Tennis Channel's "100 Greatest of All Time"

[mrzz]

Completely agree. A lot of those "titles" are little more than club house things. The very idea we can compare them to what these guys have to face up to now is laughable surely. And we're not even talking about depth. It is a completely different world now, and even if El Dude's well intentioned attempt to normalise across era's is accurate it simply won't be able to adjust for the vast difference in depth now. It literally is the difference between amateurs and professionals

I do think it is all about depth, but in a tricky way. People who are used to statistics (as most here are) will follow the argument easily. The difference between now and "then" is that now we have a league of outliers.

An outlier is that statistical point which scapes the general rule. Usually outliers are a very small fraction of the whole population.

Maybe in the past the 100 top guys playing the tournaments available were out of a population of -- wild guess -- 10.000 players. That´s the top 1%. The top 1% sample surely still "behave" as the whole population. They are the best, fine, but they are in the curve. Like the tails of a normal distribution.

Now we have the top 100 guys out of a population of 10.000.000 players world wide, that´s the top 0,001%, so we might be talking only about outliers. They are not in the curve. They are from another league.

The popularization of the sport, together with professionalism, and easier traveling also means that our top 100 guys are in general selected only by their abilities, while in the past other factors could have had more influence.

All this to say: Yes, I agree that small tournaments "back in the day" where far easier than now. The slams, however, still probably selected the top outliers, so I would not see much (relative) difference in this category.

 

[Federberg]

Now we have the top 100 guys out of a population of 10.000.000 players world wide, that´s the top 0,001%, so we might be talking only about outliers. They are not in the curve. They are from another league.

The popularization of the sport, together with professionalism, and easier traveling also means that our top 100 guys are in general selected only by their abilities, while in the past other factors could have had more influence.

All this to say: Yes, I agree that small tournaments "back in the day" where far easier than now. The slams, however, still probably selected the top outliers, so I would not see much (relative) difference in this category.

I agree with most of this. I would still argue though that the population sample in slams then were unlikely to contain the same relative quality that we see now. Statistically that would be very unlikely. That's the problem. I'm not trying to diminish the likes of Laver and Rosewall, I've watched them play and they look like top players. But it's very unlikely that the top hundred in a population of 10,000 will have the same relative strength as the top 100 of 10 million. That shouldn't be expected. Let's cherish those guys for the contributions they made to make the sport what it is today, but let's not kid ourselves that they would have necessarily been the best of the best in their era if the population was of an equivalent size to today

 

[mrzz]

^You are right. Too many players left out of the equation, so surely this makes the overall level lower. What can soften up a bit the situation is to assume that this 10.000 population we had back then is not completely random, but with some selection in itself, but even so surely a lot of good players are left out in this cut. So slams are the selection of the selection, and were the difference between eras should be the smaller one. But the difference is still there, I agree.

In Laver´s particular case, he seems to be so above the rest that I believe he would be an outlier in any case, but of course the eras comparison is a complicated one.

 

[Federberg]

^You are right. Too many players left out of the equation, so surely this makes the overall level lower. What can soften up a bit the situation is to assume that this 10.000 population we had back then is not completely random, but with some selection in itself, but even so surely a lot of good players are left out in this cut. So slams are the selection of the selection, and were the difference between eras should be the smaller one. But the difference is still there, I agree.

In Laver´s particular case, he seems to be so above the rest that I believe he would be an outlier in any case, but of course the eras comparison is a complicated one.

Yes I agree about Laver, but the problem is that that implies that his results were likely boosted by the weakness of the era. I detest the weak era theory in principle so I'm loathe to even sound like I'm diminishing his achievements. His achievements were what they were and shouldn't be subject to comparison in my view.

Now to this sentence you wrote..."So slams are the selection of the selection, and were the difference between eras should be the smaller one." You said it yourself here..."The popularization of the sport, together with professionalism, and easier traveling also means that our top 100 guys are in general selected only by their abilities, while in the past other factors could have had more influence." And the factors are obvious... societal changes have deepened the pools, so that other nations (Eastern bloc being the most obvious) are now involved when they weren't so much before.

 

[britbox]

I agree with you to an extent, but not fully. Let me explain.

I agree in that I see the Pro Slams as being somewhat similar to today's World Tour Finals: a smaller pool of players, but generally the best on tour. I count them as equal to Amateur and Open Slams even though there were fewer rounds because you only had three a year, and because of the high level of competition. Realistically I think the Pro Slams were equal to maybe a 70% share of an Open Era slam, and an amateur Slam was even less, maybe 60%. But if we make an adjustment like that we're unduly penalizing these players, because they could only compete in the tournaments available at the time.

