128 players of all time, as selected by tennis abstract

U

urban

Legend
I recently found an uncomplete list of 128 players of the last century on tennisabstract. com. Jeff Sackmann has tried to select 128 male and women players and write articles on them all. Now the project has reached the places around Nr. 50. Its certainly an interesting project to remember some of the greatest members of the hall of fame and others, beyond the range of the big 3 of todays tennis. As far as i can see, the articles are written with expertise on the basis of literature and stats. Problem is certainly the ranking order. Maybe it would be better to give this well researched articles without a ranking order. I have never been a fan of a combined ranking of men and women, too much difference especially in the level of competition. And i cannot understand why a player with the historical significance of Jean Borotra, who was ranked by a guy like Dan Maskell among the top ten of all time, is here at Nr. 123, behind Kei Nishikori, Vitas Gerulaitis or Del Potro. And Lew Hoad at Nr. 74? Come on. At least in the article the author mentions, that Lew was selected by several different experts as best of all time. He certainly deserves better.
I don't want to be over-critical here, because i like well studied researches, and i can see the merits of this project. The articles seem to be better than the ranking order.
 
T

thrust

Legend
urban said:
I recently found an uncomplete list of 128 players of the last century on tennisabstract. com. Jeff Sackmann has tried to select 128 male and women players and write articles on them all. Now the project has reached the places around Nr. 50. Its certainly an interesting project to remember some of the greatest members of the hall of fame and others, beyond the range of the big 3 of todays tennis. As far as i can see, the articles are written with expertise on the basis of literature and stats. Problem is certainly the ranking order. Maybe it would be better to give this well researched articles without a ranking order. I have never been a fan of a combined ranking of men and women, too much difference especially in the level of competition. And i cannot understand why a player with the historical significance of Jean Borotra, who was ranked by a guy like Dan Maskell among the top ten of all time, is here at Nr. 123, behind Kei Nishikori, Vitas Gerulaitis or Del Potro. And Lew Hoad at Nr. 74? Come on. At least in the article the author mentions, that Lew was selected by several different experts as best of all time. He certainly deserves better.
I don't want to be over-critical here, because i like well studied researches, and i can see the merits of this project. The articles seem to be better than the ranking order.
Click to expand...
It seems the rankings are more favorable to the open era players. Where is the list?
 
TheFifthSet

TheFifthSet

Legend
I wish the Open Era bias was its only flaw. Wawrinka has as many majors as his contemporary Nishikori has Semi’s+, yet is ranked seven spots behind him.

I suppose it’s meant to be a fun list that doesn’t try too hard to quantify these things.
 
boredone3456

boredone3456

G.O.A.T.
MJ Fernandez ahead of Halep?
Austin ahead of Goolagong?
all the other things already pointed out by everyone else?

Predicting Serena will be #1. Will be interesting to see where the big 3 fall...and where Sampras lands
 
U

urban

Legend
In the introduction the author writes, that he uses an Elo system for his ranking. But the ranking order seems indeed sometimes weird and not based on any numbers. Its a bit sad, because the articlles themselves are pretty carefully written. I found the concept of the book of Rex Bellamy, Love Thirty, from 1990, where he tells the stories of the 30 best players of his lifetime, without any ranking, much more convincing.
 
S

Scott_tennis

Rookie
Thank you Urban for pointing out this website and project. I admire Jeff Sackman's effort and overall intent, though the order of players does seem to raise some eyebrows. The essays associated with each listed player appear to be well-researched and thoughtful. Perhaps he should have simply listed the essays in alphabetical order without the ranking. In that respect it reminds me of a Tennis Channel production years ago which featured short video features of top players and was also criticized for the order of players.
 
S

Scott_tennis

Rookie
For what it is worth, the Bellamy book covered 30 players from 1960 to 1990 and seems to present the essays in somewhat of a chronological order. I think his intent was to list the thirty most memorable players during that span. In the introduction he mentions that while there were no chapters devoted to Hoad or Bueno, although those two players were mentioned in other chapters.

The list of chapters/essays is as follows:

1 Gonzales
2 Rosewall
3 Emerson
4 Santana
5 Laver
6 Jones
7 Court
8 Ashe
9 King
10 Newcombe
11 Wade
12 Kodes
13 Nastase
14 Panatta
15 Goolagong
16 Vilas
17 Connors
18 Evert
19 Borg
20 Navratilova
21 McEnroe
22 Lendl
23 Mandlikova
24 Mecir
25 Wilander
26 Edberg
27 Becker
28 Graf
29 Sanchez Vicario
30 Chang

(20 men and 10 women)
 
U

urban

Legend
Yes. Bellamy's chapters are chronologically structured, and he gives fine accounts on other players like Hoad or Gimeno in certain chapters. Some chapters of his book are now available on the internet, i had pointed in other threats to those of Laver and Santana. Bellamy could really write, fo my money the best tennis writer of all tennis journalists. And there were quite a few good writers in those days, like Collins, Gray, Tingay, Pingeon, Tomassi, Barrett, Danzig, Wilson, Grimley, Amdur, Deford Wind, Trengove and mayn others. As longtime tennis correspondent of The Times, Rex saw a lot of tennis in his life. I once had the opportunity to speak shortly to him at the World Team Cup at Düsseldorf.
 
