Round of 32 by age

[manley0702]

It'd be interesting if the first man born in the 1990s to win a Slam was born in the late 1990s. It's looking quite possible at this point.

At this point, I wouldn't be at all shocked in the years 89-92 never have a slam winner. That's 4 years. In my opinion, 4 years is really an entire generation, and that generation might go slamless. The older generation will keep winning in their prime, and by the time the older guys start to fade, the younger generation might block them from winning any.

 

[helterskelter]

At this point, I wouldn't be at all shocked in the years 89-92 never have a slam winner. That's 4 years. In my opinion, 4 years is really an entire generation, and that generation might go slamless. The older generation will keep winning in their prime, and by the time the older guys start to fade, the younger generation might block them from winning any.

That's plausible - and it might even go deeper into the 1990s, although I would agree that Thiem is likely to win a Slam at some point.

By the way, I believe that Gaston Gaudio is the only man born in any of 1977, 1978, and 1979 to win a Slam. So, those three years could - and probably should - have gone Slamless.

 

[helterskelter]

@manley0702

N.B. I'm not sure how you're defining "generation." In general, a generation is the period between birth and reproduction, so I would have thought that, by analogy, it is the period between emergence and maturation. But you're probably using it as the period between maturation and decline, in which case I would not agree that a four-year period is a full generation. I think players can expect to have significantly more than four years at the top of their game. There will, of course, be fluctuations of form, and I agree that players can't expect to have four years in a row on top form. But I think that many posters mistake loss of form for permanent decline.

 

[Thundergod]

It's also interesting to point out that in the first 3 slams this year, there's been only 2 SFists under 28, Dmitrov (25 at the time) and Thiem (23).

The last time there has been 2 SFists under 28 was 2015 AO (Djoko and Murray). There have only been 7 SFists(5 players) under 28 since then. That's the last 10 slams I think.

 

[JMR]

At this point, I wouldn't be at all shocked in the years 89-92 never have a slam winner. That's 4 years. In my opinion, 4 years is really an entire generation, and that generation might go slamless.

A tennis "generation" is typically defined as roughly five years, although naturally there are no predetermined beginning or ending points. But this term enables us to state, for example, that Federer is one generation before Nadal, two generations before Dimitrov, and about three generations before A. Zverev.

 

[helterskelter]

A tennis "generation" is typically defined as roughly five years, although naturally there are no predetermined beginning or ending points. But this term enables us to state, for example, that Federer is one generation before Nadal, two generations before Dimitrov, and about three generations before A. Zverev.

"Typically" meaning on here? What random posters on TTW think is really neither here nor there. Five years is too short. It should be more like 8-10.

 

[helterskelter]

It's also interesting to point out that in the first 3 slams this year, there's been only 2 SFists under 28, Dmitrov (25 at the time) and Thiem (23).

The last time there has been 2 SFists under 28 was 2015 AO (Djoko and Murray). There have only been 7 SFists(5 players) under 28 since then. That's the last 10 slams I think.

- One teen Slam finalist since the start of 1991 (Nadal at RG 2005).
- No Slam finalists younger than 24 since Australian Open 2011.
- One Slam finalist younger than 25 since Australian Open 2012 (Nishikori at US Open 2014).

 

[Jonas78]

I've done this for the last few Slams. Here it is this time:

21: Hyeon Chung, Karen Khachanov
22: Kyle Edmund
23: Jiri Vesely, Dominic Thiem, Lucas Pouille
24: Diego Schwartzman
25: Pablo Carreno Busta, Nikoloz Basilashvili
26: Milos Raonic, Grigor Dimitrov, David Goffin
27: Kei Nishikori, Steve Johnson
28: Juan Martin Del Potro, Marin Cilic
29: Albert Ramos-Vinolas, Roberto Bautista Agut
30: Andy Murray, Fabio Fognini, Richard Gasquet [turns 31 in June], Gael Monfils, Rafael Nadal [turns 31 on Saturday], Novak Djokovic
31: Pablo Cuevas, Kevin Anderson
32: Horacio Zeballos, Stan Wawrinka, John Isner
33: Fernando Verdasco, Guillermo Garcia Lopez [turns 34 on Sunday]
35: Feliciano Lopez

By age category:
Teenagers: 0
20-24: 7
25-29: 11
30-34: 13
35+: 1

Says more about weak era then. Which of the players play at a higher level post 30? Federer, Nadal, Djokovic, Murray, Berdych, Monfils, Gasquet etc all played at a higher level pre 30 than post 30. So very hard to see another explanation than lack of competition from the younger players.

I have seen the same misunderstanding when it comes to for example mens 100m running. Usain Bolt won the olympics at 30y, but he peaked many years earlier. He could never touch his own world records from as early as 2009.

 

[manley0702]

A tennis "generation" is typically defined as roughly five years, although naturally there are no predetermined beginning or ending points. But this term enables us to state, for example, that Federer is one generation before Nadal, two generations before Dimitrov, and about three generations before A. Zverev.

This was the definition I was using. Granted, it's arbitrary, but I think it's commonly accepted. I consider Nadal - Delpo/Cilic to be one generation. The next generation starts with guys like Nishikori and Raonic.

"Typically" meaning on here? What random posters on TTW think is really neither here nor there. Five years is too short. It should be more like 8-10.

Like I say, it's all arbitrary, but 8-10 years seems too long. The vast majority of players don't peak for 8-10 years. Historically, I'd say players have a peak of around 4-5 years. This is the period in which they're playing their best tennis before they slowly (or quickly) start to decline.

 

[JMR]

"Typically" meaning on here? What random posters on TTW think is really neither here nor there. Five years is too short. It should be more like 8-10.

No, not just here. In other contexts too, going back many years. I think the idea is that five years is about the length of a top player's proverbial "peak" period, so that if the age difference is five years or more, you won't have the two players at their peaks at the same time. Hence, different generations. Ten years would be more like an "era" than a generation, e.g., "the Lendl era." It includes more stages of a career, i.e., a couple years coming up, the so-called peak, and a couple years coming down.
Of course, none of this is officially defined anywhere, so ...

