Federer, then and now...

mike danny

Bionic Poster
I don't compare to Nadal nor Djokovic, I try to see it from outside, against "the field". Federer won't have more problems with Novak or Nadal, it's the "rest" that challanges him more. Who has Federer lost to in 2014?
Hewitt
Nadal
Nishikori
Djokovic (x2)
Wawrinka
Chardy
Gulbis
Tsonga
Cilic

Out of those loses, there was a) one match where he had absolutely no chance (Cilic US Open)
b) two further matches that didn't go the full distance (Nadal @ AO, Tsonga @ Toronto)

All other matches were extremely tight and could've easily been his if he had upped or held his game just a tiny little bit. A few single points can change a lot. Likewise Federer won some matches veeery clutch (or lucky, name it how you want).
If he cannot sustain this level, he'll not only lose to Nadal or Djokovic but to many more top50 players.

Against someone like Cilic, I'm 99% sure, Federer in 2005 would not have lost in straights, no way. There were many players trying to "bomb" away Federer, some of them prevailed in tight games but not without putting up a fight.
Those kind of matches will happen more often as he gets older if he runs into someone on fire. Tsonga was also a little bit like that @ Toronto.

I think 2013 gave us a good example how it looks when a player is ageing and declining. It's not like he sucks completely but there's just an edge missing.
2014 showed that it had nothing to do with age but healthiness. Which is also the reason I believe most top players retire from tennis.
Federer in 2005 might even have beaten Cilic. He probably wouldn't have let Monfils take him to 5 in the first place and preserved his energy for Cilic.

I just don't see Cilic beating 2005 Fed in a major
 

NatF

Bionic Poster
Cilic doesn't beat any Federer of 04-08 at the USO IMO. I have doubts he beats even 09-11 Federer too. Even in 2003 I think Federer would of taken a set.
 

falstaff78

Hall of Fame
I haven't done that, and I would except for the fact that I'm not done with the stats I am working on now, and it takes a LOT of time.

But if you want to share more stats, I would love to see them.

The numbers I used indirectly indicate scores. Obviously 200 for a match would be a double bagel, and you can't get more dominant in tennis that that.

I don't know how TBs are scored as wins/losses, but if a TB is simply scored as a break, the score there would be 114 because 6 games are held, only one is broken. A set with three TBs are going to average out very low, at something a wee bit under 105. Isner is at 93/10 or 103 for the year.

So far the magic number seems to be around 120. This year Novak is only 115 on grass, quite low for this era, Fed 117 on grass. Dimitrov in this weak year was 116. Murray was 119 in 2013, good but not terribly strong.

Fed was 116 in 2012, even though he won Wimbledon.

Compare that to Fed 2006, 124 and an amazing 130 in 2004. That's God-mode on grass. Novak was 131 on HC in 2011 and 132 on Grass. No one else came close. But on clay in 2011 Novak, 125 ran into Nadal in 126. That's why he did not get the grand slam that year. ;)

I'd go to this:

Fed 2014 on grass: 116
Fed 2005 on grass: 119

3 points may not seem like much, but it's a definite factor.

Fed 2014 on HC: 117
Fed 2005 on HC: 121

In order to win another slam, he has to get very lucky, or he has to get zoned in one slam as Pete did in his last USO. Pete was at a horrible 107 that year, but he pulled it out in just one tournament to beat Agassi, who was 118. That had to have been Agassi's most bitter loss.

Notice that that 4% difference is very close to the score I gave. It may be coincidence, probably is, but breaks in serve have to be directly related to percentage of break points won. That score can be lower IF service games are way stronger.

Novak has a 45% breakpoint average this year. In any year that has to be higher than Fed's.So 45% is not good enough for him. In 2011 it was 48%.

Fed was at 47% of breakpoints on grass in 2004, 49% in 2006. That tells the real story, doesn't it?

Rafa had a stunning 52% of break points on clay in 2008. That's they year he hit 84/51 135 I'm using.

Hey thanks a lot for the posts.

