PDA

View Full Version : New open era weighted rankings based on masters 1000 and higher events


timnz
09-10-2012, 10:12 PM
Like to see us talk about Slams + Season end finals + masters 1000 rather than just Slams, when it comes to evaluating players careers. The season end finals is now a tournament with a rich and strong tradition with great depth of players (over 40 years and top 8 respectively) and the masters 1000's or equivalents pre-1990 have very deep fields. Also there is the WCT finals to consider.

How to go somewhere to creating a level playing field between current players who tend to play 4 slams a year vs older players of the 70's and early 80's who tended to play only 3 slams a year.

Players pre-mid 1985 tending to only play 3 Slams a year versus players today playing 4. There is also the other issue of the WCT finals which was a very important event and the need to include it. Players shouldn't get 6 events where they can gain points in this methodology because that would be unfair to modern players who only get 5 events where they can gain points. The solution proposed is to ONLY include Dallas if a player who won the WCT finals didn't play all the slams in that year. That way the modern players are not disadvantaged. So for example, Lendl's 1982 WCT finals win gets included because he didn't play all the slams that year but his 1985 win doesn't get included because he played all the slams that year. In McEnroe's case 4 out of 5 of his WCT finals get included as he played all the slams in 1983 when he won the 1983 Dallas event.

Weightings
Slams + Season End Finals and WCT finals (only if the player didn't play all the Slams that year) + Losing Finals in Slams + Masters 1000 equivalents, with a weighting factor depending on the importance of the event ie 2 x for slams, 1.5 for Season end finals (including WCT finals), 1.2 for Losing slam finals, 1 x for Masters 1000 equivalents

OP Edit UpdateI have changed the weighting of the season end finals from 1.5 to 1.4. The reason for this is that not all of the Masters Cup winners won the tournament in an unbeaten fashion. Forinstance 1 of Federer's 6 wins he lost a match in the round robin. In 2001 Hewitt was an unbeaten winner but in 2002 is lost one match. No one has lost more than 1 match and gone on to win the tournament - so I thought on average then we could weight it half way between an unbeaten winner and a one match loser but overall winner - to arrive at 1.4. (Currently in the ATP each round robin win is worth 200 points).

Calculations

Federer = (17 x 2) + (6 x 1.4) + (7 x 1.2) + (21 x 1) = 71.8

Lendl = (8 x 2) + ((5 +1) x 1.4)) + (11 x 1.2) + (22 x 1) = 59.6

Sampras = (14 x 2) + (5 x 1.4)) + (4 x 1.2) + (11 x 1) = 50.8

Nadal = (11 x 2) + (0 x 1.4) + (5 x 1.2) + (21 x 1) = 49

McEnroe (7 x 2) + ((3 + 4) x 1.4)) + (4 x 1.2) + (19 x 1) = 47.6

Borg = (11 x 2) + ((2 + 1) x 1.4)) + (5 x 1.2) + (15 x 1) = 47.2

Connors = (8 x 2) + ((1 + 2) x 1.4)) + (7 x 1.2) + (17 x 1) = 45.6

Agassi = (8 x 2) + (1 x 1.4) + (7 x 1.2) + (17 x 1) = 42.8

Becker = (6 x 2) + ((3 +1) x 1.4)) + (4 x 1.2) + (13 x 1) = 35.4

Djokovic = (5 x 2) + (1 x 1.4) + (4 x 1.2) + (13 x 1) = 29.2

Edberg = (6 x 2) + (1 x 1.4) + (5 x 1.2) + (8 x 1) = 27.4

Wilander = (7 x 2) + (0 x 1.4) + (4 x 1.2) + (8 x 1) = 26.8

timnz
09-10-2012, 10:13 PM
His US open final means he is finally ahead of Edberg.

jokinla
09-10-2012, 10:22 PM
That's a pretty good order there, I know your list is based on calculations, but I'd have to switch Lendl and Sampras, Borg and McEnroe and Djoker and Edberg.

timnz
09-10-2012, 10:35 PM
That's a pretty good order there, I know your list is based on calculations, but I'd have to switch Lendl and Sampras, Borg and McEnroe and Djoker and Edberg.

My instincts say the same, but when I analyse them it is because i tend to discount slam finals and masters 1000's. But why should those achievements be forgotten?

I'm curious about your ratings of djokovic and edberg. (obviously you are completely entitled to them). Just would like to understand your thinking.

jokinla
09-11-2012, 10:01 AM
My instincts say the same, but when I analyse them it is because i tend to discount slam finals and masters 1000's. But why should those achievements be forgotten?

I'm curious about your ratings of djokovic and edberg. (obviously you are completely entitled to them). Just would like to understand your thinking.

