P/AR all-time 2k leaderboard

[Cyclingman1]

Some comments:

Any tweaking of the P/AR calculations to make them more legit is a good thing. but, then again, it has to do that. Not sure that I totally grasp what the fallout will be from the proposed tweak. But I am interested in the results.

Advantage of Slides over Rowerg - a much debated notion. For me, all of my best times are on Rowerg, except for 500m and 1 min. But then there are many factors: age, where in training, how many attempts, etc. All things being equal, I lean towards Slides as having a slight advantage. But it is not significant enough to disqualify one from these sort of rankings.

#19, Cureton, in the latest list is totally bogus. There was commentary in the forums at the time of the posting. This the individual who signs up for races but either never starts or stops midway. 'nough said.

[Elizabeth]

It's going to be really hard to get widespread buy-in for a ranking system that places Rich Cureton above Josh Dunkley-Smith.

In essence, P/AR scoring yields a 'healthy ageing ranking'. As such it is intended for athletes who are well 'over the hill'. Athletes in their early thirties or younger are disadvantaged as the P/AR algorithm is based on a power decline linear with age.

A simple fix would be not to show athletes younger than say 35 in P/AR rankings. A better alternative would be to top off the age reserve for younger athletes. Something like age reserve is 80 years for athletes aged 40 and younger, and drops one year for every year the athlete is older than 40. Will do some analysis (determine the best fit to the online C2 data), and publish this revised list when ready.

One of the American Olympians is 40 and about to compete in her third Olympics. I'm not sure what her erg performance has looked like over the last decade+, but I do wonder when age-related decline starts for someone consistently training.

[MudSweatAndYears]

One of the American Olympians is 40 and about to compete in her third Olympics. I'm not sure what her erg performance has looked like over the last decade+, but I do wonder when age-related decline starts for someone consistently training.

The analysis I am currently doing should shed some light on this. In essence, I take the top 1% scores for the various ages in the C2 database and apply P/AR equalizer algorithms using different age decline (age reserve as function of age) models. The model that creates the best level playing field is the one to be used in the P/AR metric. Up till now I have used a simple linear age decline model, but currently I am considering a model that assumes a flat performance till a certain age, beyond which it starts to decline linearly. Of course we should be sensitive to the fact that there is a trade off between accuracy and transparency/simplicity, but I think this added complexity will be worth it.

[Cyclingman1]

One of the American Olympians is 40 and about to compete in her third Olympics. I'm not sure what her erg performance has looked like over the last decade+, but I do wonder when age-related decline starts for someone consistently training.

That question is complicated by the fact that better training and/or more efficient actions can mask decline. But under the covers, it is fairly well accepted that physical decline will begin at age 40, if not before. Of course the slope of the decline is minimal for a number of years but will increase over time. By 70 the declines are substantially greater.

One thing of interest is how each individual declines. Some athletes are toast by 40 or shortly thereafter; others are winning Olympic marathon medals.

[MudSweatAndYears]

One of the American Olympians is 40 and about to compete in her third Olympics. I'm not sure what her erg performance has looked like over the last decade+, but I do wonder when age-related decline starts for someone consistently training.

That question is complicated by the fact that better training and/or more efficient actions can mask decline. But under the covers, it is fairly well accepted that physical decline will begin at age 40, if not before. Of course the slope of the decline is minimal for a number of years but will increase over time. By 70 the declines are substantially greater.

One thing of interest is how each individual declines. Some athletes are toast by 40 or shortly thereafter; others are winning Olympic marathon medals.

To complicate things further: age decline is also dependent on race distance. The general trend being: the shorter the distance, the stronger the year-on-year decline. The anaerobic engine is more difficult to maintain than the aerobic engine.

[Cyclingman1]

Hope you can post your revised P/AR method. Course, as always, it will undoubtedly knock me way back, which is OK.