Ranking analysis in Big3 H2H

Never wrote that.
But that is the implication when you assume that you can tell whether Big 3 were in good or bad form when they lost to each other. I mean in most seasons those losses are a small percentage of all matches played. To believe that the players' form was corresponding to their ranking in those particular matches you need to think that it is the case for all matches.

But if you don't think that then how'd you draw conclusions from the stats you posted?
 
hmmm ... 25 years as an average peak??



hope you know that average years in the top10 is 30,54. and it is getting older.

7 of the top10 players are over 30 (70%). and just 1 is less than 25 (10%).

in top 25, 10players are over 30 (40%) and just 7 are under 25 (28%).

in top 50, 21 players are over 30 (42%) and only 11 are under 25 (22%).

in top 100, 34 players are over 30 and 29 are under 25.

if 25 was realy an average peak than are to supposed that abouth 50% of players in the top should be over, and abouth 50% under 25.
 
The geometric does not really make sense at all. The geometric mean should be used to determine the average if you have two different factors. F.e. if you have a camera of 500 pixels with a rating of 6/10 versus a camera with 400 pixels and a 7/10 rating. In this case, you have two factors that have different scales (pixels vs rating) and it makes sense to give both the pixels and the ratings a (somewhat) equal weight.

But using the geometric mean for just minimising rankings makes no sense IMO.

But AFAIK, Federer is the most consistent player ever and even in 2013 he was the number three seed at Wimbledon lol.

EDIT: the geometric mean is lulz for rankings: Suppose you have 10 players with 5x world no.1 and 5x world no.50. The arithmetic mean gives you a middle point of 25.5. The geometric mean gives you a 7.07 average.

Basically, number 50 is squashed to 50^(1/10) =1.479. Then this 1.479^5*1^5 = 7.07 (the geometric mean). The higher the ranking, the more it's squashed towards one. So it barely matters if your ranking is low or not in this average. Also, if your sample size is larger than 10 (I am sure that this is the case here), the number is squashed towards one even more lol. In general, it's not useful for rankings.

I am not sure if this is the way to go for rankings. I did not invent math and it's not my fault the geometric mean provides a meaningful average for the above purposes only.
 
Last edited:
I didn't claim it was. But it's the immediate aftermath, where Fed's form hadn't recovered fully and where he was pretty low on confidence after a year like that and vs. Rafa especially after 4 drubbings, which is why I mentioned it.

I could also mention that a mere 8 of their 38 meetings are post Wimbledon - and only 3 of those 8 are not at the WTF. 5 of them at the WTF (1 in 2013), 1 in Cinci (2013 again), 1 in Basel and 1 in Shanghai.

Given how Rafa and Roger generally does vs. the field post Wimbledon, I think it's safe to say, Rafa has benefitted from meeting Fed in Rafa's best period of the year more often than not.
Buah buah buah. Stop with the fanboy excuses. 2014 is not 2013, Federer was playing much better in 2014. Nadal was just too good that day, no need for an excuse.

Always that Federer loses to Nadal, Federer fans need to put an excuse.
 
Last edited:
Lew winds them up again. Lol. 8-B One thing I like about Lew though is that he doesn't insult posters who disagree with him or try to bully them when they don't see his point of view. That doesn't mean I don't think he somewhat has an agenda, lots of others do too on here, but he just ignores the insults and keeps on posting his stats. I have to admit it's a little comical how upset some of you get when it comes to this stuff.

Look, Federer could have retired in 2012 and went out with a bang at 31 like Sampras did. Federer had an amazing 2012 and won his 17th Slam. So he could easily have said I have had enough and it's time to move on to something else and enjoy the fruits of my labor. Instead, he chose to keep playing and when you keep playing you are going to have to play the top players to win big tournaments. There is no way around it. He added 3 more Slams and 6 Masters since then.

So he could have retired in 2012 with a 16-13 record against Djokovic, 17 Slams and 21 Masters, but his Slam record would all but be surpassed by now with Djokovic closing in and Djokodal would have much more Masters than him. But hey at least he would have a winning head to head against Djokovic (sarcasm). You can't have it all. Djokovic and Nadal can't either. Federer playing longer increased his legacy not hurt it and losing the head to heads comes along with that territory.
Excellent, rational, objective and well-redacted comment.
 
He would’ve had a very positive H2H vs Djokovic.

Nadal is irrelevant. Majority of their matches from 2005-2010 were on clay... where he peaked on the surface from 2005.
H2H: 16-13 (?)
Slams: 6-5
BO5: 7-5
Slam Finals: 1-0

But I think it's good he played after 2012 though. He has won the DC and the eighth Wimbledon crown.
 
I could also mention that a mere 8 of their 38 meetings are post Wimbledon - and only 3 of those 8 are not at the WTF. 5 of them at the WTF (1 in 2013), 1 in Cinci (2013 again), 1 in Basel and 1 in Shanghai.

