Level playing field across ages and across events

[MudSweatAndYears]

I spent some time over the holidays on analysing C2 ranking data for endurance rowing, and I guess some of you might be interested in the outcomes.

What I tried to do was to combine algoritms for age correction with algoritms for comparison across distances. My hope was that in doing so one could achieve a level playing field for endurance performance (regardless the event for distances 5k and up) across a wide range of ages. This would allow a fair competition across events and across ages, and in particular would give all of us the opportunity to compete against our former selfs.

I soon discovered that existing algoritms have their flaws (for instance: the order in which you apply the age and distance corrections does matter), and I set out devising a more integrated approach for creating a level playing field. I won't bother you with the details, and just present the simple algoritm I landed on.

The ACEP (Age-Corrected Endurance Performance) for athletes 20 years and older is defined as:

ACEP = (100/H).d/t^(23/24)

Here d is the distance rowed, t is the time it took, and H is an age correction factor:

H =1 - ((age-20)/134,5)^2

With the three inputs d (in meters), t (in seconds), and age (in years), you can calculate for each of your PBs what ACEP scores they translate into. The highest of these scores is your ACEP PB.
[Note: the whole approach is constructed with endurance activities (5k and up) in mind. If you apply it to shorter distances you will not get meaningful results.]

How well does all of this work? Do ACEPs indeed create a level playing field?
To check this, I calculated ACEP PBs for the top-performing male heavyweights across events (5k to marathon) on the Concept2 site. The top-10 looks as follows:

rank) ACEP . age . event . name

1) 742 . 65 . 30m . TJ Oesterling
2) 741 . 65 . 5k . Brad Brinegar
3) 738 . 32 . 6k . Eric Murray
4) 733 . 47 . 60m . Luke Wollenschlaeger
5) 730 . 33 . 6k . Niksa Skelin
6) 730 . 34 . 6k . Hamish Bond
7) 729 . 34 . 5k . Graham Benton
8) 728 . 26 . mar . Ben de Wit
9) 727 . 50 . hm . Andrew Benko
10) 726 . 59 . 60m . James Crawford

A spread across events and across ages is achieved, a sign that a level playing field is indeed created. The real proof of the pudding is in the eating: many of us are struggling to beat PBs achieved years back. Now you can participate in a much fairer competition against your former self: simply calculate the ACEP of your current activity and compare it against the ACEP of your PB (calculated using your age at that time). Enjoy!

[robbiep]

Might I suggest the nonathlon website, where they've been doing this sort of maths for years to produce rankings where anybody can compete against anyone else - age, weight, sex, all taken into account

www.nonathlon.com

[MudSweatAndYears]

Might I suggest the nonathlon website, where they've been doing this sort of maths for years to produce rankings where anybody can compete against anyone else - age, weight, sex, all taken into account

www.nonathlon.com

Yes indeed, Nonathlon uses an algoritm that tries to achieve a level playing field across ages and across events. I was refering also to this algoritm when I mentioned that "existing algoritms have their flaws". I think I am not saying anything controversial or anything new when I mention that the Nonathlon algoritm is flawed as it heavily favors older athletes. In fact, over the last ten editions of Nonathlon, no person younger than 48 years of age managed to enter the top three (for males this is even worse: no male younger than 60 years of age managed such a top-3 ranking).

[jamesg]

heavily favors older athletes.

Would seem times and curves have not been updated in the last few years, so now the faster and more numerous ergers who've become OAPs now show well over the WR 1000 points mark. It must have been very laborious, as no doubt you've found.

[KeithT]

Interesting but seems some competitors would be higher as they have created a large gap between themselves and all others in their age group. Where younger rowers are much closer.

[hjs]

Re level playing fields, for some reason when is looked at something like this, never the overall best gets on top. While the 2k and Olympic athletes put in the most effort and 2k is what is trained for. No age grouper should come up above them.

[sjors]

I have been busy for a while and did not read this threat. Very interesting formula's.
Did you find something similar for the shorter work where it seems that age does not have the same impact on performance as does endurance?
As you maybe know I'm trying to solve this issue as well for my site.

[sjors]

No age grouper should come up above them.

Henry, that is a fair point. (Maybe the formula has to take this more into account).
But on the other hand some do have extraordinary talent to achieve great performances but were not in the position to do that at their 20-30's.

[hjs]

No age grouper should come up above them.

Henry, that is a fair point. (Maybe the formula has to take this more into account).
But on the other hand some do have extraordinary talent to achieve great performances but were not in the position to do that at their 20-30's.

Absolutely true Siebe, but have you ever seen a age group athlete who puts in the same amount of time, has the same resources, the same support etc as an olympian top athlete? Those do not exist, so per definition they can’t perform at a same level. And apart from that, I do think that absolute performance is in the end what sport is about. If you for some reason you could not perform in your prime, you missed the boat.

Think to come up with a formula that takes that into account is more or less impossible. For me the top of the piramide always the absolute Wr, anything else should be ranked lower.