Wander

Wander

Hall of Fame
Jeff's list now has only 31 players left to be revealed, and I think that by this point it has finally become easy to determine exactly who the final bunch will be. Only the order is a mystery. I must say that Jeff's Elo-based formula throws you some wicked curveballs, but you have to consider that this kind of a mathematical formula doesn't know or care about the existence of things like titles, let alone the extra significance people tend to place on Major championships. It's a purely mathematical calculation based on the player's win-loss record and the quality of those opponents as based on the very same Elo calculations. Someone beating Steffi Graf or Novak Djokovic is as big an accomplishment for the purposes of this formula regardless of whether it happened in the final of a Grand Slam or in the 3rd round of a 500.


Anyway, the list of players revealed so far is as follows:

128. Beverly Baker Fleitz
127. Stan Wawrinka (podcast)
126. Jean Borotra
125. Li Na
124. Betty Nuthall
123. Michael Stich
122. Ashley Cooper
121. Angela Mortimer
120. Kei Nishikori
119. Adrian Quist
118. Bill Johnston
117. Darlene Hard
116. Ted Schroeder
115. Rosie Casals
114. Andrea Jaeger
113. Karel Koželuh
112. Shirley Fry
111. Goran Ivanišević
110. Frank Kovacs
109. Anita Lizana
108. Molla Mallory
107. Jim Courier
106. Sarah Palfrey Cooke
105. Petra Kvitová
104. Vinnie Richards
103. Tony Roche
102. Jadwiga Jędrzejowska
101. Ashleigh Barty
100. Dorothy Round
99. Tom Okker
98. Zina Garrison
97. Frank Parker
96. Elena Dementieva
95. Vitas Gerulaitis
94. Kitty McKane Godfree
93. Simona Halep
92. Gottfried von Cramm
91. Ann Jones
90. Caroline Wozniacki
89. Michael Chang
88. Mary Joe Fernández
87. Juan Martín del Potro
86. Margaret Osborne duPont
85. Svetlana Kuznetsova
84. Lleyton Hewitt
83. Jack Crawford
82. Maria Esther Bueno
81. Budge Patty
80. Andy Roddick
79. David Ferrer
78. Simonne Mathieu
77. Henri Cochet
76. Pam Shriver
75. Virginia Wade
74. Lew Hoad
73. Elizabeth Ryan
72. Stan Smith
71. Tony Trabert
70. John Bromwich
69. Nancy Richey
68. Manolo Santana
67. Mary Pierce
66. Vic Seixas
65. René Lacoste
64. Bobby Riggs
63. Ora Washington
62. Amélie Mauresmo
61. Ilie Năstase
60. Frank Sedgman
59. Evonne Goolagong
58. Pancho Segura
57. Louise Brough
56. Tracy Austin
55. Roy Emerson
54. Jana Novotná
53. Hilde Krahwinkel Sperling
52. John Newcombe
51. Hana Mandlíková
50. Mats Wilander
49. Helen Jacobs
48. Arthur Ashe
47. Jennifer Capriati
46. Victoria Azarenka
45. Conchita Martínez
44. Jaroslav Drobný
43. Guillermo Vilas
42. Althea Gibson
41. Doris Hart
40. Stefan Edberg
39. Kim Clijsters
38. Andre Agassi
37. Fred Perry
36. Maria Sharapova
35. Pauline Betz
34. Ellsworth Vines
33. Justine Henin
32. Boris Becker
31. Gabriela Sabatini


www.tennisabstract.com

Introducing The Tennis 128 - Heavy Topspin

Scroll down or click here for the list of players published so far. * * * You know what tennis really needs? More arguments about the greatest players of all time. Really! I could take or leave the Djokovic-Federer-Nadal debate, and I don’t need to read another word about Serena versus Margaret...
www.tennisabstract.com
 
Wander

Wander

Hall of Fame
crimson87 said:
Sabatini better ranked than Vilas? Doesnt make much sense.
Click to expand...
Not just Vilas, also Becker, Edberg, Agassi, Henin, Sharapova and so on and so forth.

But if you read the article, it explains a little bit how this kind of thing can theoretically be justified. In this case the formula rates Sabatini so high mainly because of how well she did against Graf between 1990 and 1992.
 
