PDA

View Full Version : 1982 - How is this possible?


Benhur
03-18-2008, 03:38 PM
I've just realized that McEnroe got the year-end number 1 in 1982. I'd always assumed Connors had that, on account of his Wimbledon and US Open wins.

http://www.tennis-x.com/stats/atprankhist.shtml

I find this completely incomprehensible. Mind-boggling. Subreal. McEnroe was clearly not even number 2 that year, since Lendl had far, far better results than him, entering 23 tournaments and reaching the finals of 20 of them, ending the year with 15 titles (including 9 WCT + Master's) and 5 runner-ups -- to McEnroe's 5 titles and 5 runner-ups. McEnroe didn't win any slams, and none of his 5 titles were even WCT. In addition, he lost to Lendl all 4 times they played that year, including the US Open SF and the Master's final.

Does anybody have a clue what the rationale was behind the ranking system that came up with McEnroe as number one for that year? Did they pull his name out of a hat? It makes no sense whatsoever.

Moose Malloy
03-18-2008, 03:50 PM
I posted this in the '77 thread - WCT events & the Year End Masters did not count for ATP Ranking pre 1990.

Mac finished at #1 with a 122.58 pts average for the ATP ranking & Connors finished with a 121.47 average for #2, a very close margin. Lendl had 115.

In the Year in Review Issue of Tennis magazine for 1982, they have an interview with Connors in which they ask him how it feels to be #1 again, which is funny since he technically wasn't. I guess the players/media knew well enough to not care too much about it back then.

Mac in interviews at the time sounded like he considered Connors #1 for '82.

Connors did finish #1 in the Volvo Grand Prix Standings that year (which determined the field for the Masters) Vilas was 2, Lendl was 3.

Benhur
03-18-2008, 04:43 PM
I posted this in the '77 thread - WCT events & the Year End Masters did not count for ATP Ranking pre 1990.

Mac finished at #1 with a 122.58 pts average for the ATP ranking & Connors finished with a 121.47 average for #2, a very close margin. Lendl had 115.

In the Year in Review Issue of Tennis magazine for 1982, they have an interview with Connors in which they ask him how it feels to be #1 again, which is funny since he technically wasn't. I guess the players/media knew well enough to not care too much about it back then.

Mac in interviews at the time sounded like he considered Connors #1 for '82.

Connors did finish #1 in the Volvo Grand Prix Standings that year (which determined the field for the Masters) Vilas was 2, Lendl was 3.

I'd always thought that WCT events and the Master's were the most important tournaments after the slams in the 80s. If they don't count for anything, is there any validity to the ATP rankings in those years? Was it like a parallel system? Somehow this does not make sense.

jeffreyneave
03-19-2008, 04:46 AM
The Atp rankings are a joke. As noted Lendl, won 15 out of 23 events. However, according to the Atp he only played 12. ie neither his wins in the WCt finals, Masters or any other WCT event counted. There was a political war between ATP anD WCT in 1982.

Mcenroe did well partly because the ATp was an average system (not 14 best plus majors), finalists got 75% (he was runner-up at wimbledon) and semi-finalists 50% (a semi-finalist at US open) devaluing wins a bit, but primarily because the slams only had about a 1/3 mark-up on the next tier of events. Now the mark up is 100%. With today's ratio for slams, Connors would be an easy winner over Mcenroe.

jeffrey

jeffrey

Tennis old man
03-30-2008, 08:13 AM
and in 75 and in 77 same problems...

Gizo
03-30-2008, 09:02 AM
McEnroe in 1982 is weakest year end no. 1 since the ATP computer rankings were introduced.

Wuornos
03-31-2008, 07:25 AM
Deleted as found unuseful by board members. :cry:

jeffreyneave
03-31-2008, 10:47 AM
Are these 1982 and 1975 results purely based on the 4 grand slams



Are these rankings weighted towards participation or an average preformance

ie if you play 4 evewnts and your results are w,w, f ,sf

do you score more than a 2 event player whose results are w,w

Since your ranking are based purely on the quality you beat what are the scores for beating no1, no2, no10,no20,no 50 ,no100 in these two years


jeffrey

Wuornos
03-31-2008, 11:25 PM
Deleted as found unuseful by board members. :cry:

jeffreyneave
04-01-2008, 10:46 AM
Your amswers are clear.

