Mens Tennis Forums banner

The Rankings Database [version 2]

32K views 154 replies 10 participants last post by  Slasher1985 
#1 · (Edited)
Research continuing in another form
 
#7 · (Edited)
Second Category Based Additive Best 18 System Part II [2014- ]

2014: (RANKING SYSTEM - ACTUAL SYSTEM)
06/01/2014
13/01/2014
27/01/2014
03/02/2014
10/02/2014
17/02/2014
24/02/2014
03/03/2014
17/03/2014
31/03/2014
07/04/2014
14/04/2014
21/04/2014
28/04/2014
05/05/2014
12/05/2014
19/05/2014
26/05/2014
09/06/2014
16/06/2014
23/06/2014
07/07/2014
14/07/2014
21/07/2014
28/07/2014
04/08/2014
11/08/2014
18/08/2014
25/08/2014
08/09/2014
15/09/2014
22/09/2014
29/09/2014
06/10/2014
13/10/2014
20/10/2014
27/10/2014
03/11/2014
10/11/2014
17/11/2014
24/11/2014 [YEAR END RANKINGS]
01/12/2014
08/12/2014
15/12/2014
22/12/2014
29/12/2014

2015: (RANKING SYSTEM - ACTUAL SYSTEM)
05/01/2015
12/01/2015
19/01/2015
02/02/2015
09/02/2015
16/02/2015
23/02/2015
02/03/2015
09/03/2015
23/03/2015
06/04/2015
13/04/2015
20/04/2015
27/04/2015
04/05/2015
11/05/2015
18/05/2015
25/05/2015
08/06/2015
15/06/2015
22/06/2015
29/06/2015
13/07/2015
20/07/2015
27/07/2015
03/08/2015
10/08/2015
17/08/2015
24/08/2015
31/08/2015
14/09/2015
21/09/2015
28/09/2015
05/10/2015
12/10/2015
19/10/2015
26/10/2015
02/11/2015
09/11/2015
16/11/2015
23/11/2015
30/11/2015 [YEAR END RANKINGS]
07/12/2015
14/12/2015
21/12/2015
28/12/2015

2016: (RANKING SYSTEM - ACTUAL SYSTEM)
04/01/2016
11/01/2016
18/01/2016
01/02/2016
08/02/2016
15/02/2016
22/02/2016
29/02/2016
07/03/2016
21/03/2016
04/04/2016
11/04/2016
18/04/2016
25/04/2016
02/05/2016
09/05/2016
16/05/2016
23/05/2016
06/06/2016
13/06/2016
20/06/2016
27/06/2016
11/07/2016
18/07/2016
25/07/2016
01/08/2016
08/08/2016
15/08/2016
22/08/2016
29/08/2016
12/09/2016
19/09/2016
26/09/2016
03/10/2016
10/10/2016
17/10/2016
24/10/2016
31/10/2016
07/11/2016
14/11/2016
21/11/2016
28/11/2016 [YEAR END RANKINGS]
05/12/2016
12/12/2016
19/12/2016
26/12/2016

2017: (RANKING SYSTEM - ACTUAL SYSTEM)
02/01/2017
09/01/2017
16/01/2017
30/01/2017
06/02/2017
13/02/2017
20/02/2017
27/02/2017
06/03/2017
20/03/2017
03/04/2017
10/04/2017
17/04/2017
24/04/2017
01/05/2017
08/05/2017
15/05/2017
22/05/2017
29/05/2017
 
#12 ·
Status update on 1982:

1982:
[PRELIMINARY RUN]

This preliminary run has the purpose of employing the rules of the rankings in order to determine a mockup of the stars assigned to tournaments in 1982. This initial model will then be run through the algorithm to determine the actual stars ATP awarded to tournaments. You will notice (if you access the files via OneDrive) that the points will differ significantly from the official figures, although placement of players may not differ by much. The succession of the #1 player is exactly the same as the official records show.
 
#13 ·
Whilst doing the 1982 preliminary run, I discovered a unique set of circumstances which applies after 1981 (because 1981 has a lot of ATP missing data), which would enable me to solve 1982-1985 star systems in one go. For that though, I need a piece of the puzzle which I was not yet able to obtain.

I asked through email already, but I believe it can help even better if more people get involved. If anyone has this possibility, it would move the research on fast forward (well, it would still require some work, but it would be faster). I need as many 1985 intermediary rankings with points (not just averages) as possible. For it to have any effect, perhaps at least one ranking set per month of 1985 should do, and it doesn't have to be more than Top 50, but it does need to show the full amount of points per player.