U

urban

Legend
I am a big fan of Gaby, but this makes no sense, with Maria Bueno at Nr. 82. I thought, Gaby would have been forgotten, like Guga Kuerten or Sergi Bruguera or Thomas Muster, who are nowhere to be found. It seems indeed, that is is better here follwing this Elo-system, to beat a Nr. 1 than to be a Nr. 1.
 
Wander

Wander

Hall of Fame
urban said:
I am a big fan of Gaby, but this makes no sense, with Maria Bueno at Nr. 82. I thought, Gaby would have been forgotten, like Guga Kuerten or Sergi Bruguera or Thomas Muster, who are nowhere to be found. It seems indeed, that is is better here follwing this Elo-system, to beat a Nr. 1 than to be a Nr. 1.
Click to expand...
Well, it depends on the #1. Wins over Graf were very valuable because her rating was usually so high compared to anyone else.

This kind of formula really doesn't care for surface specialists which is why Guga, Bruguera and Muster don't make the top 128. If you look at the performance timeline of a player like Sabatini, she was very consistent for a very long time, which this kind of a mathematical formula likes. It doesn't matter that she didn't win that many big titles if she made a lot of semi-finals all season long for many years. Players like Muster, Kuerten and Bruguera were never like this.

Again, remember that for this type of formula it would actually be better to get to 10 finals but lose all of them than to win five tournaments but lose the rest of the time in the 1st round.
 
U

urban

Legend
I see and understand the arguments, but something is wrong with the whole for,mula and the ranking order. Guga had also some vital wins and finals on hardcourt, including a year end final win vs. Agassi indoors, which secured his Nr. 1 position for the year A Ferrer is in, a Del Porto, or Roddick or Nishikori, but Guga is obviously out of the overall 128, if he would not placed in the top 15 male, which is certainly too high. Muster won 12 titles in 1995 alone, beat Nr. 1 Sampras indoors, got to Nr. 1, won 40 titles overall on clay and hardcourt, and still is completely out..

On the other hand, Boris Becker, one could call him a fast court specialst, is higher placed than contemporary Andre Agassi, which is unusual in most rankings, although Agassi has the more rounded and consistent record on all surfaces, has more titles (60-49), has a 10-4 in hth and was ATP Nr. 1 much longer.
 
Last edited:
TheFifthSet

TheFifthSet

Legend
urban said:
I see and understand the arguments, but something is wrong with the whole for,mula and the ranking order. Guga had also some vital wins and finals on hardcourt, including a year end final win vs. Agassi indoors, which secured his Nr. 1 position for the year A Ferrer is in, a Del Porto, or Roddick or Nishikori, but Guga is obviously out of the overall 128, if he would not placed in the top 15 male, which is certainly too high. Muster won 12 titles in 1995 alone, beat Nr. 1 Sampras indoors, got to Nr. 1, won 40 titles overall on clay and hardcourt, and still is completely out..

On the other hand, Boris Becker, one could call him a fast court specialst, is higher placed than contemporary Andre Agassi, which is unusual in most rankings, although Agassi has the more rounded and consistent record on all surfaces, has more titles (60-49), has a 10-4 in hth and was ATP Nr. 1 much longer.
Click to expand...

Sackmann is obsessed with ELO and Becker’s superior Top 10 record probably gives him the bump over Agassi on his list.

Misses the forest for the trees, I think.
 
Wander

Wander

Hall of Fame
Third Serve said:
that’s insane lol
Click to expand...
It's not insane at all when you consider the parameters that the list is made with. Like I said earlier, the algortihm does not even "know" that such things as titles exist. It simply treats every single match as an individual event and makes a calculation of how good a player was based on the full list of their professional matches and their outcomes as individual events.

Nishikori has a career win percentage of 67.1%. Kuerten's is 64.7%. That may not seem like a big difference but for a purely Elo based calculation it's clearly enough to skew things Nishikori's way.

In a podcast I listened to some time ago Jeff said that depending on how he weighed the calculation, certain players who just missed could've made it in (obviously). In fact, Daniil Medvedev would already make it somewhere in the top 120 right now, but he just barely missed the cut at the beginning of the year. Of course also players like Swiatek and Alcaraz are on a trajectory to make the cut soon and so on and so worth. I think he said Thomas Muster could've made it in with a different weighing. I don't remember anything about Kuerten.
 
Wander

Wander

Hall of Fame
bigbadboaz said:
You're only making it sound more insane.

The list is billed as "128 greatest players", not "Tennis ELO #s 1-128".
Click to expand...

Jeff's argument is that Elo is a better metric than the number of Majors or weeks spent at #1 or whatever else. You can disagree with that premise, but I don't think it's completely insane.

Other people have come up with other mathematical calculations that are based on a combination of different things.

https://www.ultimatetennisstatistics.com/goatList is one such list for men's tennis which might be more agreeable to a lot of people since it heavily weighs big tournament titles and particularly Grand Slams in its metric, but even so it comes with its own issues. The primary one being that it only considers tennis after 1968, aka the Open Era.