First about 1982 . Wilander should out score KRIEK EASILY. His win at the French open where he beat 4 top ten players should be worth much more than either of kriek's wins. Kriek only beat mcnanmee (top 40) and denton (top 20) at the 82 Aus. Actually kriek's 81 Aus came thru a much stronger field (vilas , Mcnamara and Tanner) although he only played Mayotte (top 30), Edmondson (top 20) Denton again (top 50). I suspect that denton becu he did well at the 81 aus was a worth a lot more points in 82 and boosted kriek's score. its not very acceptable logic. You really can't just give points on the basis of who you play; the original strength of field counts as well.

Wilander backed up his french win with 2 L16 at wimbledon and Us; kriek backed up his aussie with a qf at Wimb. I really can't see why Wilander is not way ahead of kriek.

The other problem with these rankings in the open era is that they are solely based on slam performance and given the wealth of knowledge available about other results in these years you can't ignore them. Lendl' wins at the masters and WCT finals are regarded as far more significant than kriek's wins at the Aussie open. Of the rankings I have seen Lendl is always at least 2 or 3; Kriek is not in the top ten.

When doing rankins you always have to take notice of the historical context.
In 1975 the WCT was a major event; far more signicant than the aussie open. The 1975 masters was a big event. Thus rankings of the time reflected the fact that that Nastase would be ranked ahead of Necombe and Ashe ahead of Connors. Ashe's two big wins and at L16 are better than connor's 3 runner-ups

Open era rankings can not just be done on the 4 slams; all the other tournaments in a year count as well. This stops historical comparisons bceause there is no way you can get the drawa of all pre-open era events.

You mentioned that Trabert did well in comparison with open era peaks. I presume you are referring to his 1955. He turned pro in 1956 and was well beaten by Gonzales, Sedgman and Segura.: in otherwords he was really only the world number 4. So to do a comparison with the open era you would have to knock off the points share of a strong top 3 from 1955; eg the share today of Fedrerer, Nadal and Djokovic or say the share of Samparas, Agassi and Becker in 1994/95.

laurie
04-01-2008, 11:13 AM
I've never been into the calculation of ranking points - I do know that ranking systems are always controversial. Mcenroe finishing number 1 in 1982 indeed does seem strange. When the ATP really took over matters in 1990 is that when they gave slams more weight in terms of points?

Also, I remember in 2000 Hingis finsihed the year as number 1 without winning a slam and that was pretty controversial. Venus won Wimbledon, the US open and an Olympic gold medal that year - although I remember Venus had wrist problems the first few months of 2000.

Wuornos
04-01-2008, 11:54 AM
Deleted as found unuseful by board members. :cry:

jeffreyneave
04-01-2008, 12:34 PM
stop being rude. your original gives no worked out examples. and is not at all clear. put in simple english not statistical jargon

As far as I'm concerned your points for each match are 2* rating of opponent player. that's all.; then you add of the points for each player across each match; nothing else.

Wilander's win should be worth a lot more than krieks at the Aussie. Show your workings from scratch for kriek and wilander's win. if wilander is not higher your rankings are nonsense ; look at every journalist at the time wilander is top 5 and kriek outside the top ten

you don't know anything about 1982 history do you . I've seen loads of journalists rqankings and the wct and masters are very importqant.; the aussie is not. kriek's 81 should be better than kriek's '82 if the system is based on the quality of the entry field.

Ashe was ranked top by every journalist at the time . they knew the historical reality ie the wct counted a lot more than the aussie open. neither you or the atp do.


.jeffrey

Wuornos
04-01-2008, 01:12 PM
Deleted as found unuseful by board members. :cry:

Moose Malloy
04-01-2008, 01:48 PM
put in simple english not statistical jargon


But that is his background, & the whole point of his systems is to create a statistical database that calculates best players based on elo system, which was used in chess. see link below.