 

[helterskelter]

We are unlikely to convince each other!

But that said, I strongly believe that posters (here and perhaps elsewhere - I don't read Men's Tennis Forum or others) tend to misunderstand how "peak" is used by sports commentators and analysts. It's usually used here to refer to a continuous period, as you used it. But it's used by more analytic thinkers - I know that some posters think the average level of analysis is higher here than amongst commentators but I disagree: the level of knowledge here is high, the level of analysis is generally very low - to denote brief purple patches of excellent form, not a prolonged period or stage of a career. "Prime" would be used in the latter sense, but much more loosely. The problem with using "peak" in the sense used on the forum is that it simplifies our understanding of results, and makes them appear more closely determined by factors such as age than they actually are. Players lose form all the time. Certainly, nobody ever has maintained, or ever will maintain, good form for four or five years. But given fluctuations in form, many players have maintained a career at the top for eight or ten years, and it is possible and in fact frequent to get two players playing close to their best at the same time despite an age difference of much more than five years. For example, although Lendl was seven and a half years older than Becker, they were both very close to their best for much of 1989.

No, not just here. In other contexts too, going back many years. I think the idea is that five years is about the length of a top player's proverbial "peak" period, so that if the age difference is five years or more, you won't have the two players at their peaks at the same time. Hence, different generations. Ten years would be more like an "era" than a generation, e.g., "the Lendl era." It includes more stages of a career, i.e., a couple years coming up, the so-called peak, and a couple years coming down.
Of course, none of this is officially defined anywhere, so ...

 

[helterskelter]

This was the definition I was using. Granted, it's arbitrary, but I think it's commonly accepted. I consider Nadal - Delpo/Cilic to be one generation. The next generation starts with guys like Nishikori and Raonic.



Like I say, it's all arbitrary, but 8-10 years seems too long. The vast majority of players don't peak for 8-10 years. Historically, I'd say players have a peak of around 4-5 years. This is the period in which they're playing their best tennis before they slowly (or quickly) start to decline.

As I said in my reply just now, I don't think players have continuous peaks. I think peaks are brief moments that are discontinuous with each other. Nadal was on peak form in the spring and summer of 2008, but not for much of 2009, then close to it in 2010, etc. Peak is not best understood as a prolonged period, because doing so downplays the significance of inevitable fluctuations in form. By contrast, players are in their physical prime (the prolonged period during which it is possible to hit peak form) for much more than 4-5 years.

 

[helterskelter]

Says more about weak era then. Which of the players play at a higher level post 30? Federer, Nadal, Djokovic, Murray, Berdych, Monfils, Gasquet etc all played at a higher level pre 30 than post 30. So very hard to see another explanation than lack of competition from the younger players.

I have seen the same misunderstanding when it comes to for example mens 100m running. Usain Bolt won the olympics at 30y, but he peaked many years earlier. He could never touch his own world records from as early as 2009.

Tennis isn't usefully understood by analogy with running, because tennis is a much more complex sport, and the skills required are neither purely physical, nor stable over time. @Sysyphus has argued cogently that the skills required of contemporary tennis are becoming more like those of long-distance and endurance sports, which tend to favor players in their late 20s, rather than the quick explosive movements that favor players in their early 20s. This sort of change is very unlikely in a running event, because there aren't such drastic changes in strategy, technology, and training. I think Sysyphus is right to some extent, but I also think that going alongside those changes are changes to the structure of the tour that make it more difficult to break through, and perhaps some other changes, too.

By the way, the rise in age among tennis players goes well beyond the top few players. It is clearly visible across the rankings as a whole, which makes the explanation that you offered about lack of competition much less likely to be purely coincidental. You can look it up on Tennis Abstract, but there is also a journal article about it that Sysyphus shared a link to a while back.

 

[JMR]

We are unlikely to convince each other!

But that said, I strongly believe that posters (here and perhaps elsewhere - I don't read Men's Tennis Forum or others) tend to misunderstand how "peak" is used by sports commentators and analysts.

The wrong word is under scrutiny here. It doesn't matter whether you accept the traditional definition of "peak" or not (I find it flawed as well). The point is that the common usage of "generation" as about five years has been shaped by the conventional understanding of "peak." Even if that understanding is wrong, it doesn't change how "generation" has been traditionally used. This is descriptive lexicography, not normative.

 

[Jonas78]

Tennis isn't usefully understood by analogy with running, because tennis is a much more complex sport, and the skills required are neither purely physical, nor stable over time. @Sysyphus has argued cogently that the skills required of contemporary tennis are becoming more like those of long-distance and endurance sports, which tend to favor players in their late 20s, rather than the quick explosive movements that favor players in their early 20s. This sort of change is very unlikely in a running event, because there aren't such drastic changes in strategy, technology, and training. I think Sysyphus is right to some extent, but I also think that going alongside those changes are changes to the structure of the tour that make it more difficult to break through, and perhaps some other changes, too.

By the way, the rise in age among tennis players goes well beyond the top few players. It is clearly visible across the rankings as a whole, which makes the explanation that you offered about lack of competition much less likely to be purely coincidental. You can look it up on Tennis Abstract, but there is also a journal article about it that Sysyphus shared a link to a while back.

I agree tennis is much more complex than running, but my point was peak age. The list in the OP with 11 players at 25-29 and 13 players at 30-34 give a wrong impression, simply because players arent better at 30-34 than 25-29. This skewed list is because the extremely strong 85-88 generation and the historically weak 89-92 generation, and this will change in a few years when big4+ retires. When the 89-92 players (Nishikori and Raonic) are in their early thirties, your list will look different. It wont be like in the 80s and early 90s, when players mostly won slams in their early 20. This is probably because of a combination of several factors; nutrition, surgery, changes in playing style etc. But no way 30-34 will be the new peak age for most players.