Since Jeff's system only considers matches themselves independently from tournaments, it can compare pre-Open and Open era players much more seamlessly than systems that are based on weighing different titles at different values. This is its greatest strength, and I think it's the first list I've seen that allows us to see how Suzanne Lenglen or Bill Tilden or Ken Rosewall (and so on and so on) stack up to the modern Greats of the game according to an algorithm that treats both equally since its independent from the constraints of different eras.

Maybe most people don't see any value in that, but I find it well worthwhile and can overlook something like it feeling a bit funny that Nishikori is ranked above Wawrinka (and Kuerten, Muster etc.) since I understand how this position can be mathematically defended.

To understand this list you just have to pretend that there's a reality where tennis is not about the number of Grand Slam titles. For the purposes of this list, Grand Slams (and everything else) might as well just be 250 tournaments with some extra rounds since that's how the algorithm treats them. After you can accept that, everything starts to make more sense.
 
U

urban

Legend
I would strongly disagree, that we can compare eras alone on the basis of match percentage stats. The structure of the circuit is to be reckoned with. We had that am-pro split, and on the pro tour since the 1930s, the main circuit consisted of a one on one, mano a mano match series between two top players or a small tournaments system, both with over 100 matches for the top players. Result was, that the top player had to play a top ten player virtually every day. This structure hampered his match winning percentage, even if he was the best player of the day. Result was, that the top amateurs had way better percentages than the top pros, even if the pros were better at that time.

Look at am and pro champion Ellie Vines. In 1932 as top am he had a 90% or so match win percentage, but in 1937, when he was playing better, he had only a little over 50 %, when all his matches came in a close hth series with Fed Perry. That says very little about his real form and level. Or Tilden, who had a 93% winning percentage over his 10 amateur years in the 1920s. Over his pro career in the 1930s, with more hth tours, his percentage sank to a low 60% or so.

Pancho Gonzalez, the pro king of the 1950s, had his best year in 1956, when he played around pro 150 matches, but his percentage was only a low 67% or so (too lazy to look it up). he had way better percentages in a year, when he was playing as amateur or when he was playing a lighter schedule as a pro. Rod Laver had the best percentages of the pros in the years 1964-1967, aound 78 % or so, but he won over 100 matches in each of the years. He won over 100 matches in 8 of his season as am, pro or open era player. I always perfered for seasons evaluation the difference between match wins and losses over the percentage of wins. The more matches you will play, the more you will lose.

Result is further, that tennis abstract places among 1950s players, Drobny, Seixas (both amateur only players), Sedgman and Trabert all over Lew Hoad (Nr. 74 on the list), although Hoad was the only one, who could really challenge Gonzalez in the 1950s and had the best pro record, and the best combined pro and amateur record of them all. That is simply not true.

Imagine for a moment, if in the 2010 years, Nadal, Djoker, Federer, Murray (and say Zverev, Thiem, Wawrinka, and Raonic) would have played a structure like the old pro days, with a tour of small tournaments inclusive round robins with say 8 top players, and would compete every day and week against each other, at least personal 20 hth matches against each other on every surface, Nadal vs. Djoker, Djoker-Fed, Fed-Nadal, Murray- Djoker and so on. They would never reached 80% match wins under those cicrumstances., and nevertheless they had to play a higher level over this virtual circuit than under ATP conditions.

And a last note: If the match win percentage is decisive for the elo ranking, why is Jean Borotra, who has the best male career match win percentage with 83, 7% since 1920, is ranked only at Nr. 127?
 
Last edited:
Wander

Wander

Hall of Fame
@urban

Well, the point is that it isn't just about win percentage stats (and I never meant to claim that it was) because Elo is far more sophisticated than that. It is able to distinguish between opponents of different quality so that Pancho Gonzales going 67% or whatever against top competition will be theoretically valued just as high (or higher) by the system as going 87% against weaker (amateur) competition. This is why the algorithm has also managed to rate Borotra at #127 while Pancho Gonzales and Bill Tilden are yet to appear on the list.

If you read Jeff's articles he fully acknowledges the difference between the amateur and pro circuits in the pre-open era and even talks about how Elo sometimes struggles to accurately rate these players since the system was only bleeding players one way (from amateur to pro). Some pretty good players never turned professional and are as a result difficult to fully assess.

Jeff's list even features Ora Washington as the strongest black female player from before black people were allowed to even try to play in the Majors, but her actual level in real terms is of course impossible to really measure since she was mostly playing against players nobody knows anything about since they never played any matches against the stars of the day. In all likelihood Ora wasn't actually good enough to be winning Majors at the time (she was also playing basketball and working full-time) but writing an article about her tennis exploits is nevertheless a laudable effort for some historical perspective of the time.