First of all, if you have been trawling the ATP site for stats you will LOVE tennisabstract. It is run by a guy who collects all the info from the ATP site and allows you to run crazy intuitive searches...and also calculates dominance ratios!!!!

e.g. the URLs below link to all the analyses I shared in my previous post:

2005:
overall
vs. top 10
in semis
in finals

2014:
overall
vs. top 10
in semis
in finals

Just to clarify - I really like this summation metric. i.e. sum of pct of games won on serve and pct of return games won. I think it captures brilliantly and concisely the highs and lows of players' games. Furthermore, what this metric is for games played, dominance ratio is for points played. They complement each other really well.

My issue is with the metric % of total points won. Because total % of points won depends not only on % of points won on serve and % of points won on return, but also depends on % of total points played on serve! This is not an issue when analyzing ratios of games played, because roughly 50% of games are always played on each players' serve.

F
 

falstaff78

Hall of Fame
Falstaff, could you make a thread comparing 2006 and 2014 or something like that using dominance ratio at the end of the year? Could be great.

Hey Chanwan here you go. (2014 now includes Basel final)

overall (2006 / 2014)
Win loss: 92-5 / 66-10
Dominance ratio: 1.40 / 1.39
Tie breaks: 37-14 / 24-13



Vs top 10 (2006 / 2014)
Win loss: 19-4 / 13-4
Dominance ratio: 1.24 / 1.19
Tie breaks: 14-6 / 2-5

Vs 11+ (2006 / 2014)
Win loss: 73-1 / 53-6
Dominance ratio: 1.47 / 1.47
Tie breaks: 23-8 / 22-8




prior to semi-finals (2006 / 2014)
Win loss: 59-1 / 46-3
Dominance ratio: 1.45 / 1.52
Tie breaks: 25-7 / 14-7

semi-finals (2006 / 2014)
Win loss: 16-0 / 10-2
Dominance ratio: 1.42 / 1.19
Tie breaks: 2-1 / 2-2

finals (2006 / 2014)
Win loss: 12-4 / 5-5
Dominance ratio: 1.27 / 1.06
Tie breaks: 8-6 / 5-4

2006 breakup by round excludes 2 davis cup matches and 3RR matches for master's cup. 2014 breakup by round excludes 5 davis cup matches.
 

NatF

Bionic Poster
Hey thanks a lot for the posts.

First of all, if you have been trawling the ATP site for stats you will LOVE tennisabstract. It is run by a guy who collects all the info from the ATP site and allows you to run crazy intuitive searches...and also calculates dominance ratios!!!!

e.g. the URLs below link to all the analyses I shared in my previous post:

2005:
overall
vs. top 10
in semis
in finals

2014:
overall
vs. top 10
in semis
in finals

Just to clarify - I really like this summation metric. i.e. sum of pct of games won on serve and pct of return games won. I think it captures brilliantly and concisely the highs and lows of players' games. Furthermore, what this metric is for games played, dominance ratio is for points played. They complement each other really well.

My issue is with the metric % of total points won. Because total % of points won depends not only on % of points won on serve and % of points won on return, but also depends on % of total points played on serve! This is not an issue when analyzing ratios of games played, because roughly 50% of games are always played on each players' serve.

F

Does tennis abstract provide examples of simple winners and UE's? I know it gives the total points and other stats.
 

Steve0904

Talk Tennis Guru
Hey Chanwan here you go. (2014 now includes Basel final)

overall (2006 / 2014)
Win loss: 92-5 / 66-10
Dominance ratio: 1.40 / 1.39
Tie breaks: 37-14 / 24-13



Vs top 10 (2006 / 2014)
Win loss: 19-4 / 13-4
Dominance ratio: 1.24 / 1.19
Tie breaks: 14-6 / 2-5

Vs 11+ (2006 / 2014)
Win loss: 73-1 / 53-6
Dominance ratio: 1.47 / 1.47
Tie breaks: 23-8 / 22-8




prior to semi-finals (2006 / 2014)
Win loss: 59-1 / 46-3
Dominance ratio: 1.45 / 1.52
Tie breaks: 25-7 / 14-7

semi-finals (2006 / 2014)
Win loss: 16-0 / 10-2
Dominance ratio: 1.42 / 1.19
Tie breaks: 2-1 / 2-2

finals (2006 / 2014)
Win loss: 12-4 / 5-5
Dominance ratio: 1.27 / 1.06
Tie breaks: 8-6 / 5-4

2006 breakup by round excludes 2 davis cup matches and 3RR matches for master's cup. 2014 breakup by round excludes 5 davis cup matches.