Consider lso that Laver and Rosewall pretty much continued their Pro tour dominance in the early years of the Open Era, while the dominant amateur, Roy Emerson, declined quickly.

Laver really put a cap on it by winning all four Slams in 1969. And he wasn't beating stiffs at those Slams, either - some players being top amateurs who he had never or barely played in seven years. Consider some of the "names" he beat at each of those four Slams:
Australian Open: Andres Gimeno, Tony Roche, Fred Stolle, Roy Emerson
French Open: Rosewall, Okker, Gimeno, Stan Smith
Wimbledon: Newcombe, Ashe, Drysdale, Smith, Pietrangeli
US Open: Roche, Ashe, Emerson, Ralston

Poor Roy Emerson went from being the top amateur in 1967 to not being among the five best Open Era players, maybe barely top 10.

I

It's interesting but also complicated...

So we have Laver on an official 11 slam count. Most would argue his "value" would/should be more because he spent so many years on the Pro Tour that it impacted his official major numbers.

On the contrary, you could say his first 6 majors were bagged because he won them as an amateur, not on the pro tour. Would Laver have won those 6 majors on a combined tour? Considering how roundly he was spanked by the Pros when he first joined, you'd probably say no.

So you take with one hand and give with another.

But you also need to consider that once Laver started playing these guys regularly, he bridged the gap and became the best player. That opportunity was only available to those "invited" to partake.


...and some of these Pro Tour Slams only has 12 players in the entire field..... compare with 128 in the modern era and a lengthy qualification process... a pro tour slam is more like a YEC than a grand slam event. I'd value it in the same category.

 

[Ricardo]

I do think it is all about depth, but in a tricky way. People who are used to statistics (as most here are) will follow the argument easily. The difference between now and "then" is that now we have a league of outliers.

An outlier is that statistical point which scapes the general rule. Usually outliers are a very small fraction of the whole population.

Maybe in the past the 100 top guys playing the tournaments available were out of a population of -- wild guess -- 10.000 players. That´s the top 1%. The top 1% sample surely still "behave" as the whole population. They are the best, fine, but they are in the curve. Like the tails of a normal distribution.

Now we have the top 100 guys out of a population of 10.000.000 players world wide, that´s the top 0,001%, so we might be talking only about outliers. They are not in the curve. They are from another league.

The popularization of the sport, together with professionalism, and easier traveling also means that our top 100 guys are in general selected only by their abilities, while in the past other factors could have had more influence.

All this to say: Yes, I agree that small tournaments "back in the day" where far easier than now. The slams, however, still probably selected the top outliers, so I would not see much (relative) difference in this category.

i think tennis playing population would be more like 100 million, so the top 100 guys are in fact all one in a million talent. Every time some posters here remark 'mediocre players' (ranked 50-100 mostly) i can only sigh....you are the best out of a million and people say you are 'mediocre', and those people probably can't get past club level.

 

[mrzz]

Well, I completely guessed the number of 10 million... but, wow! 100 million? That´s roughly one person out of 70 that plays tennis... Do you found that figure somewhere or it is also a guess?

Completely agree with your last remark. We were talking about that a while ago (not sure exactly which thread, but it is recent). I get that it is all relative, but even so...

 

[El Dude]

Well again, I wouldn't dispute that the quality of tennis now is (probably) much higher than it was back in the day. This is true of probably every sport, as the pool of players deepened. But this is circling back to arguing something that no one is arguing against - mainly because it is both logically sound but not provable; that the quality of tennis players generally improves over time. We certainly see this in pure athletics, where--for instance--sprinters beat their own records, eventually someone else comes along to surpass them. Usain Bolt is the fastest recorded sprinter ever, but inevitably someone will be faster.

This does ignore the basic approach I advocate, which is to compare cross-generally through looking at relative dominance. The only way to really do that is to compare rankings. Titles change, the number and quality of tournaments - all of that is relative to an era. As I pointed out about, Laver won 22 titles in 1962 - an absurd feat by today's standards. But he did so through playing 38 tournaments, perhaps even more absurd. We just can't compare. And this also goes for Slams, whether pro, amateur, or Open. We just cannot compare; but we also can't penalize older players for playing (and winning) at the only tournaments available to them.

As an aside, one thing to support Laver's greatness is the fact that he conquered the amateur tour, then moved onto the pro circuit and after a year or so of adjustment, dominated that. Then when the Open Era started, he dominated that. So it didn't matter the context - he rose to the top, and was the best. And look at the 1969 Slams - he faced the best of the best of that era, and still won.