Hope there is a way here and thanks a million if anyone can do it.
 
#18 ·
Alright, I got huge help here, and I would like to address my thanks to Salvatore Mamola and to @Pete7. You made this all possible.:worship:

First step has already been complete. I started bridging the years. Observations:

- 1984 is not only a ranking reform, it's also a technological leap for the ATP. They basically upgraded their verification method. There's simply no error or very little errors after this.
- As suspected, the 1984-1989 bonus system remained in place without any change (1-5: 30p; 6-10: 24p; 11-15: 20p; 16-20: 16p; 21-30: 12p; 31-50: 6p; 51-75: 3p; 76-100: 2p; 101-150: 1p).
- In 1982, the ATP switched to awarding bonus points according to entry rankings, instead of the immediately preceding rankings. This will force me to not change original post-1982 bonus points, because it's impossible to trace the actual entry rankings the ATP used - the number of weeks between entry and tournament start was NOT standard yet (and since it became a rule, I am forced to go with it).

Here are the two players bridged for now:

McEnroe:
Code:
1985
Philadelphia 10* - 158 - 110 + 48
Houston 9* - 139 - 100 + 39
Milan 9* - 132 - 100 + 32
Chicago 8* - 114 - 90 + 24
Atlanta 9* - 137 - 100 + 37
Forest Hills 16* F - 158 - 128 + 30
Roland Garros 26* SF - 181 - 135 + 46
Wimbledon 27* QF - 71 - 70 + 1
Stratton Mountain 11* - 166 - 120 + 46
Canada 11* - 192 - 120 + 72
US Open 28* F - 289 - 218 + 71
Los Angeles 8* SF - 67 - 45 + 22
San Francisco 7* QF - 35 - 20 + 15
Stockholm 9* - 157 - 100 + 57
Australian Open 18* QF - 67 - 48 + 19
2063 / 2063

1984
Philadelphia 10* - 180 - 110 + 70
Madrid 7* - 164 - 80 + 84
Brussels 8* - 160 - 90 + 70
Roland Garros 24* F - 286 - 188 + 98
Queen's 10* - 155 - 110 + 45
Wimbledon 25* - 326 - 260 + 66
Canada 11* - 190 - 120 + 70
Cincinnati - 1
US Open 26* - 362 - 270 + 92
San Francisco 7* - 129 - 80 + 49
Stockholm 9* - 165 - 90 + 75
Wembley - 0
2118 / 2118

1983
Philadelphia 15* - 186 - 160 + 26
Las Vegas - 1
Roland Garros 30* (?) QF - 75 + 4 = 79 (?)
Queen's 13* (?) F - 105 + 10 = 115 (?)
Wimbledon 36* (?) - 392 - 370 + 22
Canada 17* SF - 92 - 90 + 2
Cincinnati 19* F - 159 - 150 + 9
US Open 38*(?) - 49 (?)
San Francisco 11*(?) F - 104 (?)
Sydney 9*(?) - 103 - 100 + 3
Wembley 13*(?) - 147 - 140 + 7
AUS Open 25* - 132 - 130 + 2
1559 / 1559

1982
Philadelphia 14*(?) - 150 + 14 = 164 
Memphis 12*(?) F - 98 + 21 = 119
Brussels 11* QF - 30
Queen's 9* F - 90 - 75 + 15
Wimbledon 26* F - 208 - 203 + 5
Canada 16* SF - 85 + 2 - 87
Cincinnati 16* SF - 177 - 85 + 5 - 90
US Open 28* SF - 155 - 145 + 10
San Francisco 11*(?) - 136 - 120 + 16
Sydney 9*(?) - 113 - 100 + 13
Tokyo Ind 14*(?) - 162 - 150 + 12
Wembley 10* - 117 - 110 + 7
1471 / 1471
Lendl:
Code:
1985
Delray Beach 22* R16 - 42 - 29 + 13
Fort Meyers 8* - 156 - 90 + 66
Monte Carlo 10* - 183 - 110 + 73
Forest Hills 16* - 233 - 170 + 63
Roland Garros 26* F - 269 - 203 + 66
Wimbledon 27* R16 - 43 - 35 + 8
Indianapolis 11* - 192 - 120 + 72
Stratton Mountain 11* F - 144 - 90 + 54
Canada 11* F - 139 - 90 + 49
US Open 28* - 379 - 290 + 89
Stuttgart Outdoor 4* - 95 - 50 + 45
Sydney Indoor 7* - 115 - 80 + 35
Tokyo Indoor 9* - 202 - 100 + 102
Wembley 9* - 188 - 100 + 88
Australian Open 18* SF - 105 - 95 + 10
2485 / 2485

1984
Philadelphia 10* - 125 - 83 + 42
Brussels 8* F - 126 - 68 + 58
Rotterdam 8* F - 83 - 68 + 15
Luxembourg 7* - 144 - 80 + 64
Monte Carlo 9* QF - 49 - 25 + 24
Roland Garros 24* - 374 - 250 + 124
Queen's 10* - 1
Wimbledon 25* SF - 165 - 130 + 35
Canada 11* - 1
US Open 26* F - 278 - 203 + 75
Sydney Indoor 7* F - 90 - 60 + 30
Tokyo Indoor 9* F - 90 - 75 + 15
Wembley 8* - 173 - 90 + 83
Australian Open 18* R16 - 29 - 24 + 5
1728 / 1728

1983
Philadelphia 15* F - 123 - 120 + 3
Brussels 13*(?) F - 131 - 105 + 26
Milan 16*(?) - 186 - 170 + 16
Monte Carlo - 1
Hamburg (?)- 17
Roland Garros 30* (?) QF - 75 + 3 = 78 (?)
Queen's 13* (?) SF - 70 + 10 = 80 (?)
Wimbledon 36* SF (?) - 179 (?) [Number of points is 179, but it's lower than 185, the correct SF points for 36* - POSSIBLE UNCORRECTED ERROR]
Canada 17* - 190 - 180 + 10
Cincinnati 19* F - 100
US Open 38*(?) F - 327 - 293 + 34
San Francisco 11*(?) - 132 - 120 + 12
Sydney 9*(?) QF - 25
Tokyo Indoor 14*(?) - 152 - 150 + 2
Australian Open 24* - 223 - 188 + 35
1944 / 1944

1982
Palm Springs 12* F - 98
Frankfurt 11*(?) - 127 - 120 + 7
Monte Carlo 14*(?) F - 120 - 113 + 7
Madrid 9* F - 76 - 75 + 1
Roland Garros 24* R16 - 32
Boston 10* QF - 32 - 28 + 4
Washington 11* (?) - 134 - 120 + 14
North Conway 11* (?) - 144 - 120 + 24
Canada 16* F - 149 - 128 + 21
Cincinnati 16* - 190 - 170 + 20
US Open 28* F - 227 - 218 + 9
Barcelona 12* QF - 51 - 33 + 18 (POSSIBLE ERROR: SHOULD HAVE BEEN 36p)
1380 / 1380
The ? marks are in place where the number of stars is not yet determined (but it will be by the end of the bridge process). There are possible errors noted (only in 1982 and 1983), which may not be so.
 
#20 ·
First step has already been complete. I started bridging the years.
That is great news. I am looking forward to your results.

I once tried to deduct the bonus points from total points to arrive at the tournament points in 1984 and 1985, but got contradictory results. Maybe it was because of those entry rankings in bonus point allocation that you have mentioned.
 
#21 ·
1982 and 1983 still seem highly unstable. For once, the 1983 bonus points just don't make any sense whatsoever. It seems like the ATP was experimenting the new bonus system that would start in 1984 in 1983.

1983 could very well have been the start of reform, and I'm not seeing it yet. I need to try calculations from that perspective too.

Until then, this is Connors:

Code:
1985
Philadelphia 10* SF - 77 - 55 + 22
Memphis 9* SF - 79 - 50 + 29
La Quinta 11* QF - 43 - 30 + 13
Fort Myers 8* F - 122 - 68 + 54
Chicago 8* F - 125 - 68 + 57
Las Vegas - 1
Roland Garros 26* SF - 168 - 135 + 33
Queen's - 1
Wimbledon 27* SF - 155 - 140 + 15
Washington 9* SF - 56 - 50 + 6
Stratton Mountain 11* SF - 80 - 60 + 20
Canada 11* - 94 - 60 + 34
US Open 28* SF - 176 - 145 + 31
Tokyo Indoor 9* SF - 60 - 50 + 10
1237 / 1237

1984
Memphis 9* - 140 - 100 + 40
La Quinta 9* - 174 - 100 + 74
Madrid 7* R16 - 22 - 10 + 12
Rotterdam 8* F - 122 - 68 + 54
Boca West 8* - 145 - 90 + 55 ($250000)
Roland Garros 24* SF - 153 - 125 + 28
Queen's 10* SF - 80 - 55 + 25
Wimbledon 25* F - 245 - 195 + 50
Canada 11* SF - 96 - 60 + 36
Cincinnati 12* SF - 91 - 65 + 26
US Open 26* SF - 176 - 135 + 41
Los Angeles 8* - 121 - 90 + 31
Tokyo Indoor 9* - 172 - 100 + 72
Stockholm 9* SF - 94 - 50 + 44
Wembley 8* SF - 64 - 45 + 19
1895 / 1895

1983
Memphis 13*(?) - 165 - 140 + 25
La Quinta 12*(?) R32 - 9
Brussels 13*(?) R16 - 22 - 18 + 4
??? - 1 [LATE WITHDRAWAL PENALTY]
Los Angeles 11*(?) SF - 71 - 60 + 11
Las Vegas 15*(?) - 175 - 160 + 15
Roland Garros 30* (?) QF - 79 - 78 + 1
Queen's 13* (?) - 180 - 140 + 40
Wimbledon 36* R16 (?) - 56 - 47 + 9
Canada 17* SF - 92 - 90 + 2
Cincinnati 19* SF - 101 - 100 + 1
US Open 38* (?) - 437 - 390 + 47
Dallas 10* (?) QF - 28
Tokyo Indoor 14*(?) SF - 77 - 75 + 2
Wembley 13*(?) F - 107 - 105 + 2
1600 / 1600

1982
Philadelphia 14* F - 113
Monterrey 13* - 146 - 140 + 6
Brussels 11* SF - 64 - 60 + 4
Rotterdam 12*(?) F - 106 - 98 + 8
Milan 15*(?) F - 126 - 120 + 6
Los Angeles 10*(?) - 110
Las Vegas 15*(?) - 167 - 160 + 7
Hamburg 14*(?) - 37 - 35 + 2
Roland Garros 24*(?) QF - 63
Queen's 9* - 125 - 100 + 25
Wimbledon 25*(?) - 285 - 260 + 25
Columbus 5*(?) - 66 - 60 + 6
Canada 16*(?) SF - 87 - 85 + 2
Cincinnati 16*(?) SF - 93 - 85 + 8
US Open 29*(?) - 330 - 300 + 30
San Francisco 11*(?) F - 92 - 90 + 2
Sydney 9*(?) SF - 55 - 50 + 5
2065 / 2065
 
#23 ·
Please don't take the above calculations for 1982 and 1983 final. The mathematical model is now hard at work with the data and results will probably be a bit different. 1984 and 1985 look final to me, I will publish the systems though when I get there, it is always safe to take another look when chronology is in place; once 1983 is done, it will be smooth sailing from there to 2012, the point where my live rankings started (with a more than likely break to redo 1968-1970 more realistically, as per elegos7's suggestions.:wink2:
 
#24 ·
I will be upgrading post 4 with the primary result of the mathematical model for 1982, the calendar, the ranking system and ranking bonuses. This time you will notice that points will be identical to the original year end rankings of 1982. This is because original bonus points are now being used as a result of the fact that bonus points were awarded according to entry rankings starting with 1982.

1982: (RANKING SYSTEM|RANKING BONUSES - 1982 ORIGINAL SYSTEM --- CALENDAR)
[UNCONFIRMED]
11/01/1982
18/01/1982
25/01/1982
01/02/1982
08/02/1982
15/02/1982
22/02/1982
01/03/1982
08/03/1982
15/03/1982
22/03/1982
29/03/1982
05/04/1982
12/04/1982
19/04/1982
26/04/1982
03/05/1982
10/05/1982
17/05/1982
24/05/1982
31/05/1982
07/06/1982
14/06/1982
21/06/1982
28/06/1982
05/07/1982
12/07/1982
19/07/1982
26/07/1982
02/08/1982
09/08/1982
16/08/1982
23/08/1982
30/08/1982
06/09/1982
13/09/1982
20/09/1982
27/09/1982
04/10/1982
11/10/1982
18/10/1982
25/10/1982
01/11/1982
08/11/1982
15/11/1982
22/11/1982
29/11/1982
06/12/1982
13/12/1982
20/12/1982
27/12/1982
03/01/1983 [YEAR END RANKINGS]

Here is the current mathematical model calendar:
1982 Calendar Model

Also, you can see the category ranking of tournaments here:
Ranking of tournaments


--------------------------------------------------
And now, to add on the analysis I made in the original thread:

1979:

Perfect match: 6 (2 in Top 10)
Similar match: 15 (4 in Top 10)
Far: 1 (Panatta is actually this, despite yellow color
Errors: 8 (4 in Top 10, 5 in Top 20)

1980:

Perfect match: 7 (3 in Top 10)
Similar match: 19 (6 in Top 10)
Far: 1
Errors: 3 (1 in Top 10, 2 in Top 20)

1981:

Perfect match: 8 (3 in Top 10)
Similar match: 17 (7 in Top 10)
Far: 1
Errors: 4 (0 in Top 10, 0 in Top 20)

1982:

Perfect match: 29 (10 in Top 10)
Similar match: 0 (0 in Top 10)
Far: 0
Errors: 1 (0 in Top 10, 1 in Top 20)

The 1982 is a huge leap from 1981, not because of the perfect matches (which are caused by the fact that I no longer calculated bonus points dynamically, but rather used the originals), but because there is only one single error within the Top 30, and it doesn't even cause a position shift. It's clear the ATP's technology was evolving, 1982 being the first year with a weekly release of the rankings (except for the start of January, where no tournaments were played).

After a similar 1983, the ranking reform will bring forth another giant leap in technology and errors will probably not be present within the Top 100 anymore.
 
#25 ·
The model has been found, in a sense that the Top 100 equation is satisfied. This model also "cures" Higueras (shown above) and now the entire Top 30 is 100% identical to the original rankings. Before I upgrade this to 'CALCULATED' the model must satisfy the equation for the Top 200, then the calendar will be quite final. Take note that this model also shows very little difference between the ATP stars for tournaments and the ones that should have been.

All weekly rankings have been updated.
 
#27 ·
This is going faster than in the previous years thanks to the help I got with the bridge. Gonna be one or two weeks more with the lower ranked players.

In the meantime I felt it was kinda necessary that I show you this:

Rankings | Singles | ATP World Tour | Tennis

Looks average enough, but if you scroll to rank #523 you'll see Pablo Martin-Adalia there, a player who wasn't even born yet. So, this project isn't only for the big guys' records in the Top 10, Top 100, it's also for the little guys who don't get recognized because of an error made by someone. BTW, it should be Pablo Martinez-Catalan there, who was 18 at the time. Not to mention that Ed Nagel is not displayed at all, replaced by Markus Nagel. There are more, of course, but statistics buffs definitely need clean records. Maybe the little guys aren't important for more than 99% of tennis fans, but somebody does care.

Back to work now.:wink2:
 
#29 ·
I have finished the WCT-included run and the results are quite surprising. I did not expect Lendl to control 1982 after July 26th (minus August 9th) - that's 23 weeks, although I did expect him to get some weeks at #1 if WCT is included. But because WCT had 300K tournaments, they would get many stars and are basically counting as major tournaments against their GP counterparts.

To get the tournaments in balance, 16-draw tournaments received a 1-star draw size handicap (32-draw receive 0 bonus stars, so 16-draw had to be hadicapped for balance). Draw difficulty was calculated by a formula that would have been used by the ATP in 1979 had they accepted 16-draw tournaments (the star handicap comes from there too).

So, here is the comparison:

From left to right - official 1982 - WCT-included 1982 - Nixdorf rankings


The week by week rankings will be added to the OP, along with the calendar.
 
#37 ·
Hi Slasher!

First of all, congratulations for this amazing work. Certainly it's a major source of information of the early history of ATP.

Looking at the 1974 season, I have a doubt about the Oslo Open (played in November). You placed it in the week 18-24 november (the same week with Buenos Aires and Johannesburg); however, apparently it was played one week earlier: from 11-17 november, the same week with London, Manila and Bombay.

Bob Hewitt and Joaquin Loyo-Mayo played both the Oslo and Johannesburg tournaments; however, in your Excel file from 25 nov 1974 both players have points only for the Johannesburg result. The same with Rick Fisher and Hans Kary, they played in Oslo and Buenos Aires, but in the Excel file they only got points for Buenos Aires.

Also, some newspapers apparently confirm the week of the Oslo tournament

https://news.google.com/newspapers?...AIBAJ&sjid=APsDAAAAIBAJ&pg=6189,5396157&hl=es

https://news.google.com/newspapers?...AIBAJ&sjid=jGgFAAAAIBAJ&pg=3441,2785145&hl=es

https://news.google.com/newspapers?...AAIBAJ&sjid=n54FAAAAIBAJ&pg=1017,127964&hl=es

So, if you could clarify this situation, or eventually correct it, it would be really appreciated.

Thanks!
 

Attachments

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top