Interesting. Biggest thing there is TB's IMO.

Massive difference overall and against top 10 players, and a slightly better positive record in the TB's in finals in 2006. Those things are momentum swingers in big matches.
 

dpli2010

Semi-Pro
Here are stats from the ATP site. I'm comparing Fed in 2004 and 2005.

Service Record Year-to-Date: 2005 and 2014

599 Aces 2005: .511 per game
552 Aces 2014: .628 per game

152 Double Faults 2005: .139 per game
101 Double Faults 2014: .114 per game

63% 1st Serve 2005
65% 1st Serve 2014

76% 1st Serve Points Won 2005
78% 1st Serve Points Won 2014

59% 2nd Serve Points Won 2005
57% 2nd Serve Points Won 2014

327 Break Points Faced 2005
287 Break Points Faced 2014

64% Break Points Saved 2005
70% Break Points Saved 2014

1,093 Service Games Played 2005:
879 Service Games Played 2014:

89% Service Games Won 2005
90% Service Games Won 2014

70% Service Points Won 2005
71% Service Points Won 2014

Return Record Year-to-Date: 2005 and 2014

35% 1st Serve Return Points Won 2005
33% 1st Serve Return Points Won 2014

52% 2nd Serve Return Points Won 2005
50% 2nd Serve Return Points Won 2014

733 Break Points Opportunities 2005
576 Break Points Opportunities 2014

44% Break Points Converted 2005
40% Break Points Converted 2014

1,054 Return Games Played 2005
865 Return Games Played 2014

31% Return Games Won 2005
26% Return Games Won 2014

42% Return Points Won 2005
40% Return Points Won 2014

55% Total Points Won 2005
55% Total Points Won 2014

GD, the tt analyst...
 

Gary Duane

Talk Tennis Guru
Hey thanks a lot for the posts.

First of all, if you have been trawling the ATP site for stats you will LOVE tennisabstract. It is run by a guy who collects all the info from the ATP site and allows you to run crazy intuitive searches...and also calculates dominance ratios!!!!
I'm a bit afraid to go there. I may turn into Rain Man. ;)
Just to clarify - I really like this summation metric. i.e. sum of pct of games won on serve and pct of return games won. I think it captures brilliantly and concisely the highs and lows of players' games. Furthermore, what this metric is for games played, dominance ratio is for points played. They complement each other really well.
[/quote]
It's just a fast way to get to a figure that is not there: % of games won. Since I have been adding the figures in my head, not having to divide by 2 just makes it easier. But if you have the stats and can play with them, it's simply (%A + %B)/2 = %C, where C is % of all games played.
My issue is with the metric % of total points won. Because total % of points won depends not only on % of points won on serve and % of points won on return, but also depends on % of total points played on serve! This is not an issue when analyzing ratios of games played, because roughly 50% of games are always played on each players' serve.
F
I think over time % of points played will even out, but it is a bad metric for just a year. The guys who regularly serve and hold at love or 15 will have an edge on points on serve. Every game is a minimum of 4 points, so 15, 30, 40 game can be 4/0 in points. The next best is 40/15 then game, 4/1. Any other service game will be won by two points.

The same thing is true in reverse when returning, obviously. So % of total points will reflect the game winning edge, but it can slip a point or so either way, over one year.

Novak is 85/32 for his career, so 117. Since these figures are rounded off, that could be 85.49 and 32.49, which would mean that the number should really be 118. Depending on how they round the figures.

So the total can be off a point. Sticking with 117, that's 58.5%?

And he has 54% of points won, career. If the second figure is relevant, there should be a close relationship to the first figure over many years.

Fed is 115 and 54%, career. So very close to Novak's 115 and 54%. Again since 54 is rounded, there may be an even closer correlation.

Nadal is 119 and 54%. Checking by surfaces will show career dominance on any surface.

To me the numbers that jump out at me are the range from around 115 to around 120. Figure 114 is 57%, and 120 is 60%. So that seems to be the magic number over years. To me this suggests that in the end it is mostly about winning percentage of games, no matter how they are won, so a higher percentage overall means a greater chance of pulling out close matches.

Agassi is 116 and 53
Sampras is 113 and 54

Sampras gets dragged down because his awful record on clay, and it really was awful.