Given how Rafa and Roger generally does vs. the field post Wimbledon, I think it's safe to say, Rafa has benefitted from meeting Fed in Rafa's best period of the year more often than not.
I don't understand how this hurts Federer. Federer is superior at AO, IW, Miami and Wimbledon with a combined 22 titles. On the other hand Nadal has a combined 6 titles yet Nadal is tied with him 7-7 in these tournaments. Federer was not able to gain any ground over Nadal in the head to head in these tournaments where he is clearly better and until 2017, Nadal led 7-4 in them. What is more important is how many times they met on clay in that part of the year which is 15/38 times or 39.5%. For comparison, Djokovic has met Nadal on clay 23/53 times or 43.4% so it's even more skewed.

I think where there is a real discrepancy is the number of times Fedal met in the US hardcourt season which is only a shocking 1 time, compared to Djokodal at 7 times, and that probably would have helped Federer but if we're being honest he underperformed against him in the first part of the year in tournaments where he is clearly better.
 

Lew II

Hall of Fame
He would’ve had a very positive H2H vs Djokovic.

Nadal is irrelevant. Majority of their matches from 2005-2010 were on clay... where he peaked on the surface from 2005.
Very positive h2h vs Djokovic?

7-4 with four wins when Djokovic was yet to reach the top10...
 
fed vs rafa h2h upp to 2012:

2012: 1-1 both on hard, 0-1 in slams

2011: 1-3, 1-1 on hard, 0-1 in slams

2010: 1-1, 1-0 on hard

2009: 1-1, 0-1 on hard, 0-1 in slams

2008: 0-4, 0-1 on grass, 0-2 in slams

2007: 3-2, 1-0 on hard, 1-0 on grass, 1-1 in slams

2006: 2-4, 1-1 on hard, 1-0 on grass, 1-1 in slams

2005: 1-1, 1-0 on hard, 0-1 in slams

2004: 0-1 on hard

so, 10-18:
7-6 on hard
2-1 on grass
1-11 on clay
2-8 in slams: 0-2 on hard, 2-1 on grass, 0-5 on clay
 
I don't understand how this hurts Federer. Federer is superior at AO, IW, Miami and Wimbledon with a combined 22 titles. On the other hand Nadal has a combined 6 titles yet Nadal is tied with him 7-7 in these tournaments. Federer was not able to gain any ground over Nadal in the head to head in these tournaments where he is clearly better and until 2017, Nadal led 7-4 in them. What is more important is how many times they met on clay in that part of the year which is 15/38 times or 39.5%. For comparison, Djokovic has met Nadal on clay 23/53 times or 43.4% so it's even more skewed.

I think where there is a real discrepancy is the number of times Fedal met in the US hardcourt season which is only a shocking 1 time, compared to Djokodal at 7 times, and that probably would have helped Federer but if we're being honest he underperformed against him in the first part of the year in tournaments where he is clearly better.
All those slow court events suit Nadal’s rum side to side retrieval / passing shots style vs Fed’s attacking game.
 
H2H: 16-13 (?)
Slams: 6-5
BO5: 7-5
Slam Finals: 1-0

But I think it's good he played after 2012 though. He has won the DC and the eighth Wimbledon crown.
I agree. 6 AO and 8 W thanks to the decision to play on. But I wouldn’t read anything into any of his losses to Nole post 2012 as he was a different player post racket change.
 
All those slow court events suit Nadal’s rum side to side retrieval / passing shots style vs Fed’s attacking game.
This is not supported by any facts. Federer is the far superior hardcourt player with multiple titles in AO, IW and Miami. Nadal has never even won Miami yet is tied in the head to head with Federer there. Federer is 5-6 in the head to head against Nadal in those three tournaments and by comparison Djokovic is 8-1. This is how the head to heads were won and lost with matchups with Nadal.
 
This is not supported by any facts. Federer is the far superior hardcourt player with multiple titles in AO, IW and Miami. Nadal has never even won Miami yet is tied in the head to head with Federer there. Federer is 5-6 in the head to head against Nadal in those three tournaments and by comparison Djokovic is 8-1. This is how the head to heads were won and lost with matchups with Nadal.
Yeah... because Nadal beats Fed by retrieving his “winners” and hitting topspin FH to his OHBH. Miami, AO, IW, Clay all play into Nadal’s hands.

Prime Fed could overcome this match up issue on faster courts thanks to his sublime ability. They didn’t meet enough post Wimbledon for it to be a fair h2h.
 
Yeah... because Nadal beats Fed by retrieving his “winners” and hitting topspin FH to his OHBH. Miami, AO, IW, Clay all play into Nadal’s hands.

