Age distribution of all Open Era major finalists

[falstaff78]

[edit: updated through AO 2019]

Folks

I have plotted below the age distribution of all the finalists of Majors and Season-Finales (i.e. TMC, WTF etc) in the Open Era.

There have been 251 such finals since the 1968 French Open, (plus 2 season finales decided by round robin.) That gives us a grand total of 504 winners and runners-up. What you see in the chart below is the frequency of ages for each of these 504 appearances. So for example, 66 of the 504 appearances were by 25 year olds.

To me this chart is a great visual interpretation of the effect that age has on elite tennis players’ performances. It gives historical context to two questions which are frequently discussed on these boards, and which are important to me as a Federer fan.


1. The extent of the handicap Roger faced while competing with the "big three" from 2010-2016

From 2010-2016 Federer had 27 wins and 32 losses vs. Murray, Djokovic and Nadal. The chart really puts these numbers into perspective.

After 29, players' performances decline precipitously. From 2010 to 2016, Federer was aged 29-35 while his competitors were 23-29. Now take another look at the chart and think about those win-loss numbers for just a few seconds....


2. The extent of the challenge facing Nadal & Djokovic in chasing Federer's major total

Djokovic is 31. Look at the drop-off from 31 to 32.


Few sundry notes:

- You may have a concern that the distribution would be different for the 204 major winners, which is a more exclusive club than the 504 finalists of majors and season finales. So here is the age distribution for only for winners of the 204 Majors in the Open Era. The short answer is, nothing much changes.

- You may also have a concern that the distribution may have changed over time. In a subsequent post I will share the age-distribution of major finalists by decade. Once again, the main message doesn't change. This histogram is very consistent across time.


Enjoy,
Falstaff78

 

[falstaff78]

[EDIT: Updated through AO 19]

Here's the age distribution of all the major finalists in the Open Era, broken out by decade

Observations:

1. Peak performance consistently remains in the early and mid twenties.

2. There are interesting variations between decades, which convince people that "something is changing" which it really isn't. The 80s and 90s were "young" and the 2010s have been "old". But you can be sure when Nadal, Djokovic and Federer retire, the 2020s will mean-revert rapidly.

 

[falstaff78]

[EDIT: Updated through AO 2019]

And finally, here's the age distribution of finalists at the 4 majors & the YEC

 

[Quite Please]

Great statistics, really shows you how big of an anomaly Agassi was during his last years.

 

[DropShotArtist]

Yup I've pretty much said this all along. Biological age is pretty much an absolute. You can't fight it.

 

[tudwell]

Holy crap, who's the 37-year-old that won a major (or the year-end final)?

 

[beernutz]

This list covers grand slam winners but itis interesting: http://www.tennis28.com/slams/agerecords_winners.html

 

[Phoenix1983]

Holy crap, who's the 37-year-old that won a major (or the year-end final)?

Ken Rosewall (37 when winning the AO in 1972).

He was also the finalist at W and the USO aged 39, in 1974.

If you want to know more about him, ask BobbyOne in the former players section. :twisted:

 

[tudwell]

Ah, I should have known it was Rosewall. His last three slams were the three oldest in the Open Era!

 

[mattennis]

Great stuff. Thank you so much!!

 

[zagor]

Great thread, it is something most us fans knew of course but it's still nice to have it displayed in such format, might be an eye opener for some here (though I rather doubt it).

 

[Cosmic_Colin]

Interesting stats and nice analysis, thanks for taking the time to put it all together.

I guess the question is:
Is Federer an outlier because his level has continued into his 30s...

or...

has he followed the normal path of decline, but his mid-20s level was so high that he is still a slam contender.

 

[kishnabe]

Brilliant use of Data. Can't wait to see him reach 2-3 finals next year to offset the 32 year old data.

 

[zagor]

Interesting stats and nice analysis, thanks for taking the time to put it all together.

I guess the question is:
Is Federer an outlier because his level has continued into his 30s...

or...

has he followed the normal path of decline, but his mid-20s level was so high that he is still a slam contender.

As far as I'm concerned it's not really a question, it's the 2nd option.

That said, I'm a Fed fan so you can call me biased.

 

[TMF]

If nadal wants to catch Roger, he better start playing and not wasting precious time.

 

[tudwell]

Interesting stats and nice analysis, thanks for taking the time to put it all together.

I guess the question is:
Is Federer an outlier because his level has continued into his 30s...

or...

has he followed the normal path of decline, but his mid-20s level was so high that he is still a slam contender.

