The best Open Era players, a statistical approach

[money_ball]

I asked myself the following question: Who can be considered the best Open Era players?

I want the question to be backed by stats, and for me, the simplest criteria are the number of titles won. Now the question is which titles? The immediately obvious one are Grand Slam titles. Next would be Masters titles. But since this tournament series started in 1990, there will be a lot of players left out. So in addition to the Masters series would be the Grand Prix Tennis Championship Series (1973-1989):

http://en.wikipedia.org/wiki/Grand_Prix_Tennis_Championship_Series_1970%E2%80%931989

So the following are my criteria:

1. A player must have won at least 1 Grand Slam title, or
2. A player must have won at least 2 Masters titles, or
3. A player must have won at least 2 Grand Prix titles

The following are the lists of players in each category, sorted by number of titles descending, with the player's career match win %:

Rank Player Grand Slam Titles Career Win %
1 Roger Federer 16 80.98%
2 Pete Sampras 14 77.44%
3 Bjorn Borg 11 82.72%
4 Rafael Nadal 10 82.62%
5 Jimmy Connors 8 81.76%
6 Ivan Lendl 8 81.76%
7 Andre Agassi 8 76.05%
8 John McEnroe 7 81.55%
9 Mats Wilander 7 72.01%
10 Boris Becker 6 76.91%
11 Stefan Edberg 6 74.91%
12 Rod Laver 5 79.54%
13 John Newcombe 5 74.83%
14 Ken Rosewall 4 74.37%
15 Novak Djokovic 4 78.23%
16 Jim Courier 4 68.10%
17 Guillermo Vilas 4 76.47%
18 Arthur Ashe 3 75.71%
19 Jan Kodes 3 63.08%
20 Gustavo Kuerten 3 64.74%
21 Lleyton Hewitt 2 72.98%
22 Yevgeny Kafelnikov 2 66.56%
23 Marat Safin 2 61.25%
24 Stan Smith 2 70.86%
25 Johan Kriek 2 62.75%
26 Illie Nastase 2 72.32%
27 Patrick Rafter 2 65.21%
28 Sergi Bruguera 2 62.26%
29 Andy Roddick 1 75.03%
30 Roscoe Tanner 1 67.09%
31 Manuel Orantes 1 72.21%
32 Michael Stich 1 68.63%
33 Andres Gimeno 1 64.92%
34 Yannick Noah 1 69.39%
35 Richard Krajicek 1 65.24%
36 Juan Carlos Ferrero 1 65.55%
37 Vitas Gerulaitis 1 69.77%
38 Goran Ivanisevic 1 64.27%
39 Juan Martin Del Potro 1 68.28%
40 Michael Chang 1 67.97%
41 Pat Cash 1 61.89%
42 Adriano Panatta 1 62.01%
43 Thomas Muster 1 69.50%
44 Andres Gomez 1 66.20%
45 Carlos Moya 1 64.32%
46 Petr Korda 1 62.31%
47 Albert Costa 1 58.51%
48 Brian Teacher 1 58.67%
49 Thomas Johansson 1 54.67%
50 Gaston Gaudio 1 57.94%
51 Mark Edmonson 1 51.33%

Rank Player Masters Titles Career Win %
1 Rafael Nadal 19 0.8262
2 Roger Federer 17 0.8098
3 Andre Agassi 16 0.7605
4 Pete Sampras 11 0.7744
5 Novak Djokovic 10 0.7823
6 Thomas Muster 8 0.6950
7 Andy Murray 7 0.7512
8 Michael Chang 7 0.6797
9 Andy Roddick 5 0.7503
10 Boris Becker 5 0.7691
11 Jim Courier 5 0.6810
12 Gustavo Kuerten 5 0.6474
13 Marat Safin 5 0.6125
14 Stefan Edberg 4 0.7491
15 Marcelo Rios 4 0.6707
16 Juan Carlos Ferrero 4 0.6555
17 Andrei Medvedev 4 0.6011
18 Nikolay Davydenko 3 0.6126
19 Carlos Moya 3 0.6432
20 Thomas Enqvist 3 0.6013
21 Guillermo Coria 2 0.6566
22 Lleyton Hewitt 2 0.7298
23 Michael Stich 2 0.6863
24 David Nalbandian 2 0.6762
25 Patrick Rafter 2 0.6521
26 Sergi Bruguera 2 0.6226
27 Richard Krajicek 2 0.6524
28 Andrei Chesnokov 2 0.5705
29 Alex Corretja 2 0.6092
30 Guy Forget 2 0.5663
31 Goran Ivanisevic 2 0.6427
32 Wayne Ferreira 2 0.6081
33 Albert Costa 1 0.5851
34 Petr Korda 1 0.6231
35 Thomas Johansson 1 0.5467

Rank Player Grand Prix Titles Career Win %
1 Ivan Lendl 22 0.8176
2 John McEnroe 19 0.8155
3 Jimmy Connors 18 0.8176
4 Bjorn Borg 15 0.8272
5 Rod Laver 9 0.7954
6 Boris Becker 8 0.7691
7 Mats Wilander 8 0.7201
8 Illie Nastase 6 0.7232
9 Stefan Edberg 4 0.7491
10 Stan Smith 4 0.7086
11 Guillermo Vilas 4 0.7647
12 Miloslav Mecir 3 0.6823
13 Manuel Orantes 3 0.7221
14 Arthur Ashe 2 0.7571
15 Vitas Gerulaitis 2 0.6977
16 Jose Higueras 2 0.6606
17 Alberto Mancini 2 0.5421
18 John Newcombe 2 0.7483
19 Yannick Noah 2 0.6939
20 Raul Ramirez 2 0.6658
21 Ken Rosewall 2 0.7437
22 Harold Solomon 2 0.6385
23 Roscoe Tanner 2 0.6709