But anyway, I am fully aware of the issues you brought up and my point is that Elo is actually the best system out there for ranking these players across eras exactly because it rates the quality of opponents and even recognizes the relative differences in the strengths of different eras.
 
U

urban

Legend
Problem is, that Tennis abstract gives no Elo numbers for the pre open era players, like Vines, Segura, Perry, Hoad and so on. So why Murray is ranked above Perry, has here no basis in numbers. At least, as far as they are published.

The Ora Washington article has great merits for cultural history, but i would not rank her above Maria Bueno.
 
NatF

NatF

Bionic Poster
Wander said:
@urban

Well, the point is that it isn't just about win percentage stats (and I never meant to claim that it was) because Elo is far more sophisticated than that. It is able to distinguish between opponents of different quality so that Pancho Gonzales going 67% or whatever against top competition will be theoretically valued just as high (or higher) by the system as going 87% against weaker (amateur) competition. This is why the algorithm has also managed to rate Borotra at #127 while Pancho Gonzales and Bill Tilden are yet to appear on the list.

If you read Jeff's articles he fully acknowledges the difference between the amateur and pro circuits in the pre-open era and even talks about how Elo sometimes struggles to accurately rate these players since the system was only bleeding players one way (from amateur to pro). Some pretty good players never turned professional and are as a result difficult to fully assess.

Jeff's list even features Ora Washington as the strongest black female player from before black people were allowed to even try to play in the Majors, but her actual level in real terms is of course impossible to really measure since she was mostly playing against players nobody knows anything about since they never played any matches against the stars of the day. In all likelihood Ora wasn't actually good enough to be winning Majors at the time (she was also playing basketball and working full-time) but writing an article about her tennis exploits is nevertheless a laudable effort for some historical perspective of the time.

But anyway, I am fully aware of the issues you brought up and my point is that Elo is actually the best system out there for ranking these players across eras exactly because it rates the quality of opponents and even recognizes the relative differences in the strengths of different eras.
Click to expand...

Elo is a good predictor for matches between peers in the same era but that's it. It's a system designed to be used in chess. It's often slow to adjust and inflates over time. The sheer number of odoties it comes up with should show that:

Towny said:
Elo seems to be one of the more popular ratings systems used on this board, especially by certain Novak apologists. It has been used to determine not only peak level but also the level of competition faced by a given player. While I'm not saying the system is total garbage, I thought it'd be interesting to look at a few anecdotal findings of the system to show just how inherently flawed it is. I'll be looking at the Big 3 for this. All data from Ultimatetennisstatistics.com

Federer
Feb 2nd 2004 - 2303
Feb 3rd 2014 - 2321
Immediately following his respective AOs, Fed of 2004 held 2 slams and TMC and was world number 1. Fed of 2014 had just had his awful 2013 season and was now world number 8. 2014 Fed was considered better. For year end Elo ratings, let's compare 2004 to 2014 (3 slams, YEC and 3 masters compared to 0 slams, 0 YEC and 2 masters)

Nov 22nd 2004 - 2420
Nov 16th 2014 -2408
Apparently fairly comparable years.

Nadal
Jan 31st 2011 - 2379
Feb 2nd 2015 - 2383
In the former, Nadal had just won 3 slams and was world number 1. In the latter, Nadal had just one slam to his name (RG) and was world number 3 after being beaten by Berdych at AO. Let's compare the Nadal that Berdych faced in 2015 to the Nadal that Djokovic faced in 2012:

Jan 16th 2012 - 2377
Jan 19th 2015 - 2397
Apparently, Berdych faced the stronger Nadal

Djokovic
September 12th 2011 - 2522
January 30th 2017 - 2516
One of them had just won 3 slams and 5 masters. The other had 1 slam and 4 masters to his name, and had just been bounced by Istomin. In fact, prior to facing Istomin, on January 16th 2017, Novak's rating was 2538, better than it had ever been in 2011

For further comparison, Novak at his lowest Elo in 2012 (August 12th) had 2431
Novak following his garbage 2017 had 2417 (Dec 1st 2017), only 14 points less

There's no real method to my madness but it's simply to illustrate that totting up 'list of players with ???? Elo points or more beaten' is a waste of time, as the system sometimes can't even tell the difference between beating 2013/4 Federer and beating 2004 Federer. But then, what can you expect from a system designed to be used in Chess?

Discuss
Click to expand...

Also IIRC Murray in 2009 had more Elo points than Sampras at any point in his career. It's pretty bad at cross era comparisons IMO.
 
Wander

Wander

Hall of Fame
urban said:
Problem is, that Tennis abstract gives no Elo numbers for the pre open era players, like Vines, Segura, Perry, Hoad and so on. So why Murray is ranked above Perry, has here no basis in numbers. At least, as far as they are published.

The Ora Washington article has great merits for cultural history, but i would not rank her above Maria Bueno.
Click to expand...
You'd have to ask Jeff why his Elo ratings aren't available on the website for pre-open Era stuff. My guess would be that it's because the match records are largely incomplete (especially the further back you go), which makes it difficult to be completely accurate. With many players for the smaller tournaments we don't even always know who they played in the early rounds.


NatF said:
Also IIRC Murray in 2009 had more Elo points than Sampras at any point in his career. It's pretty bad at cross era comparisons IMO.
Click to expand...

At Tennis Abstract at least for the year end rankings Murray in 2009 was at 2238 Elo. Sampras' year end Elo was above that at the end of 94 (his highest year end Elo of 2274) and also at the end of 95, 96 and 97 so I'm not sure that's true. Andy did have a higher peak Elo of 2347 in early 2017 which is at least defensible since he went on a tear at the end of 2016 and had a higher win% that year than Sampras did in any single season of his career. (Again, to make it clear I know that Elo is not just about win percentage, but at least with players like Sampras in the 90s and Murray in the 2010s we can compare these numbers relatively well since the tour structure and match numbers were similar)
 
NatF

NatF

Bionic Poster
Wander said:
At Tennis Abstract at least for the year end rankings Murray in 2009 was at 2238 Elo. Sampras' year end Elo was above that at the end of 94 (his highest year end Elo of 2274) and also at the end of 95, 96 and 97 so I'm not sure that's true. Andy did have a higher peak Elo of 2347 in early 2017 which is at least defensible since he went on a tear at the end of 2016 and had a higher win% that year than Sampras did in any single season of his career. (Again, to make it clear I know that Elo is not just about win percentage, but at least with players like Sampras in the 90s and Murray in the 2010s we can compare these numbers relatively well since the tour structure and match numbers were similar)
Click to expand...

I guess TA has a different formula to others I've seen. Even so Murray and Sampras competed in completely different fields with rather different surfaces, really don't think we can compare. Tennis is a zero sum game, we can only ever say whether one player is more dominant in his field than another - but that leaves out the question of the relative quality of the fields. Also Murray's higher win percentage leaves out that he only won a single slam versus Sampras winning multiple (three in a row from 1993-1994 as well).
 
Wander

Wander

Hall of Fame
Of course it's ultimately still just an exercise in hypotheticals, or as one might call it, just a bit of fun. We also know that in absolute terms, David Goffin is probably a better tennis player than Bill Tilden ever was. But relative to their eras Tilden is an all-time great while Goffin is just an another pretty good current player.

But what I don't understand is why people should think that Jeff's Elo formula is much worse than any other system for ranking players.

As I pointed out earlier, usually when people make lists based on some mathematical formula, they have to start at the beginning of the Open Era to avoid the eternal amateur tour vs pro tour problem. Whatever problems you might think Elo has, at least it sidesteps that issue better than other mathematical formulas. You will see that Roy Emerson is ranked #55 - somewhere in the top 30 for men - with his 12 (pretty weak era) amateur Slams. Still a great player, but that ranking seems more appropriate than ranking him in the top 10 or whatever just because of the absolute number of Majors.
 
U

urban

Legend
I assume, that Tilden, Budge or Gonzalez and Kramer are high on this tennis abstract list, not because of Elo rankings, but more because of common sense. That Emmo isn't top ten, i agree with, i have him around Nr. 20 male. But he, here at Nr. 55, is certainly not worse than Jaroslav Drobny, who is ranked Nr. 44, 11 places higher. Both remained amateurs (Emmo turned pro in 1968), but in most departments, majors (12 to 3), big tournaments, great seasons, Nr. 1 positions, variety of surfaces, Davis Cup play (ok, Drobny suffered here in exile), Emmo has the far better amateur record. Drob won a lot of titles, but most on the lucrative, but pretty light Riviera and Near East clay tours.

Cross era rankings are always hard to prove with the last certainty, you find no definite, dogmatic answer. I find more problematic the TA rankings of contemporaries, say Seixas, Segura, Drobny over Hoad, or Ashe over Newcombe, or Becker over Agassi, when almost all parameters point to the different direction.
 
NatF

NatF

Bionic Poster
Wander said:
Of course it's ultimately still just an exercise in hypotheticals, or as one might call it, just a bit of fun. We also know that in absolute terms, David Goffin is probably a better tennis player than Bill Tilden ever was. But relative to their eras Tilden is an all-time great while Goffin is just an another pretty good current player.

But what I don't understand is why people should think that Jeff's Elo formula is much worse than any other system for ranking players.

As I pointed out earlier, usually when people make lists based on some mathematical formula, they have to start at the beginning of the Open Era to avoid the eternal amateur tour vs pro tour problem. Whatever problems you might think Elo has, at least it sidesteps that issue better than other mathematical formulas. You will see that Roy Emerson is ranked #55 - somewhere in the top 30 for men - with his 12 (pretty weak era) amateur Slams. Still a great player, but that ranking seems more appropriate than ranking him in the top 10 or whatever just because of the absolute number of Majors.
Click to expand...

I think when you end up with results like Nishikori over Courier, or Ferrer over Courier, Roddick, Hewitt etc...Murray over Agassi, Becker, Edberg, Wilander, Becker over Agassi etc...among many other strange rankings (to say nothing of the older ones urban has picked up on) there's a problem with the system. Measuring greatness can't be done with a single metric and despite the trend towards "objectivity" and bean counting, subjectivity is a part of these debates at least in how we weigh up all the multitude of factors.

I like Jeff, having spoken to him a little bit through my participation in the match-charting project, and I understand these aren't necessarily even his own rankings as much as what his algorithm has come up with. I just think think this whole excercise, fun as it is and well written as the articles are, just highlights the problems with Elo and this numbers only approach.

I do however think it's good that it includes pre-open era players as any platform which draws attention to these past greats is good - I just think the algorithm could use updating :p
 
Wander

Wander

Hall of Fame
I believe that Nishikori, Ferrer and Murray end up surprisingly high because Elo thinks that the early 2010s was an extraordinarily strong era. I think subjectively speaking, tennis fans tend to agree, but they don't translate that opinion into thinking that Nishikori and Ferrer were equivalent or better than someone like Courier who won a bunch of Slams at an opportune time, because they see 4 Slams and peak ATP ranking of #1 next to someone's name and that trumps everything.

I think that you believe that there has to be some Elo inflation involved, but seeing that Elo thinks current era is much weaker than the era 10 years ago (it for example rates current Novak Djokovic over 200 Elo points lower than Djokovic of 10 years ago although I think some kind of inactivity penalty may be applied to his current numbers - I'm not entirely sure), the numbers can clearly go both ways so who's to say?
 
U

urban

Legend
BJK at Nr. 28, behind Sanchez-Vicario at Nr. 27. Not many would rank BJK so low, given her status and legacy in America and Wimbledon. Bud Collins would turn in his grave, i assume. The ranking is bold, to say the least.
 
King_olaf_the_hairy

King_olaf_the_hairy

Semi-Pro
Just out of curiosity... does this tennis Elo thing, whether it's Jeff Sackmann's, UTS's, or whoever's... does it take account of different surfaces?

An obvious example I can think of would be Thomas Muster in 1995. What was his Elo after he won Estoril, Monte Carlo, Rome, Roland Garros, and St Polten (37 consecutive clay wins)? UTS says he peaked at 2287 in October that year. Yet he didn't even enter Wimbledon, and had he done so would likely have lost in R1 like he normally did. Whoever drew him would have been playing an opponent who was 2250+ on paper, but was likely playing at a 1750-level on grass.

Elo is a chess thing. It works for that game because no matter where you are in the world, it's still the same 32 pieces on the same 8x8 grid. That can't apply to tennis. Not even today, never mind pre-homogenization when you had surface specialists.
 
U

urban

Legend
Even with Elo as parameter, i have doubts about the accuracy. Look at Fed Perry. Between late 1933 and 1936, he won about everything. Not only 8 majors, and the full house of all 4 majors, what was sensational given the travel conditions. He also won the British hard (clay) court 5 times, in importance the Monte Carlo of the day, and the Cannes clay tournament as well, also the South West Pacific at LA on hard court several times, which was the preemeninent hard court event in those days, and the Berkeley South. Pacific on hard court. His DC record 1933-36 was immaculate. He beat on his way greats like Vines (in DC), Budge, von Cramm, Riggs, Quist, Crawford, you name them. He must have had a higher Elo than for instance a Andy Murray, who never had that kind of record over several years.
 
Last edited:
M

MissileBackHand99

Semi-Pro
This ranking system is a dumpster fire. And just goes to show that there is no "objective", analytics-based formula to accurately calculate the history of tennis. Tennis' history is too varied, too non-uniform, etc to have one way of comparing players and eras. It's always gonna be subjective (which makes convos about it much more fun in my opinion)

Also, the obsession sports and sports fandom has with pure analytics these days...is truly exhausting
 
U

urban

Legend
In the standard tennis rankings by US writers, BJK was usually ranked very highly, often above Margaret Court, with many political arguments mixed within the debate She certainlly has more stadiums and more books behind her name, than any other tennis women player. Here Bille Jean at Nr. 28, is ranked behind "all time winners" Sanchez Vicario and even Davenport. Now, thats certainly difficult to comprehend and much too low..
 
King_olaf_the_hairy

King_olaf_the_hairy

Semi-Pro
Coming to the home straight now. I'd be astonished if the remaining 23 aren't the twelve men and thirteen women I expect, although I wouldn't fancy my chances of ordering them in agreement with the author...