Prime Fed could overcome this match up issue on faster courts thanks to his sublime ability. They didn’t meet enough post Wimbledon for it to be a fair h2h.
So why doesn't Nadal have more than 4 titles in these tournaments if he is superior in these types of conditions, and why does Federer have 14 titles? There is no facts that support that Federer is at disadvantage at these tournaments against Nadal. I don't know how you cannot see that he underperformed against him there even as far back as 2004.
 
.
So why doesn't Nadal have more than 4 titles in these tournaments if he is superior in these types of conditions, and why does Federer have 14 titles? There is no facts that support that Federer is at disadvantage at these tournaments against Nadal. I don't know how you cannot see that he underperformed against him there even as far back as 2004.
Because he ran into Federer, Djokovic etc and blew some matches to likes of Ljubicic and Roddick.

Nadal has the match up advantage vs Federer there in terms of h2h. Quite clear to anyone who’s ever watched a Fedal match.
 
.

Because he ran into Federer, Djokovic etc and blew some matches to likes of Ljubicic and Roddick.

Nadal has the match up advantage vs Federer there in terms of h2h. Quite clear to anyone who’s ever watched a Fedal match.
I've watched many Fedal matches and that is not clear to me. Even if we conceded that was the case that still doesn't explain how Federer was losing matches to him in places like Dubai, which is a fast hardcourt, back in 2006 in his best year on record.
 
I've watched many Fedal matches and that is not clear to me. Even if we conceded that was the case that still doesn't explain how Federer was losing matches to him in places like Dubai, which is a fast hardcourt, back in 2006 in his best year on record.
Clearly an implosion from Roger. If he brings his Canada, USO, YEC or Madrid indoors form he likely takes that in 2.

Why couldn’t world number 2 reach Fed at USO in 2005-2009? Or Cincy in 2005,2007,2009, Madrid indoors 2006/2007?
 
Fedfans: Screw your objective stats, here are my more important subjective views.
Stats are either accurate or inaccurate, they have nothing to do with objectivity. People and their judgment can be objective but it is rather laughable to use this in support of the OP's thread(s) since his bias can be seen from space.
 
Clearly an implosion from Roger. If he brings his Canada, USO, YEC or Madrid indoors form he likely takes that in 2.

Why couldn’t world number 2 reach Fed at USO in 2005-2009? Or Cincy in 2005,2007,2009, Madrid indoors 2006/2007?
Look, I agree that they should have met more in the USO hardcourt series and Federer would most likely have benefited but I also can see that he should have done better than a 5-6 record against Nadal in the AO, IW and Miami tournaments. 2017 helped him a lot because the record there was worse before then, and he won 3 matches in a row against him there.
 
Look, I agree that they should have met more in the USO hardcourt series and Federer would most likely have benefited but I also can see that he should have done better than a 5-6 record against Nadal in the AO, IW and Miami tournaments. 2017 helped him a lot because the record there was worse before then, and he won 3 matches in a row against him there.
Yeah true. Should have won a couple more clay matches too. Rome 2006, Hamburg 2008.
 

Lew II

Hall of Fame
The geometric does not really make sense at all. The geometric mean should be used to determine the average if you have two different factors. F.e. if you have a camera of 500 pixels with a rating of 6/10 versus a camera with 400 pixels and a 7/10 rating. In this case, you have two factors that have different scales (pixels vs rating) and it makes sense to give both the pixels and the ratings a (somewhat) equal weight.

But using the geometric mean for just minimising rankings makes no sense IMO.

But AFAIK, Federer is the most consistent player ever and even in 2013 he was the number three seed at Wimbledon lol.

EDIT: the geometric mean is lulz for rankings: Suppose you have 10 players with 5x world no.1 and 5x world no.50. The arithmetic mean gives you a middle point of 25.5. The geometric mean gives you a 7.07 average.

Basically, number 50 is squashed to 50^(1/10) =1.479. Then this 1.479^5*1^5 = 7.07 (the geometric mean). The higher the ranking, the more it's squashed towards one. So it barely matters if your ranking is low or not in this average. Also, if your sample size is larger than 10 (I am sure that this is the case here), the number is squashed towards one even more lol. In general, it's not useful for rankings.

I am not sure if this is the way to go for rankings. I did not invent math and it's not my fault the geometric mean provides a meaningful average for the above purposes only.
I used geometric mean because low rankings would skew the stat. Ultimate Tennis Statisttcs uses geometric mean too.

Arithmetic average would make look Djokovic look even better, because Federer and Nadal beat Djokovic twice when he was no.67/63, plus 5 other times when he was outside top10.
 
Last edited:
Lowest ranking they ever beat each other:

Federer b. Djokovic no.67
Federer b. Nadal no.31

Nadal b. Djokovic no.63
Nadal b. Federer no.7

Djokovic b. Federer no.8
Djokovic b. Nadal no.8
Doesn't matter. Djokovic was taking sets off Federer in 2006 and 2007 in their first meetings. It's not like he was some random scrub.
 