 

[Gary Duane]

This discussion is all over the place.

You can still look at all the winners of majors in the Open era and get a pretty good idea of what is normal. Until recently a graph would show that the chances peaked at around 24-25. Of course you had players winning much younger and much older, but it was clear when guys on the tour were most likely to win.

The question is: has this permanently changed?

And the answer is clear: We don't know yet.

Fed, by himself, has not greatly skewed the results because for the most part he has not won majors since the age of 30. Of course he will push the average up a bit with the AO win this year, and if he wins Wimbledon, he'll push it up more, but that's only one player. We could say the same thing about Rosewall.

However, when you have several ATGs continuing to play at a very high level at 30 and later, all at the same time, that WILL skew the results for awhile.

I've looked at this whole matter again and again, from many angles.

There is a definite spike in age at the beginning of the OE, and that lasted roughly until the time of Connors, when it reversed.

We are right in the middle of another age spike, but this one is even more egregious and extends quite far down the rankings. If it continues this way, even to some degree, for the next 10 or 20 years, we will know that the change is most likely permanent.

 

[Gary Duane]

As I said in my reply just now, I don't think players have continuous peaks. I think peaks are brief moments that are discontinuous with each other. Nadal was on peak form in the spring and summer of 2008, but not for much of 2009, then close to it in 2010, etc. Peak is not best understood as a prolonged period, because doing so downplays the significance of inevitable fluctuations in form. By contrast, players are in their physical prime (the prolonged period during which it is possible to hit peak form) for much more than 4-5 years.

No, players do not have continuous peaks, but if you graph their results over a career in most cases you can see a clear pattern. Many people have talked about Fed from 2003 to 2007, and that's about right.

Just look at his performance timeline:

https://en.wikipedia.org/wiki/Roger_Federer_career_statistics#Singles[/QUOTE]

#1 for 5 of 6 years, then never again. Yes, he could mathematically get to #1 again this year, but he will never again be the same beast he was when younger.

Connors, same thing:

https://en.wikipedia.org/wiki/Jimmy_Connors_career_statistics#ATP_Tour_Singles_timeline

Look at his #1 years.

McEnroe, same:

https://en.wikipedia.org/wiki/John_McEnroe_career_statistics#Singles_performance_timeline

Sampras pushed it:

https://en.wikipedia.org/wiki/Pete_Sampras_career_statistics#Singles_performance_timeline

6 years.

And Lendl:

https://en.wikipedia.org/wiki/Ivan_Lendl_career_statistics#Singles_performance_timeline

 

[Gary Duane]

Duplicate post...

 

[helterskelter]

The wrong word is under scrutiny here. It doesn't matter whether you accept the traditional definition of "peak" or not (I find it flawed as well). The point is that the common usage of "generation" as about five years has been shaped by the conventional understanding of "peak." Even if that understanding is wrong, it doesn't change how "generation" has been traditionally used. This is descriptive lexicography, not normative.

But for the common usage of "generation" to have been shaped by the conventional understanding of "peak," the conventional understanding of "peak" has to be what you said it was. My argument wasn't just that the posters' definition of peak isn't a coherent one but also that it isn't the conventional one.

As for generation: you may be right that it's usually used to mean five years. I don't know. I don't hear analysts talk about it much. You are certainly right that, with regard to generation, my argument was only normative, not both normative and descriptive.

 

[helterskelter]

I agree tennis is much more complex than running, but my point was peak age. The list in the OP with 11 players at 25-29 and 13 players at 30-34 give a wrong impression, simply because players arent better at 30-34 than 25-29. This skewed list is because the extremely strong 85-88 generation and the historically weak 89-92 generation, and this will change in a few years when big4+ retires. When the 89-92 players (Nishikori and Raonic) are in their early thirties, your list will look different. It wont be like in the 80s and early 90s, when players mostly won slams in their early 20. This is probably because of a combination of several factors; nutrition, surgery, changes in playing style etc. But no way 30-34 will be the new peak age for most players.

My point is that, because tennis is more complex than running, there is no "peak age" that is stable across time. You are probably right that it's not going to be the case that 30-34 becomes the peak age. That I agree with. However, 25-29 is now and has been for some time clearly dominant over 20-24, but that wasn't always the case.

As an aside: let me reiterate that it is rather misleading to think solely in terms of top players and who is winning Slams, because the sample is too small to draw any firm conclusions. Yes, of course it is possible for a "generation" of players at the very top level to be weaker or stronger than another. But it is rather less likely that, if you look all the way down the rankings, there will be as much variance in natural talent.

Anyway, that said, I will myself indulge in looking at Slam winners to explain the point that the ideal age clearly changes over time:

Teenage Slam winners 1981-90: Wilander (Roland Garros 82), Wilander (Australian Open 83), Becker (Wimbledon 85), Edberg (Australian Open 85), Becker (Wimbledon 86), Chang (Roland Garros 89), Sampras (US Open 90). [And it wasn't just Slam winners: the rankings were littered with teens. Agassi never won a Slam as a teenager, but he was world #3 at 18].

Teenage Slam finalists 1991-2017: Nadal (Roland Garros 05).

There has been a clear and stable shift such that teenagers are no longer competitive, because the game has changed. If that can happen, it should be clear that, even if the concept of peak age is meaningful, it is not stable over time. I think it's neither, but that's another matter.

 

[helterskelter]

My interpretation of those data is rather different from yours. My sense is that for several of the players you discussed, with the partial exception of McEnroe, the peaks and troughs are actually much more discontinuous than you make them out to be.

For example, Sampras's best two years were clearly 1994 and 1997. In neither 1995 nor 1996 was his play at anything like the same level much of the time. (By the way, the rankings are also misleading here: his play was much better in 1999 than in 1998, but he was injured for much of 1999. Of course, as players get older, injuries are more likely - that's why the notion of physical prime is meaningful, even though the idea of a continuous peak is not).