[faach1]

No age grouper should come up above them.

Henry, that is a fair point. (Maybe the formula has to take this more into account).
But on the other hand some do have extraordinary talent to achieve great performances but were not in the position to do that at their 20-30's.

Absolutely true Siebe, but have you ever seen a age group athlete who puts in the same amount of time, has the same resources, the same support etc as an olympian top athlete? Those do not exist, so per definition they can’t perform at a same level. And apart from that, I do think that absolute performance is in the end what sport is about. If you for some reason you could not perform in your prime, you missed the boat.

Think to come up with a formula that takes that into account is more or less impossible. For me the top of the piramide always the absolute Wr, anything else should be ranked lower.

This is an interesting discussion as to a certain extent I agree with you both, however as Henry said you can’t have anything else but the absolute world record at the top of the pyramid.

[iain]

Many thanks for all your hard work and interesting formula.

The data is also missing from most of the very best as there erg performances are kept from international competitors. So data will always be deficient. Personally I see the Nonathlon as something for older athletes to aspire to when there performances will never be that competitive with those whom "age has not withered" (to misuse the quote). So some older bias is to be welcomed as a way to maintain motivation.

Regarding your formula, I am surprised that distance performances decline after 20 given many world champions are older and older athletes seem to decline a little slower in the longer events. Also the research suggests that there is acceleration in performance lost through aging somewhere around 50 and again around 70 IIRC, your formula has no such discontinuity. Similarly, anecdotally people seem to slow down less quickly somewhere between 10k & an hour (I have assumed that this is where CV fitness becomes less important than muscular endurance) and then lose more before the marathon (presumably when you need to choose between slowing to drink or losing strength through dehydration) and again between there and 100k (maybe maximum distance before you need to use more fat as a food source due to limited storage and ability to absorb carbs). Interesting experiment and will investigate your result, but I think what you are trying to achieve will always be a significant approximation.

[sjors]

Many thanks for all your hard work and interesting formula.

The data is also missing from most of the very best as there erg performances are kept from international competitors. So data will always be deficient. Personally I see the Nonathlon as something for older athletes to aspire to when there performances will never be that competitive with those whom "age has not withered" (to misuse the quote). So some older bias is to be welcomed as a way to maintain motivation.

Regarding your formula, I am surprised that distance performances decline after 20 given many world champions are older and older athletes seem to decline a little slower in the longer events. Also the research suggests that there is acceleration in performance lost through aging somewhere around 50 and again around 70 IIRC, your formula has no such discontinuity. Similarly, anecdotally people seem to slow down less quickly somewhere between 10k & an hour (I have assumed that this is where CV fitness becomes less important than muscular endurance) and then lose more before the marathon (presumably when you need to choose between slowing to drink or losing strength through dehydration) and again between there and 100k (maybe maximum distance before you need to use more fat as a food source due to limited storage and ability to absorb carbs). Interesting experiment and will investigate your result, but I think what you are trying to achieve will always be a significant approximation.

I have read a lot of researches, but if I analyse the giant source of ranked workouts on concept2/rankings(it took me some days, trying te find a common ground for my rankings-site) I don't see those discontinuity's.
My theory is that all those researches are based on too little data.
Some of my assumptions:
1. The best age-range for WR's is between 22 and 33. Before the age of 22 we
2. The decline in performance in shorter distances (like 100m, 1min and 500m) is not as high as longer distrances.
3. LW-class probably need a slightly different formula
4. Woman probably needs a slightly different formula

To get some comparable data we need to consider all 4

[ArmandoChavezUNC]

I don't have links or papers in hand to back it up, but I'm fairly confident I've heard and read numerous times that for endurance sports like rowing you don't actually peak physiologically until your mid to late 30s. It takes such a long time to build the aerobic base needed to be at the absolute top of the game.

I think the reason we don't see as many athletes at these ages posting WRs is because very few stay in the game long enough - most burn out well before their physiological peak or simply move on to do other things. Even so, I think you'll find most erg WRs are done by athletes in the 30-40 age group than the 20-30.

[faach1]

I don't have links or papers in hand to back it up, but I'm fairly confident I've heard and read numerous times that for endurance sports like rowing you don't actually peak physiologically until your mid to late 30s. It takes such a long time to build the aerobic base needed to be at the absolute top of the game.

I think the reason we don't see as many athletes at these ages posting WRs is because very few stay in the game long enough - most burn out well before their physiological peak or simply move on to do other things. Even so, I think you'll find most erg WRs are done by athletes in the 30-40 age group than the 20-30.

Good points Armando. I too have read and heard this.

[sjors]

For anybody who is interested, I have made csv-export's of all best performances for the normal rower.
They are split into: distance/time, weight-class and gender
In each file there are rows for the best performance per age as long as there are more than 5 performances for that age.
You can download it from here: https://www.rankedworkouts.com/docroot/ ... mances.zip