Wins over Big3 ranked outside the top5:

Federer 9
Nadal 6
Djokovic 5

Wins over Big3 ranked outside the top10:

Federer 5
Nadal 3
Djokovic 0
Yep. And had Nadal taken out the much older Fed in 2017 at the AO, IW, and Miami, then Nadal would have passed Fed in wins outside the top 10 and matched him in wins outside the top 5. But since the much older Fed won, he looks worse in this useless cherry-picked stat.

Djokovic skates by on this stat too by losing to Kyrgios and Istomin so that he didn’t have to face the low-ranked Fed in early 2017. Djoker getting by Krygious and Istomin and then later beating Fed and his lower ranking would have worsened Djoker’s rating in this idiotic stat.

And lastly, the guy that is the oldest that also holds the record for the most weeks at number one is obviously going to have the best average ranking. This is so obvious.

But while we are at it, here are some stats that look Fed look bad. I will reference his peak year of 2017:

# of times Fed beat Donskoy while Evgeny was ranked in the top 10: 0
# of times Donskoy best Fed while he was ranked in the top 10: 1
# of times Donskoy beat Fed while he held a slam title: 1
# of career wins over Donskoy while he was ranked outside the 100 for his entire career: 0

What’s funny is that if Fed had beaten Donskoy, then Fed plummets here because now, Fed beat Donskoy when he was ranked outside the top 100. So now, Fed gets accused of beating yet another player while he’s down.

Thank god Fed lost that match. Also, I am glad that Fed lost to #302 ranked Tommy Haas. Imagine what Haas’ average ranking would have been during matches in which Fed beat him, had Fed won that match. Thankfully, Fed lost to make Lew’s bogus cherry picked stat look better.
 

ABCD

Hall of Fame
Yep. And had Nadal taken out the much older Fed in 2017 at the AO, IW, and Miami, then Nadal would have passed Fed in wins outside the top 10 and matched him in wins outside the top 5. But since the much older Fed won, he looks worse in this useless cherry-picked stat.

Djokovic skates by on this stat too by losing to Kyrgios and Istomin so that he didn’t have to face the low-ranked Fed in early 2017. Djoker getting by Krygious and Istomin and then later beating Fed and his lower ranking would have worsened Djoker’s rating in this idiotic stat.

And lastly, the guy that is the oldest that also holds the record for the most weeks at number one is obviously going to have the best average ranking. This is so obvious.

But while we are at it, here are some stats that look Fed look bad. I will reference his peak year of 2017:

# of times Fed beat Donskoy while Evgeny was ranked in the top 10: 0
# of times Donskoy best Fed while he was ranked in the top 10: 1
# of times Donskoy beat Fed while he held a slam title: 1
# of career wins over Donskoy while he was ranked outside the 100 for his entire career: 0

What’s funny is that if Fed had beaten Donskoy, then Fed plummets here because now, Fed beat Donskoy when he was ranked outside the top 100. So now, Fed gets accused of beating yet another player while he’s down.

Thank god Fed lost that match. Also, I am glad that Fed lost to #302 ranked Tommy Haas. Imagine what Haas’ average ranking would have been during matches in which Fed beat him, had Fed won that match. Thankfully, Fed lost to make Lew’s bogus cherry picked stat look better.
Usual response to hard/cold Lew II numbers is 1) Cherry picking, 2) If X had happened then Y would be the case (woulda, shoulda, coulda) , 3) 20>17>15, 4) Eye test, etc. Djokovic has better H2H numbers whatever cherries you pick.
 

Lew II

Hall of Fame
hmmm ... 25 years as an average peak??



hope you know that average years in the top10 is 30,54. and it is getting older.

7 of the top10 players are over 30 (70%). and just 1 is less than 25 (10%).

in top 25, 10players are over 30 (40%) and just 7 are under 25 (28%).

in top 50, 21 players are over 30 (42%) and only 11 are under 25 (22%).

in top 100, 34 players are over 30 and 29 are under 25.

if 25 was realy an average peak than are to supposed that abouth 50% of players in the top should be over, and abouth 50% under 25.
Good post.

Many say that only Big3 played good in their late 20s or 30s, but this shows it is actually false..
 
Lowest ranking they ever beat each other:

Federer b. Djokovic no.67
Federer b. Nadal no.31

Nadal b. Djokovic no.63
Nadal b. Federer no.7

Djokovic b. Federer no.8
Djokovic b. Nadal no.8
Doesn't matter. Djokovic was taking sets off Federer in 2006 and 2007 in their first meetings. It's not like he was some random scrub.
Yeah and Fed was still top 10 in 2013 when he was taking 4 losses from Nadal, 2 from Nole... top 3 in 2018 when he lost twice to in form Nole (Paris he actually fought well though)
 
Top