As for Connors, it is obvious from even a cursory glance that he played better in 1982 than he had done in several years - and that his form fluctuated within his #1 years. (The rankings were very controversial back then).

But we have deeper philosophical disagreements here that I'll address in my response to your other post.

No, players do not have continuous peaks, but if you graph their results over a career in most cases you can see a clear pattern. Many people have talked about Fed from 2003 to 2007, and that's about right.

 

[helterskelter]

My question is not "has this permanently changed?" And I think that the fact that you think that is the question illustrates what I called just now the "philosophical disagreements" between us. I don't think that any set of conditions is ever likely to be permanent. What we are instead likely to see are continuous fluctuations over time. There will be some sort of patterns, but they won't be nearly so stable as the notion of a permanent change might suggest.

Calling this a "philosophical disagreement" might seem rather grandiose, but I don't think it is. (Bear in mind that I'm a professor of a discipline that is cognate to philosophy). Because what drives our differences of opinion is a difference in what we can expect to learn from data, or what we can take data to mean. At a rather trivial level, this shows up in my repeated claim that it is a mistake to think that we can learn anything from the list of Slam winners, because the numbers involved are so small that no big lessons can be learned. But you agree about that, and the disagreement goes far deeper than sample size. To put it in terms of philosophical label, your view is a "positivist" one and mine is "anti-positivist." (positivism)

In essence, that means that when I read your posts, it seems to me that underlying them is the view that there is a sort of natural order to tennis success of which the results are epiphenomena. Clearly, you acknowledge that the order could change - the post to which I'm responding makes that explicit. But what I'm suggesting is that the order doesn't exist in anything like the stable way in which you think it does. Tennis results are the product of a lot more than physical talent, so it's a mistake to think that we should expect to find a pattern to them that correlates with the physical abilities of particular age cohorts.

Anyway, I should stop procrastinating and do some actual work.

This discussion is all over the place.

You can still look at all the winners of majors in the Open era and get a pretty good idea of what is normal. Until recently a graph would show that the chances peaked at around 24-25. Of course you had players winning much younger and much older, but it was clear when guys on the tour were most likely to win.

The question is: has this permanently changed?

And the answer is clear: We don't know yet.

Fed, by himself, has not greatly skewed the results because for the most part he has not won majors since the age of 30. Of course he will push the average up a bit with the AO win this year, and if he wins Wimbledon, he'll push it up more, but that's only one player. We could say the same thing about Rosewall.

However, when you have several ATGs continuing to play at a very high level at 30 and later, all at the same time, that WILL skew the results for awhile.

I've looked at this whole matter again and again, from many angles.

There is a definite spike in age at the beginning of the OE, and that lasted roughly until the time of Connors, when it reversed.

We are right in the middle of another age spike, but this one is even more egregious and extends quite far down the rankings. If it continues this way, even to some degree, for the next 10 or 20 years, we will know that the change is most likely permanent.

 

[manley0702]

As I said in my reply just now, I don't think players have continuous peaks. I think peaks are brief moments that are discontinuous with each other. Nadal was on peak form in the spring and summer of 2008, but not for much of 2009, then close to it in 2010, etc. Peak is not best understood as a prolonged period, because doing so downplays the significance of inevitable fluctuations in form.

Nadal was pretty peak in 2009, he just got injured. I'd say 08-11 was his best form. The only exception being 2013, and I credit that to the fact that he learned to serve out of nowhere that year (then subsequently unlearned it).

By contrast, players are in their physical prime (the prolonged period during which it is possible to hit peak form) for much more than 4-5 years.

I don't think players are in their physical prime for much more than 4-5 years. It depends on the player, but I think most players peak around 24, then start to decline after 28. There are very few exceptions.

 

[helterskelter]

Nadal was pretty peak in 2009, he just got injured. I'd say 08-11 was his best form. The only exception being 2013, and I credit that to the fact that he learned to serve out of nowhere that year (then subsequently unlearned it).

I don't think players are in their physical prime for much more than 4-5 years. It depends on the player, but I think most players peak around 24, then start to decline after 28. There are very few exceptions.

You are, again, trying to read off "physical prime" from results, which is just a straightforward mistake. There are many reasons why players might do worse that have nothing to do with their physical abilities: loss of interest in the game, level of competition, failure to continue to adapt to new technology or changing strategy. It is also just not true that most players have a five-year level of continuously high play that they don't match before or after, and that that is almost always the period from 24-28. Many players have a six-month purple patch at 23, another one at 26, a return at 28, and so on. Only in a very general sense can we extrapolate to a continuous period, and when we do so, the generality tells us as much about us as observers as about the players as players.

Individual examples never tell us much, but think about Agassi: he was world #3 at 18. He had probably his best year of consistent play at 25, having at 23 slipped out of the top 20. Then at 27, he fell all the way to #141, before winning three of four Slams at 29. There is no way to understand his career trajectory without paying serious attention to things such as form, interest level, and dedication. Serena Williams has a similar arc.

By the way, if I were to accept your story about Nadal - I don't think he was any worse in 2012 than in the other years, but I do buy the claim about 2009 (although I think it was also loss of form, as well as injury, which supports my story more than yours) - his "peak" was from the age of 21 to 25, so he doesn't fit your general rule.

I'm also interested to compare to other sports. I don't follow enough to have a very informed opinion, but I can say that in snooker, admittedly far less physical a sport than tennis, the average age has risen massively in the last 15-20 years. And in soccer, the two best players in the world continue to be Cristiano Ronaldo (32) and Lionel Messi (30), and Luis Suarez (30) probably has a good claim to be third best.

 

[Gary Duane]

My interpretation of those data is rather different from yours. My sense is that for several of the players you discussed, with the partial exception of McEnroe, the peaks and troughs are actually much more discontinuous than you make them out to be.

I just linked to timelines showing when these guys reached #1 and how long they stayed there.

For example, Sampras's best two years were clearly 1994 and 1997. In neither 1995 nor 1996 was his play at anything like the same level much of the time. (By the way, the rankings are also misleading here: his play was much better in 1999 than in 1998, but he was injured for much of 1999. Of course, as players get older, injuries are more likely - that's why the notion of physical prime is meaningful, even though the idea of a continuous peak is not).

But you have to look at the whole picture. Yes, there are injuries, and there are peaks and valleys so far as zoned playing is concerned. But one of the characteristics of a period of dominance by the very best players is that they find a way to win when they have to "win ugly". So I would say ranking, number of majors and number of important titles per year is a good metric, and I don't see a period larger than around 5 years.

Sampras is complicated. He had health issues that might be better managed today, and I find everything about his career misleading. For instance, of all the ATGs I've looked at there is the most significant difference for him between the very biggest matches, mostly majors, and others. Then his stats are VERY high at the end of majors. So more than anyone else I've looked at closely he seemed to coast. It is as if he had an extra gear that no one else had.

As for Connors, it is obvious from even a cursory glance that he played better in 1982 than he had done in several years - and that his form fluctuated within his #1 years. (The rankings were very controversial back then).

But I still think in general that his #1 years together with his results in majors and together with the number of matches he played shows that his peak was somewhere in the early to mid 70s.

I don't know how old you are, so you may or may not have personal experiences with this period. By 79 80 and 81 Borg was sucking the air right out of the whole tour, with JMac then joining him. Borg was not fully gone until 82, so as I remember it there was a window for Connors to reinsert himself, and at that time he was doing a lot of what Nadal and Federer have done this year.

I'm not saying that ATGs can't have another amazing year or two near the end of their careers, but by that time they have to plan their schedules much more carefully, and they have to be much smarter.

But we have deeper philosophical disagreements here that I'll address in my response to your other post.

That's pretty deep for a superficial discussion of how long players are on top...

 

[Gary Duane]

My question is not "has this permanently changed?" And I think that the fact that you think that is the question illustrates what I called just now the "philosophical disagreements" between us. I don't think that any set of conditions is ever likely to be permanent. What we are instead likely to see are continuous fluctuations over time. There will be some sort of patterns, but they won't be nearly so stable as the notion of a permanent change might suggest.

Calling this a "philosophical disagreement" might seem rather grandiose, but I don't think it is. (Bear in mind that I'm a professor of a discipline that is cognate to philosophy). Because what drives our differences of opinion is a difference in what we can expect to learn from data, or what we can take data to mean. At a rather trivial level, this shows up in my repeated claim that it is a mistake to think that we can learn anything from the list of Slam winners, because the numbers involved are so small that no big lessons can be learned. But you agree about that, and the disagreement goes far deeper than sample size. To put it in terms of philosophical label, your view is a "positivist" one and mine is "anti-positivist." (positivism)

In essence, that means that when I read your posts, it seems to me that underlying them is the view that there is a sort of natural order to tennis success of which the results are epiphenomena. Clearly, you acknowledge that the order could change - the post to which I'm responding makes that explicit. But what I'm suggesting is that the order doesn't exist in anything like the stable way in which you think it does. Tennis results are the product of a lot more than physical talent, so it's a mistake to think that we should expect to find a pattern to them that correlates with the physical abilities of particular age cohorts.

Anyway, I should stop procrastinating and do some actual work.

I didn't know you were going to take this discussion to a "doctoral level".

I didn't realize that I needed to be very careful and precise when expressing a general idea, so let me reword:

What I read in this forum about "weak era" seems to suggest that the only reason older players are winning right now is because all the rest of the field is weak, and often the idea is that the rest of the field is at an all time low.

When I said a "permanent change" I was being very sloppy. What I meant is that it will take some time to analyze or understand what is happening right now and what seems to have happened for a few years.

If I'm right (only a guess at this point), what we are seeing right now may be an unusual year for older players, but the aging of the whole tour, along with the aging of the general population of the whole planet, is not going to suddenly reverse back to what it was a couple decades ago.

In other words, if we find out that the average age of the top 20 players remains higher than in the past, that the average age of major winners starts to move towards age 27 or 28 at its peak, I think that is significant.

So by "permanent" I meant long range as opposed to a very short period.

Now, if you are saying we can learn nothing from looking at the ages players have won majors for around 50 years, I disagree. We have already learned something.

Now, does this apply to this year? Obviously not. But if the average or mean - or whatever - obviously is higher for the next 50s years, then I think it means something.

 

[helterskelter]

@Gary Duane

I hope I didn't offend you, as it was certainly not my intention.

As you ask: I turned 39 a month ago, so, no, I don't have personal memory of Connors until what was clearly after his best years. I did watch his run at Wimbledon 1987 and much of his later career. Because I was only 13 at the time and living in the UK, and so it was late at night, I didn't get to watch much of his run at the US Open 1991, but I did follow it. A year later, my parents had had a loft conversion and so I was less supervised than I had been, so I did stay up to watch his loss against Lendl, which was his final Slam match.

 

[helterskelter]

@Gary Duane

I have previously said what my view is not, so let me now say what it is:

- I think there is a much longer time window during which players could be competitive than we had previously thought. At the very least, it seems as though players can compete at the very top between the ages of 20 and 35, and probably a couple more years either side of that. Now, that doesn't mean that there are no differences within that range, but within it, they are not absolute restraints as they would be at, say, 14 or 50. Some players will do better when they are younger, others when they are older, some more continuously, some more discontinuously. Importantly: the variations will not merely be idiosyncratic, but will linked to other developments in tennis and, indeed, in society as a whole. I absolutely agree with you that it is not coincidental that when people in many parts of the world are living longer than they did, sports stars are able to go on longer than they did. (Sadly, some portions of the USA are now starting to see a declining life expectancy again, though).
- Different eras will have different styles that favor different age ranges. I think it's almost inevitable that, in the future, there will once again be a period when younger players dominate, but that there will after that again be a period when older players do better again.