Definitely the latter. Agassi was an outlier in that he won most of his slams aged 29 and older. Rosewall was an outlier just by his sheer longevity, winning slams consistently into his mid-late 30s! Federer has not proven to be an outlier as such (ignoring for now that his peak level of play was very much an outlier) and has declined at a pretty normal rate, it seems.

 

[tudwell]

Is there any chance of getting the values in the first graph expressed as percentages of total Open Era slams? I think that would make it even easier to compare the likelihood/difficult of winning a slam at various ages.

 

[powerangle]

Thanks for the very interesting graphical presentation!

 

[Cosmic_Colin]

Someone can correct me if I'm wrong here, but I was trying to work out the probability of winning a final, once it has been reached.

So there are 444 finalists, which means 222 winners, which means each of the winner bars should be half of the finalist bars.

Finalists at 24-25 = 64 + 64
Winners at 24-25 = 33 + 31
(exactly half)

Finalists at 31-32 = 8 + 2
Winners at 31-32 = 4 + 1
(exactly half)

Because the counts of 31 & 32 year olds are so low that there aren't enough data points, here are the combined counts for 30+ year olds:

Finalists at 31 & up = 24
Winners at 31 & up = 10
(just under half)

So what I think this means is that if someone in their 30s reaches a slam final (particularly their early 30s) then they are just as likely to win as they would have been at a younger age - about 50%. Otherwise, we'd see fewer winners in comparison to finalists at that age; less than 50%.

Does this mean that as players age, it isn't winning finals which is the challenge, but getting there?

 

[NRod2]

Watching Federer gas vs. Murray supports this thread 100%.

 

[Phoenix1983]

Interesting stats and nice analysis, thanks for taking the time to put it all together.

I guess the question is:
Is Federer an outlier because his level has continued into his 30s...

or...

has he followed the normal path of decline, but his mid-20s level was so high that he is still a slam contender.

I think looking at the graph, the 'normal' age range is 17-31, with 24 being the mid- (and high) point. A number of greats i.e. Federer, Sampras, Connors are still good enough to win and reach slam finals at 31.

Men reaching slam finals at 32 or older are very rare in the Open Era (in fact I think it's only Rosewall, Agassi and Gimeno).

Thus if Federer continues to do so from now on, he can be considered one of the 'longevity greats' as well as the 'dominance GOAT'.

 

[TMF]

It would be nice to see the trend for the WTA too.

 

[zam88]

Given that fitness and nutrition are reaching new levels that have never been seen before in tennis, and the lack of being able to compete much anymore at the highest level when under 21, would we expect to see more dominance between 23-30 rather than 20-27 like it used to be?

I do realize DelPo and Djoker won titles when they were just 20, and that Nadal was winning titles as a teen....

So... is it that we have a lack of ultra premium players in the pipeline right now? or should we expect the ages for title contenders to rise?

I expect Djoker and Murray and Delpo to be winning titles for the next 3-4 years, with a Federer title in there somewhere and a Nadal French or 2.

 

[Goosehead]

another older player winning stuff was laver in 1970 i think..

at the age of 31/32 he won 5 masters/equivilent events which is tied with connors and djokovic with most in a season.

it'd be like federer winning 5 masters THIS year :shock:

 

[travlerajm]

Thanks for the contribution. Much appreciated.

 

[THE FIGHTER]

great stuff!

it would be awesome to have different charts that showed the stats for each decade. perhaps we can see a shift from when young guys were winning majors regularly in the 80s to when it's become an anomaly for anyone under 20 to win a slam after 2005.

 

[Polaris]

Thanks for putting your time into this Falstaff. Very nice work and mature description. This thread should be "stickied" and linked to, whenever unreasonable posters with half a dozen usernames barge into threads and disparage any given player based on cherry-picked examples which are actually statistical anomalies.

 

[falstaff78]

Great statistics, really shows you how big of an anomaly Agassi was during his last years.

thanks a lot!

 

[falstaff78]

Great stuff. Thank you so much!!

you're welcome Matt!

 

[falstaff78]

Great thread, it is something most us fans knew of course but it's still nice to have it displayed in such format, might be an eye opener for some here (though I rather doubt it).

Hey Zagor - thanks for the kind words ! Means a lot

 

[Prisoner of Birth]

Great thread, one of the best I've seen. Basically reaffirms my stance that 22-26 is when most players have their primes. They generally hit their prime at around 22, with a fairly quick improvement, improve gradually till 25, drop fairly substantially at 26, and then continue to drop quickly. The fact that the graph is a progression till 25 and then drops is really telling. Shows you that it holds significance.