Anyhow, if we can agree that tennis generally improves over time, but that it is ultimately impossible to compare across broad generations, the only way to really assess historical greatness is relative dominance against one's peers - that is, against the context in which one actually played. And given that we can't really compare tournaments, it is also pointless to compare Slam and title counts. This leaves us with nothing...or does it? Actually, we have rankings. We only have ATP rankings from 1973 on, but we do have unofficial rankings before then: we have sportstwriters giving their subjective opinion going back to 1877 on the #1 and sometimes #2 players, and we have a site like TennisBase.com that actually uses a detailed formula that gives us rankings for all of tennis history. As far as I know, there isn't another site or source like TB.

Now if we say, "so what?" We're back where we started: we can't compare players across very separate eras, or at least before and after the Open Era. So there's no point in discussing it, because it is just throwing around subjective opinions that will ultimately get us nowhere. But if we're willing to say, "OK, maybe we can at least compare relative greatness across eras through looking at TB's rankings, which probably give us our best window into the past."

Again, this doesn't deny that tennis players improve over time. But it does say that the definition of "historical greatness" is not measuring how good every tennis player ever was on the same court at the same time - which is both impossible to do, and also unfair to older players, because they adapted to the context they played in - but rather, it is comparing relative greatness. And the only way to do that in a somewhat accurate way, without getting bogged down into trying assess the relative strengths of tournaments, is to look at rankings only. And to do that, TB is our best bet.

(We could also look at what the sportswriters had to say, but then we get into subjectivity and politics; it is similar to MVP awards in baseball..the sportswriters often didn't vote for players they didn't like out of spite).

Anyhow, I'll come up with a ranking of the greatest players of all time based upon TB's year-end rankings. You can also look at their ranking list, but they take into account other data - titles, Slam results, etc. I'll be teasing out just the rankings for purity's sake.

 

[Federberg]

^I still think this misses the point... we can't make a direct comparison between the titles that Laver was winning in the 60s with the ones these guys are winning now. I very much doubt that they had as many rounds as the current ones do for a start. But a simple common sense query will confirm it. Are we really suggesting that the likes of Novak or Federer at their peak would ever be able to play 38 tournaments in one year? No way no how, and that tells you right away that we're talking about a different sport. I really think it's a waste of time to try too hard to manufacture an equivalence between then and now. Full credit to the old greats for what they did, but it's not the same as what the guys of the last couple of decades are doing... not even close

 

[mrzz]

but inevitably someone will be faster

Translating it to tennis, I admit that it is the general rule to assume that, but I do believe there is a boundary to human performance, much as in skill sports as tennis as there is in "pure" athletics like sprint races. It is a long discussion, a lot of factors come in to play, but I do believe we reached the ceiling in tennis. I do not think in 50 years people will be saying that the current top players then would come here and crush the big 4. This is not the point of the discussion, but I wanted to raise it. As you know, I am against the current most of the time...

Again, this doesn't deny that tennis players improve over time. But it does say that the definition of "historical greatness" is not measuring how good every tennis player ever was on the same court at the same time - which is both impossible to do, and also unfair to older players, because they adapted to the context they played in - but rather, it is comparing relative greatness

I think this is a fine point, and I agree with it completely. I hear Federberg´s remarks, but I do not think they are completely in disagreement with your general way of thinking, as I put in the next reply.

Full credit to the old greats for what they did, but it's not the same as what the guys of the last couple of decades are doing

Physically, yes. Technically, also yes, but with some considerations. Modern technology made the game improve, physical conditioning made the game improve, so, yes, tele-transport Laver to our days and, yes, he gets beaten. But you could turn the argument the other way around: the guys back then managed to play with far worst physical training, and, maybe even more important, with far worst equipment. I would like to see today´s pros trying one of those old racquets... of course they would be able to play, but I bet you can forget all those 20+ shots rallies. So, yes, it is really not the same. Of course I still believe modern players are, in a word, better, I only think it is not that simple.

And there is also the known fact that the old greats elevated the sport. This is not just a pretty phrase. To change the way a sport is played is an immense thing. A talented guy now only need to go through the motions. What some guys in the past did is of a completely different caliber. Again, I know this is not exactly the point, but...

 

[El Dude]

OK, I'm a bit baffled, Federberg. You restated what I just said about not being able to compare tournaments across era - how am I missing that point? I said that above and clearly fully agree with you.