I reckon we should start to prepare "top ten" lists of our own to see who's best at predicting Sackmann's Folly™, i.e. what you think the author will choose, rather than who you believe yourself. Post them when he gets to #11 (which I think will be on November 20th?).

Come on now, for all its faults, this project has done exactly what JS wanted: have a lot of eyeballs on his site, gazing in horrified wonder at this slowly unfolding car-crash.

~~~~~~~~~~


EDIT: further to my previous post, I did some poking around the tennisabstract website, and it turns out that Sackmann does indeed factor different surfaces into his calculations; see here ("Adjusting Elo for surface - For most sports, we could stop here. A match is a match, with only minor variations. In tennis, though, ratings and predictions should vary quite a bit based on surface. My solution is a bit complicated. For each player, I maintain four separate Elo ratings: overall, hard court only, clay court only, and grass court only").

Well, that's good, until he messes it up by combining them ("Single-surface ratings aren’t great at predicting match results. A better solution is to take a 50/50 blend of single-surface and overall ratings"). Arrgghh. FIDE maintains three Elo ratings for its top players, in classical, rapid, and blitz time controls. Some players are better at slow time controls, e.g. Fabiano Caruana, while others are better at faster ones, e.g. Hikaru Nakamura. Likewise for our sport; you can't mix'n'match. Tennis is never played on a mixture of surfaces*, and so the Elo ratings for each should remain discrete.

Using Muster as an example as I did before (and pretending that hardcourts don't exist for a second)... if TM's clay Elo is 2250 and his grass Elo is 1750, then his "overall" Elo is going to be c. 2000. But he will never play at a 2000-level. He will either play at 2250 (on clay) or 1750 (grass).


* I know, I know...
 
Last edited:
V

venciso

New User
I like the project. Of course any algorithm you use will throw out unexpected results because tennis historically has varied so much. I still think it's a worthy project though and leaving the rankings aside the actual articles are very well researched and I have personally learned a lot about players I didn't know much about other than their names.

Are these the 22 that people think are left? I've ordered them according to what I think they should be but only within their own gender and open/amateur group! I think Novak has to be number 1 if the formula takes into account peak ELO, and also career ELO, considering his performance against Fed and Nadal but we'll see.

Djokovic 1
Nadal 2
Federer 3
Borg 4
Sampras 5
Lendl 6
Mcenroe 7

Laver 1 or 2
Tilden 2 or 1
Gonzales 3
Budge 4
Rosewall 5

Serena 1
Graf 2
Navratilova 3
Evert 4
Court 5
Seles 6
Venus 7

Wills moody 1
Connolly 2
Lenglen 3
 
U

urban

Legend
Recently i read the article on Don Budge, and i find it a bit inconsistent and not that good at all. Instead of describing his career, it begins with evaluation, that he is overrated, mainly for his "short" career. Now, in Elo terms, he must have been pretty high numbers between 1937-1940, but here no number of Elo is given. He had a pretty long streak in tournament singles, some speak of 92 matches. Maybe the competition in 1938, when he made the Grand Slam, was not the highest, but in 1937 he also played some of the finest matches in history, on which still books are written. This number of his streak i cannot find here in the article, nor i see his greatest achievement mentioned, the fast transition between amateur and pro tour. He was one of the very few, only Kramer i see here, who became pro leader in his very first pro year 1939, beating Vines, Perry and others in match play and main tournaments, and that on indoor courts, he had never played as amateur. And his career wasn't as short, as the article suggests. He was still a force on the pro tour in the late 1940s, battling strong champs like Kramer, and even in 1954 he still challenged the great Gonzalez.

The whole TA ranking seems inconsistent, if obviously McEnroe and Borg are ranked here over Budge, although they have shorter careers, and it is very debatable, if their peak was higher than that of Budge with his 92 match winning streak including the Grand Slam plus 6 majors in a row.
Dan Maskell in his 1989 list had Budge as Nr. 2 all time, and in the 1986 poll of 37 experts by Inside Tennis, the biggest of all those all time ranking polls, he still was ranked Nr. 3. Maybe today with more detalis of the pro tours researched,, some see Gonzalez or Rosewall with their great longevity above him, but i do think, that Don Budge still belongs in the top ten all time.
 
Last edited:
You must log in or register to reply here.

Similar threads

U
Maria Esther Bueno. Style or substance?
Replies
18
Views
594
PDJ
PDJ
S
  • Poll
Who are the final 2 top 10 female players of all time
2
Replies
62
Views
3K
PDJ
PDJ
U
Tennis abstract on the year 1973
Replies
4
Views
364
urban
U
M
Who is the 10th best female and male players of all time
Replies
39
Views
865
thrust
T
Bruno Santiago
Top25 Best Tennis Players of All Times - 1874 to present:
Replies
0
Views
243
Bruno Santiago
Bruno Santiago
Top