As to your point about domination, you might be right that it's very difficult for a player to continue dominating the tour for more than five years. That seems plausible. My suggestion - here I don't know whether I'm disagreeing with you or just with the other posters in this thread - is that there are non-physical reasons for that. A player can stop dominating without necessarily having declined physical abilities. It's really difficult to be the best in the world at something and all it takes is one player to come along, and you'll no longer be dominant. Had Monica Seles never taken up tennis, it's plausible that Steffi Graf would have dominated on and off (more on) for 10 years. Had Rafael Nadal never taken up tennis, we might now think that 2008 and 2009 were still dominant years for Roger Federer. Sure, he wouldn't have been as dominant in minor events as he had been in 2005 and 2006, but it was already obvious in 2007 that he couldn't sustain the interest in those events necessary to keep dominating them.

Plus there are things like loss of motivation that seem less prevalent now, but clearly hampered Borg, McEnroe, Wilander, Becker, Sampras in minor tournaments, and many players on the women's tour.

The rankings of lower-ranked players has always fluctuated a great deal. That's probably because it takes much less change in form to signal a great change in ranking at that level.

 

[Gary Duane]

I almost decided not to say anything except that I agree, because that's really the bottom line:

@Gary Duane
- I think there is a much longer time window during which players could be competitive than we had previously thought. At the very least, it seems as though players can compete at the very top between the ages of 20 and 35, and probably a couple more years either side of that. Now, that doesn't mean that there are no differences within that range, but within it, they are not absolute restraints as they would be at, say, 14 or 50. Some players will do better when they are younger, others when they are older, some more continuously, some more discontinuously. Importantly: the variations will not merely be idiosyncratic, but will linked to other developments in tennis and, indeed, in society as a whole. I absolutely agree with you that it is not coincidental that when people in many parts of the world are living longer than they did, sports stars are able to go on longer than they did. (Sadly, some portions of the USA are now starting to see a declining life expectancy again, though).

I'm very aware of the reversal of life expectancy in parts of the US, but I think that is linked to poor diet, too little exercise and horrible weaknesses in the present healthcare system. Add to that increased use of drugs. We have a minority of people who look like they plan to live forever, but others seems as though they don't care or can't care until it is too late.

I can't add much to what you said.

There is no reason to believe all adults hit a physical peak at the same age, and I think careful studies would show something very contrary. I looked like everyone's baby brother in high school, so unlike many kids that I went to school with I was probably a good five years behind in growth and physical maturation. I looked around 21 years old when I was in my mid to late 20s. This is common for my family. We all look like we are on a slower time table.

There is no reason why one man might not be at the same stage physically as another several years later or earlier.

So that's part of it.

Another is wear and tear on the body. Connors has had at least one hip operation, and I think two. I'm four years older than him, and my lower body is still fine. But I haven't (obviously) worn down my body with a competitive sport such as tennis, which is brutal.

So peaks are going to be different.

As for the time players stay on top, that's a bit more complicated. For most players I'd say that there is a clear high point, a kind of magical year. For Borg it was in the later 70s or so, but of course he quit, possibly long before his body was done. For Connors it was probably '74, close to then. He had a lot of good years, but in that year he was almost untouchable, with only four losses. One way to guess at peaks would be when the winning% is highest combined with the most matches.

For Fed it was probably 2005 or so. Nadal is tough because of his body. Djokovic probably peaked in 2011. Murry is hard to track. Wawrinka is strange - but perhaps he never came close to his potential earlier in his career.

For what I think of as "the peak" I would look for one year, then a window with that year in it.

But there is something else going on in tennis. Unlike things like swimming or running, maturity plays a huge role, so even though a player may hit a physical peak earlier, he may peak in terms of tactics, knowledge of the game and even technique at another time. We will know about Fed by tomorrow, but unlike most people here I feel like elements of his game have improved. I believe that he is a better player in many ways, but of course with his almost 36 year-old body that becomes a trade-off. I believe that if Fed could be given a magic pill that would restore his body back to what it was at around 24-25 and he could play his previous self, he would win most days.

This is where we get a player with a 2.0 career, later. I like to call it an "Indian summer". We saw it around 82 for Connors, we saw it in the early 2000s for Agassi, we saw it past the age of 30 for Rosewall, and we are seeing it right now for Fed.

- Different eras will have different styles that favor different age ranges. I think it's almost inevitable that, in the future, there will once again be a period when younger players dominate, but that there will after that again be a period when older players do better again.

I'm not going to be around 50 years from now to check. But if we could graph the average age of the top 20 players of the OE each year, either there will eventually be a gradual change in age, or there won't be. Right now we only have about 50 years to work with. Two majors have changed surfaces, and there has been a total change in equipment. No one playing with wood and guy could have anticipated modern rackets and strings, and we have only a bit more than 10 years to study how things have changed. If I am right - and I would not be one cent on being right - eventually there will be a clear movement of the average age of winning majors, but over time it will be much more gradual than we could assume from the past couple years.

It's one thing to look at the very top players and make assumptions, but the dominance of older players has now been going on a bit too long to be temporary, if I am right, and we have to pay attention to the number of guys who are winning late. I've lost track - Lopez, Karlovic, Mueller, Wawrinka, so many more. Long ago I remember Rosewall. Gimeno and Laver, but there were much younger players dominating other then. Right now it's not just Federer and a couple others. There are so many dominating, and meanwhile we are still waiting for Zverev, Thiem and other younger players to break through. What if they don't start playing at the very top of their games until 25 or past 25? What will our conclusions be if that happens?

As to your point about domination, you might be right that it's very difficult for a player to continue dominating the tour for more than five years. That seems plausible. My suggestion - here I don't know whether I'm disagreeing with you or just with the other posters in this thread - is that there are non-physical reasons for that. A player can stop dominating without necessarily having declined physical abilities.