I don't think the finals and Masters are discounted, Edberg got to #1 and was the man for a bit there, so those tourneys are factored into that ranking, but as a career I favor him, more majors and #1 ranking, when Djoker is finished he might pass him, but I won't put him above Edberg now because he had more finals or Masters or whatever, that will get worked into the ranking points, and Edberg did ascend to a higher level. To me you have to draw the line somewhere, majors are more important than Masters, and Masters are more important than 500's, etc., so we judge on majors, and all the other tourneys factor into the points that give you your ranking, so Edberg did more in the other tourneys, whatever they were, than Djoker has. Your system is what it is when you just include what you did, but now you discount the other stuff, which isn't as important, but to take someone's career into account, you have to take it all into account, so as a whole, Edberg above Djoker, but of course Djoker isn't finished yet.

fed_rulz
09-11-2012, 10:04 AM
Like to see us talk about Slams + Season end finals + masters 1000 rather than just Slams, when it comes to evaluating players careers. The season end finals is now a tournament with a rich and strong tradition with great depth of players (over 40 years and top 8 respectively) and the masters 1000's or equivalents pre-1990 have very deep fields. Also there is the WCT finals to consider.

How to go somewhere to creating a level playing field between current players who tend to play 4 slams a year vs older players of the 70's and early 80's who tended to play only 3 slams a year.

Players pre-mid 1985 tending to only play 3 Slams a year versus players today playing 4. There is also the other issue of the WCT finals which was a very important event and the need to include it. Players shouldn't get 6 events where they can gain points in this methodology because that would be unfair to modern players who only get 5 events where they can gain points. The solution proposed is to ONLY include Dallas if a player who won the WCT finals didn't play all the slams in that year. That way the modern players are not disadvantaged. So for example, Lendl's 1982 WCT finals win gets included because he didn't play all the slams that year but his 1985 win doesn't get included because he played all the slams that year. In McEnroe's case 4 out of 5 of his WCT finals get included as he played all the slams in 1983 when he won the 1983 Dallas event.

Weightings
Slams + Season End Finals and WCT finals (only if the player didn't play all the Slams that year) + Losing Finals in Slams + Masters 1000 equivalents, with a weighting factor depending on the importance of the event ie 2 x for slams, 1.5 for Season end finals (including WCT finals), 1.2 for Losing slam finals, 1 x for Masters 1000 equivalents

Calculations

Federer = (17 x 2) + (6 x 1.5) + (7 x 1.2) + (21 x 1) = 72.4

Lendl = (8 x 2) + ((5 +1) x 1.5)) + (11 x 1.2) + (22 x 1) = 60.3

Sampras = (14 x 2) + (5 x 1.5)) + (4 x 1.2) + (11 x 1) = 51.3

Nadal = (11 x 2) + (0 x 1.5) + (5 x 1.2) + (21 x 1) = 49

McEnroe (7 x 2) + ((3 + 4) x 1.5)) + (4 x 1.2) + (19 x 1) = 48.3

Borg = (11 x 2) + ((2 + 1) x 1.5)) + (5 x 1.2) + (15 x 1) = 47.5

Connors = (8 x 2) + ((1 + 2) x 1.5)) + (7 x 1.2) + (17 x 1) = 45.9

Agassi = (8 x 2) + (1 x 1.5) + (7 x 1.2) + (17 x 1) = 42.9

Becker = (6 x 2) + ((3 +1) x 1.5)) + (4 x 1.2) + (13 x 1) = 34.3

Djokovic = (5 x 2) + (1 x 1.5) + (4 x 1.2) + (12 x 1) = 28.3

Edberg = (6 x 2) + (1 x 1.5) + (5 x 1.2) + (8 x 1) = 27.5

Wilander = (7 x 2) + (0 x 1.5) + (4 x 1.2) + (8 x 1) = 26.8

the weightings mirror todays ranking points (scaled down by 1000)? if so, here's a thought -- the 1500 WTF points are maximum possible, achieved only if you win all RR matches. do you want to factor that in?

Agassifan
09-11-2012, 11:13 AM
Nice, but slams should weigh a whole lot more.

Slam: 10
Olympic Gold: 5
Slam final: 4
WTF: 4
Masters: 2

Although including OG makes sense only the past couple of olympics when players have become semi-serious.

mistik
09-11-2012, 11:18 AM
Nice, but slams should weigh a whole lot more.

Slam: 10
Olympic Gold: 5
Slam final: 4
WTF: 4
Masters: 2

Although including OG makes sense only the past couple of olympics when players have become semi-serious.

What about giving more points to career slam winners ? Lendl won less majors then Rafa.He never manage to win Wimbledon.

timnz
09-11-2012, 11:29 AM
Nice, but slams should weigh a whole lot more.

Slam: 10
Olympic Gold: 5
Slam final: 4
WTF: 4
Masters: 2

Although including OG makes sense only the past couple of olympics when players have become semi-serious.

This does not correspond to present atp rank weightings. For instance olympic gold isnt even as much as a masters 1000. And slams according to the atp are 2 x masters not 5 x as you have here. I actually think slams should be weighted much more, but the rankings are what they are.

mattennis
09-11-2012, 03:29 PM
Well, it is just one of the countless "numbers" you can use to "try to compare" players from different eras.

Yours is nice, but it should at least include points for Year-End-Nš1.