So the combined list is as follows ordered by career match win %:

The best Open Era players:
Rank Player Career Win %
1 Bjorn Borg 82.72%
2 Rafael Nadal 82.62%
3 Jimmy Connors 81.76%
4 Ivan Lendl 81.76%
5 John McEnroe 81.55%
6 Roger Federer 80.98%
7 Rod Laver 79.54%
8 Novak Djokovic 78.23%
9 Pete Sampras 77.44%
10 Boris Becker 76.91%
11 Guillermo Vilas 76.47%
12 Andre Agassi 76.05%
13 Arthur Ashe 75.71%
14 Andy Murray 75.12%
15 Andy Roddick 75.03%
16 Stefan Edberg 74.91%
17 John Newcombe 74.83%
18 Ken Rosewall 74.37%
19 Lleyton Hewitt 72.98%
20 Illie Nastase 72.32%
21 Manuel Orantes 72.21%
22 Mats Wilander 72.01%
23 Stan Smith 70.86%
24 Vitas Gerulaitis 69.77%
25 Thomas Muster 69.50%
26 Yannick Noah 69.39%
27 Michael Stich 68.63%
28 Juan Martin Del Potro 68.28%
29 Miloslav Mecir 68.23%
30 Jim Courier 68.10%
31 Michael Chang 67.97%
32 David Nalbandian 67.62%
33 Roscoe Tanner 67.09%
34 Marcelo Rios 67.07%
35 Raul Ramirez 66.58%
36 Yevgeny Kafelnikov 66.56%
37 Andres Gomez 66.20%
38 Jose Higueras 66.06%
39 Guillermo Coria 65.66%
40 Juan Carlos Ferrero 65.55%
41 Richard Krajicek 65.24%
42 Patrick Rafter 65.21%
43 Andres Gimeno 64.92%
44 Gustavo Kuerten 64.74%
45 Carlos Moya 64.32%
46 Goran Ivanisevic 64.27%
47 Harold Solomon 63.85%
48 Jan Kodes 63.08%
49 Johan Kriek 62.75%
50 Petr Korda 62.31%
51 Sergi Bruguera 62.26%
52 Adriano Panatta 62.01%
53 Pat Cash 61.89%
54 Nikolay Davydenko 61.26%
55 Marat Safin 61.25%
56 Alex Corretja 60.92%
57 Wayne Ferreira 60.81%
58 Thomas Enqvist 60.13%
59 Andrei Medvedev 60.11%
60 Brian Teacher 58.67%
61 Albert Costa 58.51%
62 Gaston Gaudio 57.94%
63 Andrei Chesnokov 57.05%
64 Guy Forget 56.63%
65 Thomas Johansson 54.67%
66 Alberto Mancini 54.21%
67 Mark Edmonson 51.33%

 

[money_ball]

What is notable are the following players who have won a Grand Slam title, but never a Masters or Grand Prix title (ordered by career match win % descending):

Juan Martin Del Potro 68.28%
Yevgeny Kafelnikov 66.56%
Andres Gimeno 64.92%
Adriano Panatta 62.01%
Pat Cash 61.89%
Brian Teacher 58.67%
Gaston Gaudio 57.94%
Mark Edmonson 51.33%

 

[fed_rulz]

I asked myself the following question: Who can be considered the best Open Era players?

I want the question to be backed by stats, and for me, the simplest criteria are the number of titles won. Now the question is which titles? The immediately obvious one are Grand Slam titles. Next would be Masters titles. But since this tournament series started in 1990, there will be a lot of players left out. So in addition to the Masters series would be the Grand Prix Tennis Championship Series (1973-1989):

http://en.wikipedia.org/wiki/Grand_Prix_Tennis_Championship_Series_1970–1989

So the following are my criteria:

1. A player must have won at least 1 Grand Slam title, or
2. A player must have won at least 2 Masters titles, or
3. A player must have won at least 2 Grand Prix titles

The following are the lists of players in each category, sorted by number of titles descending, with the player's career match win %:

Rank Player Grand Slam Titles Career Win %
1 Roger Federer 16 80.98%
2 Pete Sampras 14 77.44%
3 Bjorn Borg 11 82.72%
4 Rafael Nadal 10 82.62%
5 Jimmy Connors 8 81.76%
6 Ivan Lendl 8 81.76%
7 Andre Agassi 8 76.05%
8 John McEnroe 7 81.55%
9 Mats Wilander 7 72.01%
10 Boris Becker 6 76.91%
11 Stefan Edberg 6 74.91%
12 Rod Laver 5 79.54%
13 John Newcombe 5 74.83%
14 Ken Rosewall 4 74.37%
15 Novak Djokovic 4 78.23%
16 Jim Courier 4 68.10%
17 Guillermo Vilas 4 76.47%
18 Arthur Ashe 3 75.71%
19 Jan Kodes 3 63.08%
20 Gustavo Kuerten 3 64.74%
21 Lleyton Hewitt 2 72.98%
22 Yevgeny Kafelnikov 2 66.56%
23 Marat Safin 2 61.25%
24 Stan Smith 2 70.86%
25 Johan Kriek 2 62.75%
26 Illie Nastase 2 72.32%
27 Patrick Rafter 2 65.21%
28 Sergi Bruguera 2 62.26%
29 Andy Roddick 1 75.03%
30 Roscoe Tanner 1 67.09%
31 Manuel Orantes 1 72.21%
32 Michael Stich 1 68.63%
33 Andres Gimeno 1 64.92%
34 Yannick Noah 1 69.39%
35 Richard Krajicek 1 65.24%
36 Juan Carlos Ferrero 1 65.55%
37 Vitas Gerulaitis 1 69.77%
38 Goran Ivanisevic 1 64.27%
39 Juan Martin Del Potro 1 68.28%
40 Michael Chang 1 67.97%
41 Pat Cash 1 61.89%
42 Adriano Panatta 1 62.01%
43 Thomas Muster 1 69.50%
44 Andres Gomez 1 66.20%
45 Carlos Moya 1 64.32%
46 Petr Korda 1 62.31%
47 Albert Costa 1 58.51%
48 Brian Teacher 1 58.67%
49 Thomas Johansson 1 54.67%
50 Gaston Gaudio 1 57.94%
51 Mark Edmonson 1 51.33%

Rank Player Masters Titles Career Win %
1 Rafael Nadal 19 0.8262
2 Roger Federer 17 0.8098
3 Andre Agassi 16 0.7605
4 Pete Sampras 11 0.7744
5 Novak Djokovic 10 0.7823
6 Thomas Muster 8 0.6950
7 Andy Murray 7 0.7512
8 Michael Chang 7 0.6797
9 Andy Roddick 5 0.7503
10 Boris Becker 5 0.7691
11 Jim Courier 5 0.6810
12 Gustavo Kuerten 5 0.6474
13 Marat Safin 5 0.6125
14 Stefan Edberg 4 0.7491
15 Marcelo Rios 4 0.6707
16 Juan Carlos Ferrero 4 0.6555
17 Andrei Medvedev 4 0.6011
18 Nikolay Davydenko 3 0.6126
19 Carlos Moya 3 0.6432
20 Thomas Enqvist 3 0.6013
21 Guillermo Coria 2 0.6566
22 Lleyton Hewitt 2 0.7298
23 Michael Stich 2 0.6863
24 David Nalbandian 2 0.6762
25 Patrick Rafter 2 0.6521
26 Sergi Bruguera 2 0.6226
27 Richard Krajicek 2 0.6524
28 Andrei Chesnokov 2 0.5705
29 Alex Corretja 2 0.6092
30 Guy Forget 2 0.5663
31 Goran Ivanisevic 2 0.6427
32 Wayne Ferreira 2 0.6081
33 Albert Costa 1 0.5851
34 Petr Korda 1 0.6231
35 Thomas Johansson 1 0.5467

Rank Player Grand Prix Titles Career Win %
1 Ivan Lendl 22 0.8176
2 John McEnroe 19 0.8155
3 Jimmy Connors 18 0.8176
4 Bjorn Borg 15 0.8272
5 Rod Laver 9 0.7954
6 Boris Becker 8 0.7691
7 Mats Wilander 8 0.7201
8 Illie Nastase 6 0.7232
9 Stefan Edberg 4 0.7491
10 Stan Smith 4 0.7086
11 Guillermo Vilas 4 0.7647
12 Miloslav Mecir 3 0.6823
13 Manuel Orantes 3 0.7221
14 Arthur Ashe 2 0.7571
15 Vitas Gerulaitis 2 0.6977
16 Jose Higueras 2 0.6606
17 Alberto Mancini 2 0.5421
18 John Newcombe 2 0.7483
19 Yannick Noah 2 0.6939
20 Raul Ramirez 2 0.6658
21 Ken Rosewall 2 0.7437
22 Harold Solomon 2 0.6385
23 Roscoe Tanner 2 0.6709

So the combined list is as follows ordered by career match win %:

The best Open Era players:
Rank Player Career Win %
1 Bjorn Borg 82.72%
2 Rafael Nadal 82.62%
3 Jimmy Connors 81.76%
4 Ivan Lendl 81.76%
5 John McEnroe 81.55%
6 Roger Federer 80.98%
7 Rod Laver 79.54%
8 Novak Djokovic 78.23%
9 Pete Sampras 77.44%
10 Boris Becker 76.91%
11 Guillermo Vilas 76.47%
12 Andre Agassi 76.05%
13 Arthur Ashe 75.71%
14 Andy Murray 75.12%
15 Andy Roddick 75.03%
16 Stefan Edberg 74.91%
17 John Newcombe 74.83%
18 Ken Rosewall 74.37%
19 Lleyton Hewitt 72.98%
20 Illie Nastase 72.32%
21 Manuel Orantes 72.21%
22 Mats Wilander 72.01%
23 Stan Smith 70.86%
24 Vitas Gerulaitis 69.77%
25 Thomas Muster 69.50%
26 Yannick Noah 69.39%
27 Michael Stich 68.63%
28 Juan Martin Del Potro 68.28%
29 Miloslav Mecir 68.23%
30 Jim Courier 68.10%
31 Michael Chang 67.97%
32 David Nalbandian 67.62%
33 Roscoe Tanner 67.09%
34 Marcelo Rios 67.07%
35 Raul Ramirez 66.58%
36 Yevgeny Kafelnikov 66.56%
37 Andres Gomez 66.20%
38 Jose Higueras 66.06%
39 Guillermo Coria 65.66%
40 Juan Carlos Ferrero 65.55%
41 Richard Krajicek 65.24%
42 Patrick Rafter 65.21%
43 Andres Gimeno 64.92%
44 Gustavo Kuerten 64.74%
45 Carlos Moya 64.32%
46 Goran Ivanisevic 64.27%
47 Harold Solomon 63.85%
48 Jan Kodes 63.08%
49 Johan Kriek 62.75%
50 Petr Korda 62.31%
51 Sergi Bruguera 62.26%
52 Adriano Panatta 62.01%
53 Pat Cash 61.89%
54 Nikolay Davydenko 61.26%
55 Marat Safin 61.25%
56 Alex Corretja 60.92%
57 Wayne Ferreira 60.81%
58 Thomas Enqvist 60.13%
59 Andrei Medvedev 60.11%
60 Brian Teacher 58.67%
61 Albert Costa 58.51%
62 Gaston Gaudio 57.94%
63 Andrei Chesnokov 57.05%
64 Guy Forget 56.63%
65 Thomas Johansson 54.67%
66 Alberto Mancini 54.21%
67 Mark Edmonson 51.33%

why are the career win % the same for all events? Using career win % has a problem because current top players will have a higher % as they might be in their primes; retired players would've gone through a decline phase before deciding to quit,which would affect their win %. Borg has an inflated career win % because he quit early.

 

[veroniquem]

I'd be ready to bet Delpo will win at least one master. He's still young.

 

[TennisFan3]

I'd be ready to bet Delpo will win at least one master. He's still young.

He'll win several Masters and several Majors. Mark my words..

 

[tennisenthusiast]

statistically speaking, gap between federer and djokovic is only going to widen? is that right?

 

[money_ball]

why are the career win % the same for all events? Using career win % has a problem because current top players will have a higher % as they might be in their primes; retired players would've gone through a decline phase before deciding to quit,which would affect their win %. Borg has an inflated career win % because he quit early.

The "career win %" stat is for all career matches, which is why those numbers are the same in each category.

Your absolutely correct observation on one flaw in the "career win %" stat is one of the reasons why relying on a single metric to determine the "best" player is unsatisfactory. I only used this stat for organizational purposes, as it is better to see a list of these 67 "best" players ordered by this stat rather than alphabetically.

My main purpose was to narrow down a list of the "best" Open Era players. Now I can focus on these 67 players and start examining other stats.

For starters, what other stats on these 67 players would you be interested in finding out?

 

[money_ball]

statistically speaking, gap between federer and djokovic is only going to widen? is that right?

If you mean by the "career win %" stat, it depends. If Federer continues to keep playing, his stat will slowly drop, as he will no longer be as dominant.

As for Djokovic, he is entering his peak period, so his "career win %" will be increasing.

As to where both of their respective "career win %" stats will be at the end of their careers, I have no idea: the gap could widen, it could narrow, Djokovic might even pass Federer.

 

[fed_rulz]

If you mean by the "career win %" stat, it depends. If Federer continues to keep playing, his stat will slowly drop, as he will no longer be as dominant.

As for Djokovic, he is entering his peak period, so his "career win %" will be increasing.

As to where both of their respective "career win %" stats will be at the end of their careers, I have no idea: the gap could widen, it could narrow, Djokovic might even pass Federer.

no offense, but there have been 10000+ threads on this topic, so I doubt you'll be able to come up new material that hasn't been presented and analyzed before. if you're genuienly interested, please use the search function, you'll be able to find some history on this very topic, ans you'll perhaps be able to add your thoughts there. Again, I don't mean to trash your effort, I'm just trying to save you some time .

 

[money_ball]

no offense, but there have been 10000+ threads on this topic, so I doubt you'll be able to come up new material that hasn't been presented and analyzed before. if you're genuienly interested, please use the search function, you'll be able to find some history on this very topic, ans you'll perhaps be able to add your thoughts there. Again, I don't mean to trash your effort, I'm just trying to save you some time .

I am actually working on something that I don't think has been done on these boards. I have scraped all of the data into a database on these 67 players from the ATP website. I will actually be creating an online G.O.A.T. calculator for you to play with, where you get to assign weights to any of these following categories, in order for you to come up with your own G.O.A.T. ranking, based on those stats. The categories are as follows:

Number 1 Ranking Stats
# of Year-End No. 1 Rankings
# of times reaching No. 1
Most Consecutive weeks at No. 1
Cumulative weeks at No. 1
Avg. Consecutive Weeks at No. 1

Career Stats
Career Titles
Career Wins
Career Win %
Grand Slam Index
Grand Slam Titles
Grand Slam Wins
Grand Slam Win %

Grand Prix Tennis Championship Series Stat
Grand Prix Tennis Championship Series Titles

Masters Stats
Masters Titles
Masters Wins
Masters Win %

Surface Stats
Clay Titles
Clay Wins
Clay Win %
Grass Titles
Grass Wins
Grass Win %
Hardcourt Titles
Hardcourt Wins
Hardcourt Win %
Carpet Titles
Carpet Wins
Carpet Win %

Environment Stats
Indoor Titles
Indoor Wins
Indoor Win %
Outdoor Titles
Outdoor Wins
Outdoor Win %

Pressure Situation Stats
Tiebreak Wins
Tiebreak Win %
Top 10 Wins
Top 10 Win %
Finals Wins
Finals Win %
Deciding Set Wins
Deciding Set Win %
5th Set Wins
5th Set Win %

Other Stats
After Winning 1st Set Wins
After Winning 1st Set Win %
After Losing 1st Set Wins
After Losing 1st Set Win %
vs. Right Handers Wins
vs. Right Handers Win %
vs. Left Handers Wins
vs. Left Handers Win %
Pro Career Length

Also, just based on these 67 "best" players, I have also determined the following interesting info, which I will be releasing soon:

- % of total matches played in Grand Slams over time
- % of total matches played in Grand Prix tournaments over time
- % of total matches played in Masters tournaments over time
- average height and weight
- average age turning pro
- average retirement age
- average pro career length
- average match win % playing lefties versus righties
- average age reaching No. 1 ranking
- average years pro experience upon reaching No. 1 ranking

And the best thing about all this is that I will be releasing all of this to the public for you to download and analyze for yourself. Of the 67 "best" players, only 10 are active, and those stats will be automatically updated.

I want tennis to eventually attain the level of statistical sophistication as baseball, and I am just making one small contribution to this.

 

[vernonbc]

I think it's great you're doing all this work money ball. Comprehensive stats is something that has been missing from tennis for a long time. It would be nice if actual factual data could resolve some of the mud fights that constantly occur in this forum but I doubt it will, unfortunately. Nevertheless, some of the more reasonable, mature posters will undoubtedly find this information interesting and useful.

 

[urban]

Very ambitious task. The ATP stats are not solid up to 1985 and slowly changing and integrating more forgotten results of the early open era, as Q and M just wrote in the former players department. The numbers reflect the fact, that in many departments like percentages, overall tournaments, Grand Prix, Masters equivalents, the top players from the 70s and 80s do very well, while in the majors they are often behind the modern players. The main reason is the more solid structure of the modern circuit, which is firmly centered around the 4 majors.

 

[urban]

One final note, about the term Grand Prix, that is imo not rightly used on wikipedia and other internet sources. The Grand Prix was originated in 1970, and extended to all sanctioned tournaments of the ILTF. It was the rivalling circuit of the WCT tour. It was a tournament series, extended over the whole year, and had at the end of the year a points table, with big bonus money for the leaders. Grand Prix tournaments are not identical with Masters equivalents. For the Masters or Super Nine equivalents we can single out yearly circa 8-9 important early open era events, which came right after the majors (and often overshadowed the AO). Those events were for instance the Italian Open, the German Open, the South African Open, the Philadephia indoor, the British Indoor at Wembley, the Los Angeles South Pacifik, the Las Vegas or Tucson events. Those events with great draws and good tradition can be rightfully named equivalents of the Masters series today.

 

[Benhur]

why are the career win % the same for all events? Using career win % has a problem because current top players will have a higher % as they might be in their primes; retired players would've gone through a decline phase before deciding to quit,which would affect their win %. Borg has an inflated career win % because he quit early.

And on that note, the winning percentages of Connors, Lendl and McEnroe look especially impressive among retired open era players, considering their long careers, especially Connors.

 

[tenniselbow1]

An interesting calculator function would be to look back at the stats of each player at a certain age. Fed at 26 vs Rafa at 26, and be able to compare stats such as win percentage on each surface over time. And be able to take those numbers at any given time in a players career and compare them to the overall stats of these 67 players. Thanks for doing this, you should ask for help, surely you'll get it.

 

[sureshs]

It has been done recently with a branch of mathematics called network analysis. Jimmy Connors is the GOAT. The math takes into account the connected nature of things - who wins against who and who in turn loses against who (or is it whom) - and decides on the relative toughness of opponents. This kind of math is becoming very important today with the connected nature of social interactions through FB, Twitter etc.

 

[money_ball]

It has been done recently with a branch of mathematics called network analysis. Jimmy Connors is the GOAT. The math takes into account the connected nature of things - who wins against who and who in turn loses against who (or is it whom) - and decides on the relative toughness of opponents. This kind of math is becoming very important today with the connected nature of social interactions through FB, Twitter etc.

Please provide a source for this. I am curious on the methodology.

The closest thing that I have found on these boards is the adaptation of the ELO ratings in chess to the men's players:

http://tt.tennis-warehouse.com/showthread.php?t=162812

 

[money_ball]

An interesting calculator function would be to look back at the stats of each player at a certain age. Fed at 26 vs Rafa at 26, and be able to compare stats such as win percentage on each surface over time. And be able to take those numbers at any given time in a players career and compare them to the overall stats of these 67 players. Thanks for doing this, you should ask for help, surely you'll get it.

That would be interesting. Unfortunately my data scraping scripts are not yet sophisticated enough to get match data by age from the ATP website.

The calculations is the easy part. The hard part is scraping the data from the ATP!

 

[sureshs]

Please provide a source for this. I am curious on the methodology.

The closest thing that I have found on these boards is the adaptation of the ELO ratings in chess to the men's players:

http://tt.tennis-warehouse.com/showthread.php?t=162812

Search for "Connors" in thread titles. GOAT also, but that will be too many

 

[money_ball]

Search for "Connors" in thread titles. GOAT also, but that will be too many

I found the original paper on PLoS One:

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0017249

This is a very interesting method, and I will incorporating these results into its own separate category for the G.O.A.T. calculator.

 

[Jules]

I love stats! Great job money_ball. Keep doing your thing, I will follow it with great interest.

 

[money_ball]

I found the original paper on PLoS One:

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0017249

This is a very interesting method, and I will incorporating these results into its own separate category for the G.O.A.T. calculator.

After reading the paper, I found that the author neglected to include the "Prestige Score" that he calculated to determine his ranking of the 30 best players:

1 Jimmy Connors
2 Ivan Lendl
3 John McEnroe
4 Guillermo Vilas
5 Andre Agassi
6 Stefan Edberg
7 Roger Federer
8 Pete Sampras
9 Ilie Nastase
10 Bjorn Borg
11 Boris Becker
12 Arthur Ashe
13 Brian Gottfried
14 Stan Smith
15 Manuel Orantes
16 Michael Chang
17 Roscoe Tanner
18 Eddie Dibbs
19 Harold Solomon
20 Tom Okker
21 Mats Wilander
22 Goran Ivanisevic
23 Vitas Gerulaitis
24 Rafael Nadal
25 Raul Ramirez
26 John Newcombe
27 Ken Rosewall
28 Yevgeny Kafelnikov
29 Andy Roddick
30 Thomas Muster

I want to do the calculations myself, but unfortunately I don't have the head-to-head match data for the thousands of matches on all these players. I will try to email the paper's author to see if I can get that data set.

I don't think any one methodology can conclude on the GOAT, as in the end it is ultimately subjective, but this methodology is useful in determining the relative competitiveness of a player's opponents.

Take for example two great clay court players on this list: Borg and Vilas. This methodology ranks Vilas ahead of Borg (reflecting that Vilas has beaten relatively "tougher" opponents than Borg). But looking at one simple stat, head-to-head record, and it is overwhelmingly in Borg's favor: 17-5.

 

[sureshs]

After reading the paper, I found that the author neglected to include the "Prestige Score" that he calculated to determine his ranking of the 30 best players:

1 Jimmy Connors
2 Ivan Lendl
3 John McEnroe
4 Guillermo Vilas
5 Andre Agassi
6 Stefan Edberg
7 Roger Federer
8 Pete Sampras
9 Ilie Nastase
10 Bjorn Borg
11 Boris Becker
12 Arthur Ashe
13 Brian Gottfried
14 Stan Smith
15 Manuel Orantes
16 Michael Chang
17 Roscoe Tanner
18 Eddie Dibbs
19 Harold Solomon
20 Tom Okker
21 Mats Wilander
22 Goran Ivanisevic
23 Vitas Gerulaitis
24 Rafael Nadal
25 Raul Ramirez
26 John Newcombe
27 Ken Rosewall
28 Yevgeny Kafelnikov
29 Andy Roddick
30 Thomas Muster

I want to do the calculations myself, but unfortunately I don't have the head-to-head match data for the thousands of matches on all these players. I will try to email the paper's author to see if I can get that data set.

I don't think any one methodology can conclude on the GOAT, as in the end it is ultimately subjective, but this methodology is useful in determining the relative competitiveness of a player's opponents.

Take for example two great clay court players on this list: Borg and Vilas. This methodology ranks Vilas ahead of Borg (reflecting that Vilas has beaten relatively "tougher" opponents than Borg). But looking at one simple stat, head-to-head record, and it is overwhelmingly in Borg's favor: 17-5.

When it was first posted here, the first thing people pointed out was there was no Laver. That destroyed the credibility in most people's minds.

 

[money_ball]

When it was first posted here, the first thing people pointed out was there was no Laver. That destroyed the credibility in most people's minds.

I just spoke to Filippo Radicchi (the author of the paper) over the phone, and he told me that his data is based only on the available match data from the ATP (1968-2010):

http://www.atpworldtour.com/Scores/Archive-Event-Calendar.aspx

Since Laver was 29 at the start of 1968, his best results are not included in the data.

Filippo Radicchi has also agreed to send me his data set (for all the thousands of matches since 1968) as well as the Prestige Scores for the top 100 players, which I will incorporate into my work which I will make available to you all soon.

 

[sureshs]

I just spoke to Filippo Radicchi (the author of the paper) over the phone, and he told me that his data is based only on the available match data from the ATP (1968-2010):

http://www.atpworldtour.com/Scores/Archive-Event-Calendar.aspx

Since Laver was 29 at the start of 1968, his best results are not included in the data.

Filippo Radicchi has also agreed to send me his data set (for all the thousands of matches since 1968) as well as the Prestige Scores for the top 100 players, which I will incorporate into my work which I will make available to you all soon.

Good.

This connected graph thingie has a lot of potential. Here is a summary of Google's page rank algorithm:

A PageRank results from a mathematical algorithm based on the graph, the webgraph, created by all World Wide Web pages as nodes and hyperlinks as edges, taking into consideration authority hubs such as cnn.com or usa.gov. The rank value indicates an importance of a particular page. A hyperlink to a page counts as a vote of support. The PageRank of a page is defined recursively and depends on the number and PageRank metric of all pages that link to it ("incoming links"). A page that is linked to by many pages with high PageRank receives a high rank itself. If there are no links to a web page there is no support for that page.

It can be used to evaluate economies, predict international conflicts, model currency fluctuations and a whole bunch of stuff.

 

[jackson vile]

What ever you do, take your time. People here and else where will look for any reason to dismiss your work.

 

[DjokovicForTheWin]

What ever you do, take your time. People here and else where will look for any reason to dismiss your work.

You mean like you? When his analysis shows Nadal is not #1

 

[purge]

im not sure i get this? looks like all it does is rank players by their career win %. how is that new?

or am i missing something here?

 

[Bobby Jr]

My main purpose was to narrow down a list of the "best" Open Era players. Now I can focus on these 67 players and start examining other stats.

If the main purpose really was to narrow down the list then the statistics list should go like this:

Federer - 16
Sampras - 14.... etc.

Job done.

 

[timnz]

How will you deal with the following?

I am actually working on something that I don't think has been done on these boards. I have scraped all of the data into a database on these 67 players from the ATP website. I will actually be creating an online G.O.A.T. calculator for you to play with, where you get to assign weights to any of these following categories, in order for you to come up with your own G.O.A.T. ranking, based on those stats. The categories are as follows:

Number 1 Ranking Stats
# of Year-End No. 1 Rankings
# of times reaching No. 1
Most Consecutive weeks at No. 1
Cumulative weeks at No. 1
Avg. Consecutive Weeks at No. 1

Career Stats
Career Titles
Career Wins
Career Win %
Grand Slam Index
Grand Slam Titles
Grand Slam Wins
Grand Slam Win %

Grand Prix Tennis Championship Series Stat
Grand Prix Tennis Championship Series Titles

Masters Stats
Masters Titles
Masters Wins
Masters Win %

Surface Stats
Clay Titles
Clay Wins
Clay Win %
Grass Titles
Grass Wins
Grass Win %
Hardcourt Titles
Hardcourt Wins
Hardcourt Win %
Carpet Titles
Carpet Wins
Carpet Win %

Environment Stats
Indoor Titles
Indoor Wins
Indoor Win %
Outdoor Titles
Outdoor Wins
Outdoor Win %

Pressure Situation Stats
Tiebreak Wins
Tiebreak Win %
Top 10 Wins
Top 10 Win %
Finals Wins
Finals Win %
Deciding Set Wins
Deciding Set Win %
5th Set Wins
5th Set Win %

Other Stats
After Winning 1st Set Wins
After Winning 1st Set Win %
After Losing 1st Set Wins
After Losing 1st Set Win %
vs. Right Handers Wins
vs. Right Handers Win %
vs. Left Handers Wins
vs. Left Handers Win %
Pro Career Length

Also, just based on these 67 "best" players, I have also determined the following interesting info, which I will be releasing soon:

- % of total matches played in Grand Slams over time
- % of total matches played in Grand Prix tournaments over time
- % of total matches played in Masters tournaments over time
- average height and weight
- average age turning pro
- average retirement age
- average pro career length
- average match win % playing lefties versus righties
- average age reaching No. 1 ranking
- average years pro experience upon reaching No. 1 ranking

And the best thing about all this is that I will be releasing all of this to the public for you to download and analyze for yourself. Of the 67 "best" players, only 10 are active, and those stats will be automatically updated.

I want tennis to eventually attain the level of statistical sophistication as baseball, and I am just making one small contribution to this.

Grand Prix Tennis Championship Series should be ranked equal with Masters series - so you are fair to the older players. And also you need to do this to be fair to players who overlapped the 80's and 90's like Becker. It isn't his fault that Masters 1000's started in 1990 - equivalent Grand Prix Tennis Championship Series should be equal rated.

THe other issue is that I don't see you mentioning Season end finals - like WCT Finals & Masters Cup. Those tournaments have been rated by players as being more important than Masters level events but less than Grand Slam events. To miss them out is to miss out Federer, Sampras & Lendl's - five Masters Cups - these are very significant achievements or McEnroe's five WCT finals wins. In fact the 1970's the WCT finals was regarded by many as the 4th major. And the same could be said for the Masters Cup in the early 80's. A fair weighting would be at least to view it as 3/4s of a Grand Slam win. But regardless of your opinion of them, they should be mentioned promenently.

Hence if you view Grand Prix Tennis Championship Series as being equivalent to Masters 1000 - then Lendl is the leader. And if you add them to their Season End finals - Lendl is far and away the leader - way ahead of Nadal.

 

[money_ball]

I have created a makeshift micro-prototype of the online GOAT calculator here:

http://tt.tennis-warehouse.com/showthread.php?t=400506

Have fun!

 

[wy2sl0]

Don't want to sound like a broken record, but Roddick is doing pretty well on those lists - especially playing during the peak of the goat on clay, and goat on all other surfaces. He doesn't get enough credit.

 

[kiki]

Grand Prix Tennis Championship Series should be ranked equal with Masters series - so you are fair to the older players. And also you need to do this to be fair to players who overlapped the 80's and 90's like Becker. It isn't his fault that Masters 1000's started in 1990 - equivalent Grand Prix Tennis Championship Series should be equal rated.

THe other issue is that I don't see you mentioning Season end finals - like WCT Finals & Masters Cup. Those tournaments have been rated by players as being more important than Masters level events but less than Grand Slam events. To miss them out is to miss out Federer, Sampras & Lendl's - five Masters Cups - these are very significant achievements or McEnroe's five WCT finals wins. In fact the 1970's the WCT finals was regarded by many as the 4th major. And the same could be said for the Masters Cup in the early 80's. A fair weighting would be at least to view it as 3/4s of a Grand Slam win. But regardless of your opinion of them, they should be mentioned promenently.

Hence if you view Grand Prix Tennis Championship Series as being equivalent to Masters 1000 - then Lendl is the leader. And if you add them to their Season End finals - Lendl is far and away the leader - way ahead of Nadal.

Yes, they should weight around 66% of a GS win and they should add to the GS titles, forget about " Master Series" or former "Superseries" ( the equivalent to today´s Masters Series or Super 9)

Wimbledon,Roland Garros,US Open, Australian, ATP Finals/Masters Cup and WCT Championship final are all that should count.

Since there is no WCT since 1989, I suggest to use the Miami Super 9, since it has something the other Master 1000 events do not have: played in 128 draws, over 2 weeks and in a regular hard court.

So from 1970-1989

Wimbledon
FO
USO
AO
Masters Cup ( 66% of a GS)
WCT Finals ( 66% of a GS)

from 1990-2011

Wimbledon
FO
USO
AO
Masters Cup ( or ATP Championship Finals as it was called a few years ago)
Miami (Key Biscaine)

 

[kiki]

Yes, they should weight around 66% of a GS win and they should add to the GS titles, forget about " Master Series" or former "Superseries" ( the equivalent to today´s Masters Series or Super 9)

Wimbledon,Roland Garros,US Open, Australian, ATP Finals/Masters Cup and WCT Championship final are all that should count.

Since there is no WCT since 1989, I suggest to use the Miami Super 9, since it has something the other Master 1000 events do not have: played in 128 draws, over 2 weeks and in a regular hard court.

So from 1970-1989

Wimbledon
FO
USO
AO
Masters Cup ( 66% of a GS)
WCT Finals ( 66% of a GS)

from 1990-2011

Wimbledon
FO
USO
AO
Masters Cup ( or ATP Championship Finals as it was called a few years ago)
Miami (Key Biscaine)

Of course, the value of Masters Cup and Miami for the 1990-2011 period should be the same as the events they replace, that is a 66% of a GS title

( I must admit that the AO in the 1970´s would weight like a 70%-75% of the Wimbledon,RG or Forest Hills titles but it would create more confusion9

 

[Clay lover]

Mats is not worse than Andy Roddick... but yeah stats do not show everything...

 

[pvaudio]

I am actually working on something that I don't think has been done on these boards. I have scraped all of the data into a database on these 67 players from the ATP website. I will actually be creating an online G.O.A.T. calculator for you to play with, where you get to assign weights to any of these following categories, in order for you to come up with your own G.O.A.T. ranking, based on those stats. The categories are as follows:

Number 1 Ranking Stats
# of Year-End No. 1 Rankings
# of times reaching No. 1
Most Consecutive weeks at No. 1
Cumulative weeks at No. 1
Avg. Consecutive Weeks at No. 1

Career Stats
Career Titles
Career Wins
Career Win %
Grand Slam Index
Grand Slam Titles
Grand Slam Wins
Grand Slam Win %

Grand Prix Tennis Championship Series Stat
Grand Prix Tennis Championship Series Titles

Masters Stats
Masters Titles
Masters Wins
Masters Win %

Surface Stats
Clay Titles
Clay Wins
Clay Win %
Grass Titles
Grass Wins
Grass Win %
Hardcourt Titles
Hardcourt Wins
Hardcourt Win %
Carpet Titles
Carpet Wins
Carpet Win %

Environment Stats
Indoor Titles
Indoor Wins
Indoor Win %
Outdoor Titles
Outdoor Wins
Outdoor Win %

Pressure Situation Stats
Tiebreak Wins
Tiebreak Win %
Top 10 Wins
Top 10 Win %
Finals Wins
Finals Win %
Deciding Set Wins
Deciding Set Win %
5th Set Wins
5th Set Win %

Other Stats
After Winning 1st Set Wins
After Winning 1st Set Win %
After Losing 1st Set Wins
After Losing 1st Set Win %
vs. Right Handers Wins
vs. Right Handers Win %
vs. Left Handers Wins
vs. Left Handers Win %
Pro Career Length

Also, just based on these 67 "best" players, I have also determined the following interesting info, which I will be releasing soon:

- % of total matches played in Grand Slams over time
- % of total matches played in Grand Prix tournaments over time
- % of total matches played in Masters tournaments over time
- average height and weight
- average age turning pro
- average retirement age
- average pro career length
- average match win % playing lefties versus righties
- average age reaching No. 1 ranking
- average years pro experience upon reaching No. 1 ranking

And the best thing about all this is that I will be releasing all of this to the public for you to download and analyze for yourself. Of the 67 "best" players, only 10 are active, and those stats will be automatically updated.

I want tennis to eventually attain the level of statistical sophistication as baseball, and I am just making one small contribution to this.

You are either a mathematician, or an engineer...which is the same thing except you leave the last 5 steps of the problem with "you get the idea".

 

[hoodjem]

He'll win several Masters and several Majors. Mark my words..

Alas! Too bad with the injuries.

Delpo ends up getting to four Masters finals, winning one at 2018 Indian Wells.