But what I am saying, which you did not respond to, is that the only thing we can compare with any degree of accuracy is rankings - which determine relative dominance, and in terms of relative dominance they work pretty well. While we can probably agree that being #1 in 2017 is more impressive and difficult than being #1 in 1962, it is still the best player in the world. If Laver was the best player in 1962, how can we view that as lesser than being #1 today relative to the context he played in? He had no way to be better, so if we want to compare cross-generationally, we have to give at least similar merit as far as relative dominance goes.

And again, what I am saying that if we compare across generations, the only way to do so in a meaningful way is to focus on relative dominance.

To put it another way, rankings are somewhat absolute. #10 is tenth best in the world, regardless of year. The tournaments make up the components of the ranking, and change over time and thus are harder to compare. But rankings are more solid in that they are always relative to the field.

 

[Federberg]

OK, I'm a bit baffled, Federberg. You restated what I just said about not being able to compare tournaments across era - how am I missing that point? I said that above and clearly fully agree with you.

But what I am saying, which you did not respond to, is that the only thing we can compare with any degree of accuracy is rankings - which determine relative dominance, and in terms of relative dominance they work pretty well. While we can probably agree that being #1 in 2017 is more impressive and difficult than being #1 in 1962, it is still the best player in the world. If Laver was the best player in 1962, how can we view that as lesser than being #1 today relative to the context he played in? He had no way to be better, so if we want to compare cross-generationally, we have to give at least similar merit as far as relative dominance goes.

And again, what I am saying that if we compare across generations, the only way to do so in a meaningful way is to focus on relative dominance.

To put it another way, rankings are somewhat absolute. #10 is tenth best in the world, regardless of year. The tournaments make up the components of the ranking, and change over time and thus are harder to compare. But rankings are more solid in that they are always relative to the field.

Perhaps I misread I do agree that rankings are the one thing that are truly dimensionless. I got stuck on your comment regarding the number of titles Laver won in one year. I just don't rate those tournaments in comparison to the present for my stated reasons. It's worth noting that using @mrzz's point about the total population of tennis players then compared to now, even rankings are a sketchy comparison. After all it's far more impressive to be ranked number 1 in China for than Laos for example. But it's as good as we can make it as long as we comprehend it's limitations. Apologies for the misread

 

[El Dude]

Translating it to tennis, I admit that it is the general rule to assume that, but I do believe there is a boundary to human performance, much as in skill sports as tennis as there is in "pure" athletics like sprint races. It is a long discussion, a lot of factors come in to play, but I do believe we reached the ceiling in tennis. I do not think in 50 years people will be saying that the current top players then would come here and crush the big 4. This is not the point of the discussion, but I wanted to raise it. As you know, I am against the current most of the time...

That's a very interesting thought and discussion. On one hand, I disagree in that humans always seem to push forward, even if it is only in micro-improvements, perhaps mainly because later players have the benefit of building upon what has been previously accomplished.

On the other hand, I can agree with you that there are physical ceilings. I think I read somewhere that the human arm couldn't physically throw beyond a certain speed. Not sure why, it just can't happen. Yet. Maybe in larger time spans - say, hundreds of years - the physical body will adapt.

But as far as reaching the ceiling in tennis, I kind of question that. I mean on one hand I think we're seeing the best tennis ever played by Federer, Nadal, and Djokovic. Yet on the other, why couldn't someone come along who is even better? Who evolves certain aspects of the game?

Regardless, the speculative nature of these questions are why I think all-time rankings and discussions around GOAT should always be relative in nature - that is, based upon how good a player was in the context they played in.

[

 

[Federberg]

As it happens I do often wonder about just how much faster the current sprinters are to the ones in the past. I remember when I was in my teens, I used to be a sprinter. I remember about a decade after giving up the sport and going to watch a meet. I had the chance to check out the track. I reckon at my best I would have been able to extract half a second from the improved track and also the superior spikes. It's not clear to me how much better Usain Bolt is to say... Carl Lewis at their respective peaks. Certainly not as much as their personal bests would suggest. I do think Bolt is the greatest we've ever seen.. his dimensions, stride length and freakishly good starts for such a big man tell me that must be so. But... is a Tyson Gay or a Justin Gatlin better than Carl Lewis... I'm very suspicious

 

[El Dude]

Perhaps I misread I do agree that rankings are the one thing that are truly dimensionless. I got stuck on your comment regarding the number of titles Laver won in one year. I just don't rate those tournaments in comparison to the present for my stated reasons. It's worth noting that using @mrzz's point about the total population of tennis players then compared to now, even rankings are a sketchy comparison. After all it's far more impressive to be ranked number 1 in China for than Laos for example. But it's as good as we can make it as long as we comprehend it's limitations. Apologies for the misread

I was using Laver's titles as support as to why you can't compare titles and tournaments across generations. Maybe I wasn't clear enough about that.