For example, Lendl or McEnroe 4 Year-End-Nš1 are much more important than Nadal 2 or Agassi 1 Year-End-Nš1.

I would add something like 5 points for each Year-End-Nš1 a player achieved.

That way the usually two most important things (nš of GS and nš of Year-End-Nš1) would be represented in the computation.

timnz
09-11-2012, 06:42 PM
Well, it is just one of the countless "numbers" you can use to "try to compare" players from different eras.

Yours is nice, but it should at least include points for Year-End-Nš1.

For example, Lendl or McEnroe 4 Year-End-Nš1 are much more important than Nadal 2 or Agassi 1 Year-End-Nš1.

I would add something like 5 points for each Year-End-Nš1 a player achieved.

That way the usually two most important things (nš of GS and nš of Year-End-Nš1) would be represented in the computation.

If I went that way then rather than 5 points for each year end number 1 - I would give 5 points for each 52 weeks a player had. That would avoid controversies over silly rankings - like 1982 with McEnroe being year end number one when it is very clear to all parties that he was no better than 3 that year (he was very clearly behind Connors and Lendl on any sane criteria). But is number 1 an achievement in this sense?- I guess I set out to do these rankings to based on what players actually won. Number 1 comes along with they win enough relative to other players around them. It is an indirect consequence of winning not direct. Also it is a kind of double counting. A player got to number 1 through Slams won and Slam finals reached, plus year end championships and Masters 1000's won..... but we already have counted those.

I'll reflect on this.

clayman2000
09-11-2012, 07:52 PM
If I went that way then rather than 5 points for each year end number 1 - I would give 5 points for each 52 weeks a player had. That would avoid controversies over silly rankings - like 1982 with McEnroe being year end number one when it is very clear to all parties that he was no better than 3 that year (he was very clearly behind Connors and Lendl on any sane criteria). But is number 1 an achievement in this sense?- I guess I set out to do these rankings to based on what players actually won. Number 1 comes along with they win enough relative to other players around them. It is an indirect consequence of winning not direct. Also it is a kind of double counting. A player got to number 1 through Slams won and Slam finals reached, plus year end championships and Masters 1000's won..... but we already have counted those.

I'll reflect on this.

No 1 is flawed.

Look at Lleyton Hewitt. Dont get me wrong, he was the best player on tour for 2 years, but it was 2 weak years

timnz
09-11-2012, 09:01 PM
the weightings mirror todays ranking points (scaled down by 1000)? if so, here's a thought -- the 1500 WTF points are maximum possible, achieved only if you win all RR matches. do you want to factor that in?

You have a point here. I am going to suggest that I instead weight them at 1.4 x instead of 1.5 x - reasons below:

Winners of the end of season championships tend to either go unbeaten (5 out of 6 of Federer's wins have been as an unbeaten player) or only have the loss of 1 match. I don't think there has ever been a player who has lost 2 matches and gone on to win the whole thing (if there has been it must be very much the exception). Currently an unbeaten winner gets 1500 points and a winner of the event who loses 1 match in the round robin gets 1300 points.
Hence, if you average winners of the event of the years it will be a mixture of unbeaten winners and those who lost 1 match. Hence, 1400 points would be the average (in today's ATP terms). Therefore a weighting factor of 1.4 X

I will weight the WCT finals winners (for those who didn't compete in all the slams that year) similarly.

cork_screw
09-11-2012, 09:10 PM
You should set them more like how they are weighted pointwise. The GS should be double the pts that the 1k tournies are because that is how they are adjusted pointwise. WTF should mean a little more than 1k tournies because you need to qualify and only the top tier of players get in. On top of that, you play round robin with the top players, so you don't really get any chances to let down even if it isn't elimination playing most of the top seeds in one tournie is pretty challenging. I think nadal should be higher in the point total.

timnz
09-11-2012, 09:24 PM
Player Slam wins Weighting Season end Championship WCT Finals Weighting Slam Runner-up Weighting Masters 1000 equivalents Weighting Total
Federer 17 2 6 0 1.4 7 1.2 21 1 71.8
Lendl 8 2 5 1 1.4 11 1.2 22 1 59.6
Sampras 14 2 5 0 1.4 4 1.2 11 1 50.8
Nadal 11 2 0 0 1.4 5 1.2 21 1 49
McEnroe 7 2 3 4 1.4 4 1.2 19 1 47.6
Borg 11 2 2 1 1.4 5 1.2 15 1 47.2
Connors 8 2 1 2 1.4 7 1.2 17 1 45.6
Agassi 8 2 1 0 1.4 7 1.2 17 1 42.8
Becker 6 2 3 1 1.4 4 1.2 13 1 35.4
Djokovic 5 2 1 0 1.4 4 1.2 12 1 28.2
Edberg 6 2 1 0 1.4 5 1.2 8 1 27.4
Wilander 7 2 0 0 1.4 4 1.2 8 1 26.8

This makes McEnroe and Borg very close in ranking. Borg had 4 more slams but McEnroe had many more season end finals and more Masters 1000 equivalents.

timnz
09-11-2012, 09:27 PM
No 1 is flawed.