I don't entirely understand Wuornos sometimes, nor always agree with his rankings(& agree he should use 'layman's terms' more often when posting here, I'm sure most have no idea what he's saying most of the time), but I appreciate his efforts. It must have been a momumental task, creating a database like this.

Its like the 2 of you are speaking a different language & fighting without understanding what the other is really saying. Maybe you'll understand him a bit better if you read this, no need to fight:

http://en.wikipedia.org/wiki/Elo_rating_system

you don't know anything about 1982 history do you . I've seen loads of journalists rqankings and the wct and masters are very importqant.;

again this is irrelevant(though I agree with you, that stuff is important), we are talking about a database that is creating these numbers, not his personal opinion, etc.

and I don't think its possible to give out worked out examples that you want, the data from one year creates the next & on & on & the next player, the next. You beat one player, he beats so & so, it just seems incredibly complicated, far more than any atp ranking system I don't think he can just separate it neatly to 1982 AO, like you want. At least that is my understanding.

there was a website not too long ago that attempted this method as well, but unfortunately it is now gone. he explained it better.

jeffreyneave
04-02-2008, 12:22 PM
If you are not using indvidual player quality, you must be using overall quality of th3 field. and then you are presumably distributing points according to this 50% distribution to produce the enormous differential between no1 and no 10.


ie 3,6,12,24,50,100 (rounded) points progression for 64 man field compared to

ATP progreesion of

7,15,25,45,70,100

Tennis has never used the 50% progression, whether it be pros 1950s and 60s, WCT 1970s, Grand prix 1970s or ATP of today. My preferred dist for a 32 man field is the old WCT method ie

1,2,4,7,10

compared to the 50% which produces

1,2,4,8,16

In otherwords your method is very different for winning the final match. You get an extra 8 for winning one extra match: that's wrong. Both players have won 4 matches to win 7/8 points. Winning one extra match is no where near worth the equivalent of 4 previous wins even if you are playing a better player. both systems give higher points for each win as the players get stronger, but one final match is not worth winning say 6 which
be the case in a grand slam final.; its only one match and beating your opponent should not give you all his points from his previous wins. that's why tennis has never done so.

As to 1982 , Connors has obviously earned about 220-250 points for his 2 wins; Wilander's win at the French should be worth about the same. As far as I'm concerned Wimbledon had the weakest field of the 3 (clerc, vilas lendl of top players missing) , French slightly stronger (Mcenroe, gene Mayer missing) and Us open the strongest (Mcnamarra only mising of top ten in 1982). any realistic rating of the French in 1982 would say it desrved at least as much as Wimbledon: that's why Wilander was top 5 and kriek was nowhere near.

jeffrey.

Wuornos
04-02-2008, 11:37 PM
Deleted as found unuseful by board members. :cry:

jeffreyneave
04-03-2008, 10:15 AM
I don't think winning tennis matches is the lottery of spinning a coin. I can see why you give such high rewards if you think winning 4 four matches to reach a final is the same as getting 4 heads in a row; then winning a fifth in the final is like getting the fifth head.

Players who reach finals and win like federer have win percentages of 90%. Players who reach finals on a regular basis are not in a coin spin; they are almost certain to win; players who have 50% chances of winning matches don't win touranemts and hardly ever reach finals -see marty fish.

1,2,4,7,10 progression is to reflect the rising quality of player faced not that he's being incredibly lucky to win 3 50% coin tosses in a row. The progression for each match is 1,1,2,3,3 and thats good approximation of etxra reward for a winning a more difficult match. although i admit i am also in favour when doing top 10 rankings of giving bonuses for wins against top players which would represnt about 25% of the total points awarded.

jeffrey

Wuornos
04-05-2008, 06:00 AM
Deleted as found unuseful by board members. :cry:

Wuornos
04-08-2008, 02:36 AM
Deleted as found unuseful by board members. :cry: