in an optimized aero position on a 250m track is more challenging (even more so depending on how hard the track is to ride; Aguas is very easy), but definitely something that can be managed with any sort of reasonably disciplined training. But Aguas is expensive. Record attempts are expensive. Equipment can be REALLY expensive.
Our very own MTM is one of the most credible experts on this, so I’ll be interested if he chimes in. Hope he does. He’s recorded some of the all-time best hour performances in pursuit of the Danish record.
There’s really not a whole lot - if anything - left on the table physiologically at this point. Most training “optimizations” - USRPT, MED, etc - have proven, unsurprisingly, to be fads. And I also think that as the obsessive shift to “quantified self” has made metrics even less reliable. Since wattage matters so much in and of itself, there’s definitely a desire to favor “optimistic” power sources. Because people are so impressed(?) by these numbers.
If we see a technically optimized rider who is also a physiological phenom who also selects the fastest track, I think the record will be untouchable for a while, as it was with Merckx. We’ve only really had two of the three so far. There’s also discussion that Campanaerts now realizes how much he left on the table technically and that he may take another crack at it with a focus on technical optimization.
**TL;DR - don’t believe the wattage numbers. They’re usually wrong. **