Look at Lleyton Hewitt. Dont get me wrong, he was the best player on tour for 2 years, but it was 2 weak years

Yes, but with respect you could get into a fight about which years are weak and which years are strong. The fact that very injured Hewitt can take still take sets off peak Djokovic tells me that at full flight that Hewitt was really something.

Russeljones
09-11-2012, 09:35 PM
Slams should weigh more. I think we should ultimately think about an index for the Pre-Open Era guys. Will be much harder.

timnz
09-11-2012, 10:01 PM
Slams should weigh more. I think we should ultimately think about an index for the Pre-Open Era guys. Will be much harder.

Tbey should be weighted more, but they are not. I used the current ATP weightings. I did this to get away from people's opinion. The fact that many people tell me that the Slams should weigh more tells me that the ATP itself needs writing to. But in the meantime the weighting which I have used is the one the ATP itself uses.

Hence, according to ATP weightings Lendl is more successful than Sampras or Borg or Nadal. It's not my opinion, it's fact. In fact my opinion is that Lendl shoudn't rate higher - but the weightings of the events are what they are.

roundiesee
09-11-2012, 10:04 PM
Great post!
OP- I was thinking about the "Grand Slam Cup"; I think this was held from 1990-99 in Germany, organised by the ITF it seems.
http://en.wikipedia.org/wiki/Grand_Slam_Cup
Should this be included, or was it more like a glorified exo? I think Pete won it twice and Boris once, so these tournaments might add some points for them if included in your calculations.

timnz
09-11-2012, 10:06 PM
Great post!
OP- I was thinking about the "Grand Slam Cup"; I think this was held from 1990-99 in Germany, organised by the ITF it seems.
http://en.wikipedia.org/wiki/Grand_Slam_Cup
Should this be included, or was it more like a glorified exo? I think Pete won it twice and Boris once, so these tournaments might add some points for them if included in your calculations.

My personal view is that it should be included. But when I had a similar thread in the Former Player forum - people didn't like this event to be included. In my view it was actually stronger than the WCT finals of the 1980's. I saw most of the Grand Slam cups in the 1990's and it was a highly contested affair with Semi's and Finals being best of 5 sets. And players played their hearts out to win it. It only got depreciated in people's minds because at the time the ATP didn't recognize it. But in retrospect the ATP have now included it in their tournament history.

The other thing to note - due to the level playing field rule of only including the top 5 events a player played in a year + masters 1000 equivalents - I think that Boris played all the Slams in 1996 and Pete Played all the slams both years he won (I might be wrong about that - perhaps in 1990 he didn't play all of them).

roundiesee
09-11-2012, 10:51 PM
Thanks! I think your list seems pretty fair based on those calculations. :)

mattennis
09-11-2012, 11:44 PM
If I went that way then rather than 5 points for each year end number 1 - I would give 5 points for each 52 weeks a player had. That would avoid controversies over silly rankings - like 1982 with McEnroe being year end number one when it is very clear to all parties that he was no better than 3 that year (he was very clearly behind Connors and Lendl on any sane criteria). But is number 1 an achievement in this sense?- I guess I set out to do these rankings to based on what players actually won. Number 1 comes along with they win enough relative to other players around them. It is an indirect consequence of winning not direct. Also it is a kind of double counting. A player got to number 1 through Slams won and Slam finals reached, plus year end championships and Masters 1000's won..... but we already have counted those.

I'll reflect on this.

Exactly. If you count total weeks (or Year-End-Nš1) you are counting everything (because to be Nš1 you compute GS, WTF, M-1000, ATP-500, ATP-250 and now even Davis Cup), so an even better way (to look at how a player did against his field, by the way that is the only possible way of comparing different eras, comparing what a player did relative to his field) would be to look JUST at total weeks at Nš1 or total Year-End-Nš1.

Russeljones
09-12-2012, 12:59 AM
Tbey should be weighted more, but they are not. I used the current ATP weightings. I did this to get away from people's opinion. The fact that many people tell me that the Slams should weigh more tells me that the ATP itself needs writing to. But in the meantime the weighting which I have used is the one the ATP itself uses.

Hence, according to ATP weightings Lendl is more successful than Sampras or Borg or Nadal. It's not my opinion, it's fact. In fact my opinion is that Lendl shoudn't rate higher - but the weightings of the events are what they are.

The word 'weighted' in the thread title mislead me. You only added up ranking points. I expected this to be an attempt at comparing cross-era players.

timnz
09-12-2012, 01:53 AM
The word 'weighted' in the thread title mislead me. You only added up ranking points. I expected this to be an attempt at comparing cross-era players.

Sorry if it wasn't clear for you

Weighted refers to the weighting given events having 1000 points or more.

Masters 1000 = x 1

Slam finalist = x 1.2

Season end finals = x 1.4 (because winners generally get either 1500 or 1300 points)

Slams = x 2

These are exactly in line with current atp point allocations.

Regarding eras i said in the op. title that it was for the open era, so I don't understand your confusion.

Russeljones
09-12-2012, 01:58 AM
Sorry if it wasn't clear for you

Weighted refers to the weighting given events where 1000 points or more.

Masters 1000 = x 1

Slam finalist = x 1.2

Season end finals = x 1.4

Slams = x 2

Regarding eras i said in the op. title that it was for the open era, so I don't understand your confusion.

A lot of players and competition types missing. I am not saying the effort you put in isn't appreciated, just reiterating the word 'weighted' was used incorrectly and lead to my confusion.

kOaMaster
09-12-2012, 02:03 AM
this list shows how good actually nadal is. if it wouldn't be for federer, he would be up in the whole goat-discussion for sure.

timnz
09-12-2012, 02:04 AM
A lot of players and competition types missing. I am not saying the effort you put in isn't appreciated, just reiterating the word 'weighted' was used incorrectly and lead to my confusion.

What top players from the open era are missing in your thinking?. I have shown the top 12.

People don't tend to include events of lesser value than the masters 1000 in their asssessment of greatness. Hence that is why i only included events of masters 1000 and up. In your opinion are there any important events missed (your 'competition types')

merlinpinpin
09-12-2012, 02:09 AM
this list shows how good actually nadal is. if it wouldn't be for federer, he would be up in the whole goat-discussion for sure.

Not really, when you factor in the fact that, from this list of all-time greats, Federer is the only one that is in the GOAT discussion (because, you know, a few oldies really *could* play...) ;)

timnz
09-12-2012, 02:23 AM
Exactly. If you count total weeks (or Year-End-Nš1) you are counting everything (because to be Nš1 you compute GS, WTF, M-1000, ATP-500, ATP-250 and now even Davis Cup), so an even better way (to look at how a player did against his field, by the way that is the only possible way of comparing different eras, comparing what a player did relative to his field) would be to look JUST at total weeks at Nš1 or total Year-End-Nš1.

We already have that - see the weeks at number 1 section of

http://en.wikipedia.org/wiki/List_of_ATP_number_1_ranked_singles_players

But you end up with some odd results eg Roddick having more weeks at number 1 than Becker. McEnroe being number 1 , 1.7 x the length of Borg.

Hence I dont think measuring by weeks at number 1 is the way to measure career Achievement. I think simply measuring the top events is the way to go.

timnz
09-12-2012, 02:35 AM
I think nadal should be higher in the point total.

Why, can you let us know your thinking on this?

Monsieur_DeLarge
09-12-2012, 02:46 AM
The overall order is very close to how I'd place them overall, and even the scores match my own opinion quite closely. Then again, I've always assessed players' legacies quite objectively using their achievements, so perhaps all you've done here is quantify my own thought processes. Where I disagree with positions and/or scores, there's an obvious reason why:

Lendl: too high. His eleven losing finals count towards his legacy here more that I'd qualitatively allow. I think it's a lot to do with when he lost: if you have a pile-up of slam final losses before you start winning, people won't count them as positives. Murray's reputation to date seems to have suffered the same way.
Borg: too low. I suspect that's to do with his abbreviated career; retiring at 26 stopped him racking up better numbers, even if only as a slam loser or masters-equivalent winner. Historically he's assessed not only on his numbers of wins, but his winning percentage and his three consecutive channel slams, two achievements which score a zero in this list.
Sampras: too close to Nadal. According to this list, Rafa only needs to win three clay masters next year and he'll jump above Pete, regardless of what he does in future slams. That's obviously not the case. I think it's been skewed by Pete focusing on slam wins, while Nadal has made a career out of dominating the clay masters events to an incredible level. It might also be to do with the fact that weeks/years at #1 isn't rewarded on this list, and that's a stat where the American is well ahead of the Spaniard.


Of course, it's much easier to point out imperfections than to actually come up with solutions... ;) Despite the flaws, I'd leave this list as-is; it's a good statistical comparator, and by using the ATP's numbers it is very transparent and objective.


Regards,
MDL

mattennis
09-12-2012, 05:02 AM
Anyway, be it nš of GS, or Year-End-Nš1, or whatever, it only makes sense when comparing players from the same era.

Why?, because one player only played against the players from his era, in the conditions of his era.

Had Wimbledon and Australian Open been so slow in the 80s as they are today, do you think Lendl and Wilander would be at 8 and 7 GS respectively? They would probably have won many more GS each one of them.

Other examples: today you have to compete in the M-1000 because you earn a 0-point if you don't and the points you earn there will count to your total points of the year.

In the 90s, 80s, 70s....you didn't need to play those tournaments, you didn't get a 0-points if you did not play, so you could play with another mindset, knowing that you could win points in other tournaments (for example, Muster and many others earlier, won something like 9 non-M-1000 tournaments in one season, and all them counted in the points, so no matter for him if he lost in earlier rounds in many M-1000).

Every era had its own peculiarities, players, surfaces distribution, rules, etc...

So it does not make much sense to compare numbers from players from different eras, but we all like to do it. :)

jonnythan
09-12-2012, 05:23 AM
Has anyone ever calculated the total number of career tour points won by these guys?

jokinla
09-12-2012, 08:36 AM
Has anyone ever calculated the total number of career tour points won by these guys?

I think Connors might win that seeing as he won a boatload of tourneys.

timnz
09-12-2012, 11:09 AM
Lendl: too high. His eleven losing finals count towards his legacy here more that I'd qualitatively allow. I think it's a lot to do with when he lost: if you have a pile-up of slam final losses before you start winning, people won't count them as positives. Murray's reputation to date seems to have suffered the same way.

MDL

I take the opposite view. I truly believe that a grand slam win/runner up record of 8/11 is superior to 8/10 which is superior to 8/9 which is superior to 8/8 etc etc to 8/0. It has to be. Because otherwise you would be saying that losing in the first round of a slam event constitues a superior performance to making the final. If lendl, in those 11 events he was runner up, had instead lost in the first round, then he would be sitting on a 8/0 record, which is a 100% finals winning record. No, making finals is an achievement in itself. That is why Lendl fully deserves that 11 x 1.2 factor in his calculations

timnz
09-12-2012, 11:13 AM
Lendl: too high. His eleven losing finals count towards his legacy here more that I'd qualitatively allow. I think it's a lot to do with when he lost: if you have a pile-up of slam final losses before you start winning, people won't count them as positives. Murray's reputation to date seems to have suffered the same way.

MDL

I take the opposite view. I truly believe that a grand slam win/runner up record of 8/11 is superior to 8/10 which is superior to 8/9 which is superior to 8/8 etc etc to 8/0. It has to be. Because otherwise you would be saying that losing in the first round of a slam event constitues a superior performance to making the final. If lendl, in those 11 events he was runner up, had instead lost in the first round, then he would be sitting on a 8/0 record, which is a 100% finals winning record! No, making finals is an achievement in itself. That is why Lendl fully deserves that 11 x 1.2 factor in his calculations. (note: if one were to look for a reason for those 11 losses, then all you would have to do is look at the quality of the competition is faced. All but 1 of them was number 1 some time in their career)

JMR
09-12-2012, 11:20 AM
You have a point here. I am going to suggest that I instead weight them at 1.4 x instead of 1.5 x - reasons below:

Winners of the end of season championships tend to either go unbeaten (5 out of 6 of Federer's wins have been as an unbeaten player) or only have the loss of 1 match. I don't think there has ever been a player who has lost 2 matches and gone on to win the whole thing (if there has been it must be very much the exception).

Correct. No man has ever won the YEC with two losses. No player has ever reached the final with two losses. Only one player, Nalbandian, reached the semifinals with two losses, and he went no further.

timnz
09-12-2012, 09:37 PM
of the ranking system?

I deliberatly said only open era - because it is too hard trying to compare to the professionals only era or the amateur era. I also kept the current ATP points for each event to avoid controversy about me choosing an arbitary weighting.

What do people think of leaving off everything below masters 1000's?

oneness
09-12-2012, 09:54 PM
How many WTF titles would Rafa have if it was on clay. That is the reason why I don't agree to assigning it a higher value for the only reason being that the top 8 compete.
If we have the top 8 compete in clay, how many WTF tiles would Fed/Djoker and Murray win. Food for thought.?

timnz
09-12-2012, 10:23 PM
How many WTF titles would Rafa have if it was on clay. That is the reason why I don't agree to assigning it a higher value for the only reason being that the top 8 compete.
If we have the top 8 compete in clay, how many WTF tiles would Fed/Djoker and Murray win. Food for thought.?

Not sure what you mean by 'assigning it a higher value'. Can you elaborate on that please? I have rated the WTF at 1.4 x the Masters 1000. That is half-way between an unbeaten winner (1500 points) and a winner who only lost 1 match (1300 points) - Note: There never has been a player who lost 2 matches and went on to win the title. I didn't make this ranking points up - I used the current ATP ranking system.

re. Nadal and WTF on clay. That's rather arbitary to say. I could as easily say - how many French Open's would Federer have won if it was played on grass? Actually I think it makes perfect sense that it is indoor. If it wasn't there then tennis would be missing a major indoor championship. As it is now tennis has gone too far the other way in slowing down the surface of the majority of the tennis tournaments of the season. The WTF indoor is last bastion of some sort of balance. Aside from all that the WTF has always been there since the early 70's and has only 3 times in 42 years not been played indoors. Hence, the tradition is there. Why shouldn't it be on indoor?

Russeljones
09-13-2012, 12:03 AM
I believe there's a case for looking into a couple more. ATP titles won and weeks as #1 could also yield some sort of added coefficient I think.

Guillermo Vilas

4 x 2
4 x 1.2
1 x 1.4*
7 x 1

21,2

62 ATP titles.
--------------------------

Rod Laver

9 x 2
6 x 1.2
4 x 1.5 (23 consecutive matches won earn him a better coefficient imo)
in 1970 alone he won 5 Masters 1000 equivalent titles.
31,2 +++ (this is really one for the Laver historians to nitpick)


77(40 ATP) titles

timnz
09-13-2012, 03:20 AM
I believe there's a case for looking into a couple more. ATP titles won and weeks as #1 could also yield some sort of added coefficient I think.

Guillermo Vilas

4 x 2
4 x 1.2
1 x 1.4*
7 x 1

21,2

62 ATP titles.
--------------------------

Rod Laver

9 x 2
6 x 1.2
4 x 1.5 (23 consecutive matches won earn him a better coefficient imo)
in 1970 alone he won 5 Masters 1000 equivalent titles.
31,2 +++ (this is really one for the Laver historians to nitpick)


77(40 ATP) titles

If you use the same methodology to Laver's Open era career.

(Slams x 2) + (Season End Finals x 1.4) + (Losing Slam Finalist x 1.2) + (Masters 1000 equivalents x 1)

(5 x 2) + (0 x 1.4) + (1 x 1.2) + (14 x 1) = 25.2

That means that from the age of nearly 30 (French Open 1968 to nearly 36 he had almost as much career achievement as Wilander and Djokovic have to date!

These are the Masters 1000's equivalents I have for Laver (I don't think there is anything controversial here):

1968 Los Angeles (PSW Open)
1969 Philadelphia
1969 South African Open
1969 Boston (US Pro)
1969 Wembley
1970 Johannesburg
1970 Sydney
1970 Los Angeles (PSW Open)
1970 Wembley
1970 Philadelphia
1971 Rome
1972 Philadelphia
1974 Philadelphia
1974 Las Vegas

I didn't put Laver into the overall list because it wasn't fair to him since the majority of his career is pre-open era. Whereas all the other players on the list have had all their careers in the open era.

Re. Number 1 rankings addition into the formula - please refer to my comments earlier in the thread. Basically i dont think it is a good criteria, because of strange anomolies in it like roddick being number 1 longer than boris becker and also that it really is double counting because players got to these number 1 rankings largely because of their winning the big events i am counting.

Russeljones
09-13-2012, 03:24 AM
If you use the same methodology to Laver's Open era career.

(Slams x 2) + (Season End Finals x 1.4) + (Losing Slam Finalist x 1.2) + (Masters 1000 equivalents x 1)

(5 x 2) + (0 x 1.4) + (1 x 1.2) + (14 x 1) = 25.2

That means that from the age of nearly 30 (French Open 1968 to nearly 36 he had almost as much career achievement as Wilander and Djokovic have to date!

These are the Masters 1000's equivalents I have for Laver (I don't think there is anything controversial here):

1968 Los Angeles (PSW Open)
1969 Philadelphia
1969 South African Open
1969 Boston (US Pro)
1969 Wembley
1970 Johannesburg
1970 Sydney
1970 Los Angeles (PSW Open)
1970 Wembley
1970 Philadelphia
1971 Rome
1972 Philadelphia
1974 Philadelphia
1974 Las Vegas

I didn't put Laver into the overall list because it wasn't fair to him since the majority of his career is pre-open era. Whereas all the other players on the list have had all their careers in the open era.

I agree with everything in your methodology but why not give him 9 Slams?

timnz
09-13-2012, 05:20 AM
I agree with everything in your methodology but why not give him 9 Slams?

Because he only won 5 in the open era. These are open era rankings. Where did you get the 9 total from?

Russeljones
09-13-2012, 05:35 AM
Because he only won 5 in the open era. These are open era rankings. Where did you get the 9 total from?

My apologies, it seems the 63-67 ones made their way into my total. So what is your opinion on ATP tour/Grand Prix tournaments and how they can be weighed in? And what about weeks as #1?

timnz
09-13-2012, 11:54 AM
My apologies, it seems the 63-67 ones made their way into my total. So what is your opinion on ATP tour/Grand Prix tournaments and how they can be weighed in? And what about weeks as #1?

Re. Weeks at number 1 - problematic - due to it been a measure of the strength of other players at the time. Hence uneven playing field. For instance no one would think that Roddick was in the same class as Becker but yet he was number 1 longer. The other reason not to include it - it is a form of double counting. When you think about it, a player got to number 1 largely due to his performances in masters 1000 events , slams , masters cups etc. but we have already measured that. Also some players were number 1 when nobody thought they should be (mcenroe in 1982), or werent number 1 when many think they should be (vilas in 1977). Note these two examples were because the atp didnt count all of the events.

Being number 1 definitely is a sign that your play is very very high, but it is an indirect achievement not a direct achievement ie you get to number 1 it is because of winning the above mentioned events.

Re. Grand prix. I didnt include these because i wanted to avoid disputes about the depth of field of tournaments in the earlier times of the open eras. (have you noticed that older players seem to have many more tournaments than most of the modern players). That is why i said, well lets just measure things from masters 1000 up to try to get to a more level playing field. Then there is also the issue of how many points to assign - were Connors many many smaller tournaments rated at todays 250 points or 500 points? The only way you could do it perhaps is average them out and give the tournaments below masters 1000 a rating of 0.375. But what do you do about non atp sanctioned tournaments? Connors and Lendl could have another 40 or so tournaments added to their totals. Note that there were a number of these events with deep fields that the atp doesnt recognize that lendl, connors and mcenroe won eg EC championships. Therefore i thought it best not to go there.

JMR
09-13-2012, 12:15 PM
Re. Weeks at number 1 - problematic - due to it been a measure of the strength of other players at the time. Hence uneven playing field. For instance no one would think that Roddick was in the same class as Becker but yet he was number 1 longer.

I don't believe this objection is sound. The fact that one player is superior to another does not require that he be superior in every single measure of achievement. For example, no one would think that Chang was in the same class as Sampras, yet Chang won the FO while Sampras did not. The whole point of including multiple factors is to reward players for their accomplishments in the areas in which they happened to excel. (Throwing out a metric because you don't like the way it benefits different players) = bias.

The other reason not to include it - it is a form of double counting. When you think about it, a player got to number 1 largely due to his performances in masters 1000 events , slams , masters cups etc. but we have already measured that.

This objection has some merit, because it's true that rankings are partially a form of double-counting, since they reward performance in tournaments that are already being counted separately. However, I think you overlook two important distinctions:

1. Rankings also reward performance in smaller tournaments, which you are not counting separately. A player's ranking history is actually a handy way of taking into account his accomplishments in smaller events without going to the trouble of counting and weighing them.

2. Rankings reward concentrated performance in the bigger events, i.e., domination. Clustering big wins will earn a higher ranking than spreading them out. Domination is usually regarded as a tennis achievement.

timnz
09-13-2012, 12:26 PM
I don't believe this objection is sound. The fact that one player is superior to another does not require that he be superior in every single measure of achievement. For example, no one would think that Chang was in the same class as Sampras, yet Chang won the FO while Sampras did not. The whole point of including multiple factors is to reward players for their accomplishments in the areas in which they happened to excel. (Throwing out a metric because you don't like the way it benefits different players) = bias.



This objection has some merit, because it's true that rankings are partially a form of double-counting, since they reward performance in tournaments that are already being counted separately. However, I think you overlook two important distinctions:

1. Rankings also reward performance in smaller tournaments, which you are not counting separately. A player's ranking history is actually a handy way of taking into account his accomplishments in smaller events without going to the trouble of counting and weighing them.

2. Rankings reward concentrated performance in the bigger events, i.e., domination. Clustering big wins will earn a higher ranking than spreading them out. Domination is usually regarded as a tennis achievement.

You have some good points - though I can't get away from the feeling that it is double counting. What weighting would you give number 1 rankings? One of the deliberate things i have done is to use the atp's own ranking system of points in order to get away from arguments about it. Obviously the atp doesnt have a points ranking system for time as number 1. So what would be the most objective allocation? 1 or 2 point for every 52 weeks at number 1? The choice seems arbitary vs the rest of the formula that is based on objective criteria. (Please don't think I am dismissing your suggestion - it is just that I am not seeing some objective or at least universally agreed measure of allocating points for each 52 weeks at number 1)

Russeljones
09-14-2012, 12:40 AM
Re. Weeks at number 1 - problematic - due to it been a measure of the strength of other players at the time. Hence uneven playing field. For instance no one would think that Roddick was in the same class as Becker but yet he was number 1 longer. The other reason not to include it - it is a form of double counting. When you think about it, a player got to number 1 largely due to his performances in masters 1000 events , slams , masters cups etc. but we have already measured that. Also some players were number 1 when nobody thought they should be (mcenroe in 1982), or werent number 1 when many think they should be (vilas in 1977). Note these two examples were because the atp didnt count all of the events.

Being number 1 definitely is a sign that your play is very very high, but it is an indirect achievement not a direct achievement ie you get to number 1 it is because of winning the above mentioned events.

Re. Grand prix. I didnt include these because i wanted to avoid disputes about the depth of field of tournaments in the earlier times of the open eras. (have you noticed that older players seem to have many more tournaments than most of the modern players). That is why i said, well lets just measure things from masters 1000 up to try to get to a more level playing field. Then there is also the issue of how many points to assign - were Connors many many smaller tournaments rated at todays 250 points or 500 points? The only way you could do it perhaps is average them out and give the tournaments below masters 1000 a rating of 0.375. But what do you do about non atp sanctioned tournaments? Connors and Lendl could have another 40 or so tournaments added to their totals. Note that there were a number of these events with deep fields that the atp doesnt recognize that lendl, connors and mcenroe won eg EC championships. Therefore i thought it best not to go there.

I think we can't discriminate against the early Open Era players' weeks at #1 because clearly they could only beat what is put in front of them and, a bit like at the modern WTF, they would inevitably play the best players of their time for all major honours. We count their points won in GS wins and 2nd places but we discount their rankings, which these same GS wins and 2nd places contributed to? Seems like a bit of a paradox.

I like your idea, but I think we should have another look at the 'weighted index' idea and see if we can find a way to weigh the achievements of the early Open Era players. This thread is drifting dangerously close to the 'Past Players' territory but I find it fun and informative, so let's keep digging I say! :)