Total agreement. Think of Wilander and Agassi. Rosewall may be the most convincing example. Burnout is going to be a huge factor.

I think one reason for longer careers right now is that playing is not as lonely as it used to be. When top players are traveling with friends, it's different. I don't recall Borg every having anyone like Moya around, and travel is so much easier for the top guys. They have so much money that they can almost live like royalty.

It's really difficult to be the best in the world at something and all it takes is one player to come along, and you'll no longer be dominant. Had Monica Seles never taken up tennis, it's plausible that Steffi Graf would have dominated on and off (more on) for 10 years. Had Rafael Nadal never taken up tennis, we might now think that 2008 and 2009 were still dominant years for Roger Federer. Sure, he wouldn't have been as dominant in minor events as he had been in 2005 and 2006, but it was already obvious in 2007 that he couldn't sustain the interest in those events necessary to keep dominating them.

I agree. There are no absolutes in tennis in terms of speed or strength. There is no way of knowing exactly how good Fed is now in comparison to what he was in 2005 because the players he faces are for the most part totally different. There are the usual "weak era" arguments, but other than that it is opinion and guessing. One or two matches a year will change the view of players. If, for example, Fed wins this afternoon how can people not say this is a near peak year with two majors? We can compare to this three major years, or look to 2009, but with only a couple majors in something like six years it's pretty hard not to think that he is playing better this year at least since 2009.

The rankings of lower-ranked players has always fluctuated a great deal. That's probably because it takes much less change in form to signal a great change in ranking at that level.

I don't think one or even a couple players will skew the age that much if we take the top 20 or top 30 players, and that's probably where we should be looking. What year was the average age lowest? Has it ever been higher than this year? How does age fluctuate over decades?

 

[helterskelter]

There was I thinking that, with two teenagers (Rublev and Shapovalov) and a 20-year-old (Coric) in round 3, the US Open was much younger than the other Slams this year. But I had ignored that there are FOUR men in it aged 35 or more! Even this year, no other major had more than two men in the last 32 who could have qualified for the veteran's event in days of yore.

Here's the total list. I'm assuming that Nadal finishes off Daniel, which of course he will.

18: Shapovalov
19: Rublev
20: Coric
22: Edmund
23: Thiem [turns 24 on Sunday September 3], Pouille
25: Dzumhur, Schwartzman
26: Goffin, Carreno Busta
27: Albot
28: Dolgopolov, Millman, Del Potro, Fabbiano, Cilic
29: Bautista Agut, Mannarino, Querrey, Kukushkin
30: L. Mayer, M. Zverev
31: Nadal, Troicki, Monfils [today, Friday September 1, is his 31st birthday], Anderson
32: Isner
33: Kohlschreiber
35: Lopez, Lorenzi, Mahut
36: Federer

So:
Teenagers: 2
20-24: 4
25-29: 14
30-34: 8
35+: 4

 

[stringertom]

There was I thinking that, with two teenagers (Rublev and Shapovalov) and a 20-year-old (Coric) in round 3, the US Open was much younger than the other Slams this year. But I had ignored that there are FOUR men in it aged 35 or more! Even this year, no other major had more than two men in the last 32 who could have qualified for the veteran's event in days of yore.

Here's the total list. I'm assuming that Nadal finishes off Daniel, which of course he will.

18: Shapovalov
19: Rublev
20: Coric
22: Edmund
23: Thiem [turns 24 on Sunday September 3], Pouille
25: Dzumhur, Schwartzman
26: Goffin, Carreno Busta
27: Albot
28: Dolgopolov, Millman, Del Potro, Fabbiano, Cilic
29: Bautista Agut, Mannarino, Querrey, Kukushkin
30: L. Mayer, M. Zverev
31: Nadal, Troicki, Monfils [today, Friday September 1, is his 31st birthday], Anderson
32: Isner
33: Kohlschreiber
35: Lopez, Lorenzi, Mahut
36: Federer

So:
Teenagers: 2
20-24: 4
25-29: 14
30-34: 8
35+: 4

Two guaranteed geezers in R16 unless fedr and FeLo both get rekt tomorrow by each other's walking canes!

 

[helterskelter]

Two guaranteed geezers in R16 unless fedr and FeLo both get rekt tomorrow by each other's walking canes!

In the bottom half, the third quarter is mostly old, while the fourth quarter is young. The two R4s in that quarter are an 18-year-old v a 26-year-old and a 23-year-old v a 25-year-old. In the third quarter, it's a 31-year-old v a 35-year-old, and likely to be a 29-year-old v either a 30-year-old or a 32-year-old.

 

[stringertom]

I was amazed with the ages of the QFists at the 1984 AO and I'm sure we'll never see this again:

1 @17
1 @18
1 @19
1 @20 (champion)
2 @22
2 @26

 

[helterskelter]

I was amazed with the ages of the QFists at the 1984 AO and I'm sure we'll never see this again:

1 @17
1 @18
1 @19
1 @20 (champion)
2 @22
2 @26

In the women's game, that wouldn't have looked particularly young back in the day. But, obviously, teenage girls are more competitive with fully grown women than are teenage boys with grown men and that's not going to change anytime soon.

 

[helterskelter]

Update after round 3. The last 16 are:

18: Shapovalov
19: Rublev
23: Pouille
24: Thiem [his 24th birthday is today, Sunday September 3]
25: Schwartzman
26: Goffin, Carreno Busta
28: Dolgopolov, Del Potro
29: Querrey
30: M. Zverev
31: Nadal, Anderson
33: Kohlschreiber
35: Lorenzi
36: Federer

So:
Teenagers: 2
20-24: 2
25-29: 6
30-34: 4
35+: 2

 

[jussumman]

So average age 29?

 

[helterskelter]

With Lorenzi out, Federer's the last man aged 35+ left standing.

With Shapovalov out, Rublev is the last teenager - and in fact the last man aged less than 24.