 

[sbengte]

Nice work , OP. Great thread to delve on some interesting data points.
This reaffirms that the absolute physical peak for a player in general is around ages 24-25. The sharp decline around ages 26-27 is quite telling.

Going by this, Fed's peak was roughly between 2005-2006 , Nadal's was around 2010-2011, Djokovic's (and Murray's) around 2011-2012. No surprise that these players had their best years and won the most around that time. No amount of inane arguments like Fed is playing better now than in his prime make any sense anyway, but this kind of data proves how absurd such talk really is. Fed's sharp decline was felt sometime around 2008 and it remains to be seen if the rest of the top 4 show the same pattern as they hit 26-27 years of age.

What would also be interesting is to see this distribution by majors. I wonder if any of the majors shows a different trend from the rest.

 

[doobiedoodoo]

Some Djokovic, Murray and Nadal fans will tell you 5-6 years makes no difference.......does this graph possibly suggest they are wrong??

 

[falstaff78]

Interesting stats and nice analysis, thanks for taking the time to put it all together.

I guess the question is:
Is Federer an outlier because his level has continued into his 30s...

or...

has he followed the normal path of decline, but his mid-20s level was so high that he is still a slam contender.

Hey Colin -

Thanks for the kind words! By the way I spend about half the week in Cambridge - are you a student?

Re: the discussion between you, Zagor and Tudwell.

The data would suggest that Federer is an outlier due to the LEVEL of his performance and not the SHAPE of the age-results profile. We can see that Fed's best years occurred precisely when the data suggested they would - i.e. August 03 - August 07.

The chart below disaggregates major victories by age for all 12 guys with >5 open era majors. Red highlights are late outliers, yellow highlights are early outliers. You can see the 5 important outliers to this trend in the entire open era.

 

[falstaff78]

Brilliant use of Data. Can't wait to see him reach 2-3 finals next year to offset the 32 year old data.

Thanks. I would certainly love that.... Eff the data!

 

[falstaff78]

Is there any chance of getting the values in the first graph expressed as percentages of total Open Era slams? I think that would make it even easier to compare the likelihood/difficult of winning a slam at various ages.

Here you go sir

 

[falstaff78]

Is there any chance of getting the values in the first graph expressed as percentages of total Open Era slams? I think that would make it even easier to compare the likelihood/difficult of winning a slam at various ages.

Strictly speaking the graph above is for majors AND season finales.

 

[falstaff78]

Thanks for the very interesting graphical presentation!

You are very welcome!

 

[ShoeShiner]

You should be a researcher.

​

 

[SLD76]

Facts and examples.

cant beat it.

 

[falstaff78]

Someone can correct me if I'm wrong here, but I was trying to work out the probability of winning a final, once it has been reached.

So there are 444 finalists, which means 222 winners, which means each of the winner bars should be half of the finalist bars.

Finalists at 24-25 = 64 + 64
Winners at 24-25 = 33 + 31
(exactly half)

Finalists at 31-32 = 8 + 2
Winners at 31-32 = 4 + 1
(exactly half)

Because the counts of 31 & 32 year olds are so low that there aren't enough data points, here are the combined counts for 30+ year olds:

Finalists at 31 & up = 24
Winners at 31 & up = 10
(just under half)

So what I think this means is that if someone in their 30s reaches a slam final (particularly their early 30s) then they are just as likely to win as they would have been at a younger age - about 50%. Otherwise, we'd see fewer winners in comparison to finalists at that age; less than 50%.

Does this mean that as players age, it isn't winning finals which is the challenge, but getting there?

Hey -

Great analysis! You may be onto something....sample sizes for the over 30s are a little small but I think there's an effect. If I stop being lazy at some point I'll eventually take a look at confidence intervals to see if this effect is statistically significant. in the meanwhile I ran the numbers by decade - here they are.

 

[falstaff78]

Someone can correct me if I'm wrong here, but I was trying to work out the probability of winning a final, once it has been reached.

So there are 444 finalists, which means 222 winners, which means each of the winner bars should be half of the finalist bars.

Finalists at 24-25 = 64 + 64
Winners at 24-25 = 33 + 31
(exactly half)

Finalists at 31-32 = 8 + 2
Winners at 31-32 = 4 + 1
(exactly half)

Because the counts of 31 & 32 year olds are so low that there aren't enough data points, here are the combined counts for 30+ year olds:

Finalists at 31 & up = 24
Winners at 31 & up = 10
(just under half)

So what I think this means is that if someone in their 30s reaches a slam final (particularly their early 30s) then they are just as likely to win as they would have been at a younger age - about 50%. Otherwise, we'd see fewer winners in comparison to finalists at that age; less than 50%.

Does this mean that as players age, it isn't winning finals which is the challenge, but getting there?

By the way numbers in above post are for MAJORS only! (all 180 of them since 1968 FO)

 

[falstaff78]

Thanks for the contribution. Much appreciated.

you are welcome!

 

[Bursztyn]

Hey -

Great analysis! You may be onto something....sample sizes for the over 30s are a little small but I think there's an effect. If I stop being lazy at some point I'll eventually take a look at confidence intervals to see if this effect is statistically significant. in the meanwhile I ran the numbers by decade - here they are.

It looks like in the 1960 there were relatively more winners and runner-ups in 30+ age group as compared with 1980-2010 time period.


One has to bear in mind however that the sample size was small for the 1960 because Open era started in 1968.

 

[falstaff78]

great stuff!

it would be awesome to have different charts that showed the stats for each decade. perhaps we can see a shift from when young guys were winning majors regularly in the 80s to when it's become an anomaly for anyone under 20 to win a slam after 2005.

Here you go sir. Frequency of reaching MAJOR finals ONLY, lumped in (1) 68-79, (2) then 80-89, (3) then 90-99 and (4) then 00-13.

I was a little surprised to see these numbers as I'm sure you will be. Essentially, reaching a major final as a teenager was a 1980s only phenomenon. It has happened 11 times in the open era - and 7 of those were in the 80s!

 

[falstaff78]

It looks like in the 1960 there were relatively more winners and runner-ups in 30+ age group as compared with 1980-2010 time period.


One has to bear in mind however that the sample size was small for the 1960 because Open era started in 1968.

Agreed. Essentially those numbers are driven entirely by Rosewall and Laver

 

[falstaff78]

Thanks for putting your time into this Falstaff. Very nice work and mature description. This thread should be "stickied" and linked to, whenever unreasonable posters with half a dozen usernames barge into threads and disparage any given player based on cherry-picked examples which are actually statistical anomalies.

Hey its a pleasure. It makes all the time worthwhile to receive comments such as these.

For me its amazing, because after inflicting these insights on my long-suffering wife and disinterested friends, I have now found an entire legion of nuts like myself who actually care about stuff like this!

 

[falstaff78]

Great thread, one of the best I've seen. Basically reaffirms my stance that 22-26 is when most players have their primes. They generally hit their prime at around 22, with a fairly quick improvement, improve gradually till 25, drop fairly substantially at 26, and then continue to drop quickly. The fact that the graph is a progression till 25 and then drops is really telling. Shows you that it holds significance.

Thanks so much for your kind words - as I just said, makes investing the time well worthwhile.

 

[falstaff78]

Nice work , OP. Great thread to delve on some interesting data points.
This reaffirms that the absolute physical peak for a player in general is around ages 24-25. The sharp decline around ages 26-27 is quite telling.

Going by this, Fed's peak was roughly between 2005-2006 , Nadal's was around 2010-2011, Djokovic's (and Murray's) around 2011-2012. No surprise that these players had their best years and won the most around that time. No amount of inane arguments like Fed is playing better now than in his prime make any sense anyway, but this kind of data proves how absurd such talk really is. Fed's sharp decline was felt sometime around 2008 and it remains to be seen if the rest of the top 4 show the same pattern as they hit 26-27 years of age.

What would also be interesting is to see this distribution by majors. I wonder if any of the majors shows a different trend from the rest.

Hey Sbengte - thanks a lot for the kind words.

Totally agree with your post - with one small exception - I think Murray will be a late bloomer. It's funny how history repeats itself - e.g. Nadal and Borg. I think Murray will end up falling more into the Lendl / Agassi mold where he will likely sneak out another 4-5 majors after turning 25.

Anyway here is the analysis you requested.

 

[falstaff78]

Nice work , OP. Great thread to delve on some interesting data points.
This reaffirms that the absolute physical peak for a player in general is around ages 24-25. The sharp decline around ages 26-27 is quite telling.

Going by this, Fed's peak was roughly between 2005-2006 , Nadal's was around 2010-2011, Djokovic's (and Murray's) around 2011-2012. No surprise that these players had their best years and won the most around that time. No amount of inane arguments like Fed is playing better now than in his prime make any sense anyway, but this kind of data proves how absurd such talk really is. Fed's sharp decline was felt sometime around 2008 and it remains to be seen if the rest of the top 4 show the same pattern as they hit 26-27 years of age.

What would also be interesting is to see this distribution by majors. I wonder if any of the majors shows a different trend from the rest.

And here are season finales