But yeah, I hear your point about China vs. Laos. In the end, we don't know how the #1 player from Laos would compete vs. the #1 player in China. Comparing before the Open Era to now is even more difficult (and pointless) because not only do we have different pool sizes, but entirely different contexts of play in other ways - as mrzz pointed out, re: training and other factors.

But again, one can only be judged according to context.

One other bit. While I agree that the game and players improve over time, I think it is relatively slight. We can see this in the fact that top players stay competitive over time and across sequential eras. Roger Federer beat Agassi and Sampras in the limited time he played them, but the matches were overall competitive. Even when Roger was in his prime, a 34-year old Agassi took him to five sets - at the 2004 US Open, for instance, when Roger was 23. Or we have the benefit of Youtube and can watch a prime 20-year old Borg vs. a 37-year old Rod Laver at Hilton Head in 1976. Clearly Borg was better at that point, but Laver held his own and we can't help but wonder, what would the match have looked like if Laver had been 10 years younger?

 

[El Dude]

As it happens I do often wonder about just how much faster the current sprinters are to the ones in the past. I remember when I was in my teens, I used to be a sprinter. I remember about a decade after giving up the sport and going to watch a meet. I had the chance to check out the track. I reckon at my best I would have been able to extract half a second from the improved track and also the superior spikes. It's not clear to me how much better Usain Bolt is to say... Carl Lewis at their respective peaks. Certainly not as much as their personal bests would suggest. I do think Bolt is the greatest we've ever seen.. his dimensions, stride length and freakishly good starts for such a big man tell me that must be so. But... is a Tyson Gay or a Justin Gatlin better than Carl Lewis... I'm very suspicious

Yeah, it is hard to say. We can see this in baseball, where the fastest recorded pitch ever was by Aroldis Chapman in 2010 at 105.1 mph. Nolan Ryan, who was the fastest thrower of the 70s, was recorded at 100.8 at his fastest, but they measured the speed at a different location; evidently if they used today's method, that pitch would have been 108 mph!

 

[Obsi]

The top guys I've spoken to have a ridiculous grasp of tennis history. Their views are worthy of respect in my opinion

Look at this Tennis Channel's list. The top guys put Roy Emerson above Pancho Gonzales. None with high level knowledge of tennis history considers Roy greater than Pancho.

 

[Federberg]

I was using Laver's titles as support as to why you can't compare titles and tournaments across generations. Maybe I wasn't clear enough about that.

But yeah, I hear your point about China vs. Laos. In the end, we don't know how the #1 player from Laos would compete vs. the #1 player in China. Comparing before the Open Era to now is even more difficult (and pointless) because not only do we have different pool sizes, but entirely different contexts of play in other ways - as mrzz pointed out, re: training and other factors.

But again, one can only be judged according to context.

One other bit. While I agree that the game and players improve over time, I think it is relatively slight. We can see this in the fact that top players stay competitive over time and across sequential eras. Roger Federer beat Agassi and Sampras in the limited time he played them, but the matches were overall competitive. Even when Roger was in his prime, a 34-year old Agassi took him to five sets - at the 2004 US Open, for instance, when Roger was 23. Or we have the benefit of Youtube and can watch a prime 20-year old Borg vs. a 37-year old Rod Laver at Hilton Head in 1976. Clearly Borg was better at that point, but Laver held his own and we can't help but wonder, what would the match have looked like if Laver had been 10 years younger?

That's definitely true. I always wonder how much match up and the psychology of playing against your hero has to do with those sorts of matches. The match up thing applies when you consider how easily Sampras was always able to tune Mac for example..

 

[Federberg]

Look at this Tennis Channel's list. The top guys put Roy Emerson above Pancho Gonzales. None with high level knowledge of tennis history considers Roy greater than Pancho.

I'm not defending the list in anyway. I generally don't place much utility in those lists. I merely make the point that it's absurd (to me) to disregard what top players think. The idea they don't have a passion for the sports history isn't backed up by my experience.

While I'm ready to accept the consensus that Gonzales is a better player, it's clear that they are basing their list on slam counts

 

[El Dude]

I wish there was some way we could quantify these things - how much of a match is ability level, match-up dynamics/context (including court type), and psychology. I'm guessing something like 50/25/25, but there's no way to support that. It might be more accurate to look at it as starting with a baseline of ability level, then adjusting with "modifiers" for other factors - match-up, court, psychology, etc.