 

[helterskelter]

Update after round 4. The last eight are:

19: Rublev
25: Schwartzman
26: Carreno Busta
28: Del Potro
29: Querrey
31: Nadal, Anderson
36: Federer

Teenagers: 1
20-24: 0 (!)
25-29: 3
30-34: 2
35+: 1

On another note, unless Rublev, Schwartzman, or Carreno Busta takes home the title, we'll make it to 2018 before someone born in the 1990s wins a Slam.

 

[helterskelter]

As a result of Anderson's victory over Carreno Busta, it will be 2018 before someone born in the 1990s wins a Slam, and by the end of September 2017, every man who has ever won a Slam singles title will be 29 or more.

 

[ultradr]

You have to be over 30 to peak.

I have theory that it is due to : 1. Increased seeds (e.g.16-> 32 at slam) 2. Ranking system without bonus points for lower rankers beating significantly higher ranker. 3. Homogeneously slower surfaces.

All these changes (rather abrupt) in early 2000s, coincidentally the advent of "modern baseline era", Federer and Nadal, extended to domination of big 4 last 13 years or so.

 

[helterskelter]

Here we are again:

Men's singles at AO 2018:

20: Zverev, Rublev
21: Chung
22: Marterer, Kyrgios
23: Edmund
24: Thiem
25: Dzumhur, Schwartzman, Harrison, Basilashvili, Fucsovics, Kicker
26: Dimitrov, Carreno Busta, Sandgren
27: -
28: -
29: Dolgopolov, Cilic, Mannarino, Del Potro
30: Djokovic, Ramos-Vinolas, Fognini
31: Nadal, Gasquet
32: Tsonga, Berdych
33: Seppi
34: Muller
35: -
36: Federer, Benneteau
37: -
38: Karlovic

By age group:

Teenagers: 0
20-24: 7
25-29: 13
30-34: 9
35+: 3

Super weird that there's nobody left who is either 27 or 28, but there you go. Small sample at any one event.

 

[Harry_Wild]

The older you are in general; the higher ones ranking is!

 

[JRAJ1988]

I think Karlovic is one of the oldest players to reach the 3rd Round of the Australian Open since the 70's (Note - Just the Australian Open and ATP).

As for Benneteau, the last time he made the 3rd round of a slam was the US Open in 2013, I believe he's insinuated that he's going to retire this season but he's playing as well as he has in years .

 

[Zebrev]

If Federer wins AO he will probably play for another 3 years. We'll need a 40+ category soon!

 

[NaomiKonjuhPotapova]

Women's singles at AO 2018:

15: Kostyuk
20: Ostapenko, Osaka
22: Keys, Kontaveit, Mertens
23: Svitolina, Sasnovich, Pera
24: Garcia, Davis, Khumkhum, Allertova
25: Pliskova, Linette, Bogdan
26: Halep, Bertens
27: Wozniacki, Cornet, Martic
28: Radwanska
29: Rybarikova, Navarro
30: Kerber, Safarova, Sharapova
31: Strycova, Bondarenko
32: Kanepi, Hsieh

By age group:

Teenagers: 1
20-24: 12
25-29: 11
30-34: 7
35+: 0

28 only Radwanska.

 

[helterskelter]

Round 4:

21: Chung
22: [Kyrgios]
23: Edmund
24: Thiem
25: [Schwartzman], Fucsovics
26: Dimitrov, [Carreno Busta], Sandgren
27: -
28: -
29: Cilic
30: Djokovic, Fognini
31: Nadal
32: Berdych
33: [Seppi]
34: -
35: -
36: Federer
37: -
38: -

By age group:

Teenagers: 0
20-24: 4
25-29: 6
30-34: 5
35+: 1

(Yesterday's four losers included in brackets).

 

[helterskelter]

So, in 2018, younger players do seem to be making some headway. But the old guard are still hanging on in there, too, which means the middle players are getting squeezed. Assuming that Federer beats Struff, Raonic beats Novak, and McDonald beats Pella (the last could have gone either way but McDonald currently leads by a set and a break), then the top half of the men's draw will have five men aged 30 or more, one teenager, one 23-year-old, and one 27-year-old.

Here's the full round up for the last 32:

19: Tsitsipas, De Minaur
20: Tiafoe
21: Zverev
22: (Medvedev), Khachanov
23: McDonald, Edmund, Kyrgios
24: Novak, Vesely [turns 25 on Tuesday]
27: Raonic
28: Struff, Pella, (Albot), Nishikori
29: (Fabbiano), Gulbis, Del Potro, Paire
30: Mannarino, (Querrey), Eden
31: Monfils, Djokovic, Fognini
32: Anderson, Nadal
33: Isner, Simon
34: (Kohlschreiber)
36: Federer

By age bracket:
Teenagers: 2
20-24: 9
25-29: 9 [but nobody of either 25 or 26!]
30-34: 11
35+: 1

 

[tennis_pro]

Raonic can't age fast enough.

 

[helterskelter]

Round of 16:

19: Tsitsipas
22: Khachanov
23: McDonald
24: Vesely [turns 25 on Tuesday]
27: Raonic
28: Nishikori
29: Gulbis, Del Potro
30: Mannarino
31: Monfils, Djokovic
32: Anderson, Nadal
33: Isner, Simon
36: Federer

By age bracket:
Teenagers: 1
20-24: 3
25-29: 4 [but nobody of either 25 or 26!]
30-34: 7
35+: 1

So, 12 of the last 16 are 27 or more.

 

[helterskelter]

And here's the women's round of 16:

21: Ostapenko, Kasatkina, Bencic
22: Vekic
24: Sasnovich, Van Uytvanck
26: Pliskova, Bertens, Giorgi
29: Cibulkova, Goerges, Rodina
30: Kerber, Makarova
32: Hsieh
36: Williams

Teenagers: -
20-24: 6
25-29: 6
30-34: 3
35+: 1