Hewitt is 110 and 52: Hewitt appears weak because of all the years he was injured.

Roddick is 110 and 53: Roddick did not stay at the top enough years to score as high.

Murray is 114 and 53: To me this indicates that he has gotten totally screwed by being up against the Big Three. Compare with:

Kuerten, 106 and 52: So how in hell was he #1 in 2000?

I'm with anyone who talks about the late 90s and very early 2000s as being a very weak period.
 

SublimeTennis

Professional
This just proves that Fed hasn't really declined at all. It's just that the competition is much tougher now that prevents him from winning slam. Thank you for solidifying Fed's extremely weak era.

Just the opposite; If an old Federer is winning the same percentage of points, it means THIS ERA is weaker!
 

Chanwan

G.O.A.T.
Hey Chanwan here you go. (2014 now includes Basel final)

overall (2006 / 2014)
Win loss: 92-5 / 66-10
Dominance ratio: 1.40 / 1.39
Tie breaks: 37-14 / 24-13



Vs top 10 (2006 / 2014)
Win loss: 19-4 / 13-4
Dominance ratio: 1.24 / 1.19
Tie breaks: 14-6 / 2-5

Vs 11+ (2006 / 2014)
Win loss: 73-1 / 53-6
Dominance ratio: 1.47 / 1.47
Tie breaks: 23-8 / 22-8




prior to semi-finals (2006 / 2014)
Win loss: 59-1 / 46-3
Dominance ratio: 1.45 / 1.52
Tie breaks: 25-7 / 14-7

semi-finals (2006 / 2014)
Win loss: 16-0 / 10-2
Dominance ratio: 1.42 / 1.19
Tie breaks: 2-1 / 2-2

finals (2006 / 2014)
Win loss: 12-4 / 5-5
Dominance ratio: 1.27 / 1.06
Tie breaks: 8-6 / 5-4

2006 breakup by round excludes 2 davis cup matches and 3RR matches for master's cup. 2014 breakup by round excludes 5 davis cup matches.

what a service! thanks
Does this bit
overall (2006 / 2014)
Win loss: 92-5 / 66-10
Dominance ratio: 1.40 / 1.39
mean that he has beaten the players he's beaten more comprehensively this year given it's more or less the same despite a vastly worse W/L-record?
 

kOaMaster

Hall of Fame
Again, an interpretation of mine, reading these stats:
Federer did decline quite a bit - players reaching the semis & finals of a tournament are now strong enough to beat him. this is an obvious and huge difference compared to 2006.
This is all correlating with the rest of the data, corresponds also with the interpretation of natural "ageing".

Why didn't Federer have a higher dominance ratio in 2006? Easy: It wasn't necessary. What is the point of wasting energy and mental focussing? If any player was excellent to pace himself it was Federer. Watching his matches there was very often this moment when he just switched in an higher gear, went for the break and when he eventually got it, stayed back more or less relaxed, serving the match to win.
There is no point in trying to win every single set 6-0, this will burn you out. It's mentally extremely demanding to keep up the concentration that high. I guess everyone of us who plays tennis knows it from own experience: Once you're up against a lesser talented opponent you tend to lean back a little.

Two weeks ago in the Swiss TV, Federer was interviewed about 20mins. He said that one adaption he had to make the last 5-10 years was to play a lot more "percentage tennis". This may sound ridiculous since my image of percentage tennis is more something like Ferrer, not Rogers tennis. Imo, this change allows him to still beat the vast majority reliably.

(It's also every coaches dream - my coach often is close to killing me for my rather low-percentage style ;) )
 
Last edited:

Chanwan

G.O.A.T.
Again, an interpretation of mine, reading these stats:
Federer did decline quite a bit - players reaching the semis & finals of a tournament are now strong enough to beat him. this is an obvious and huge difference compared to 2006.
This is all correlating with the rest of the data, corresponds also with the interpretation of natural "ageing".

Why didn't Federer have a higher dominance ratio in 2006? Easy: It wasn't necessary. What is the point of wasting energy and mental focussing? If any player was excellent to pace himself it was Federer. Watching his matches there was very often this moment when he just switched in an higher gear, went for the break and when he eventually got it, stayed back more or less relaxed, serving the match to win.
There is no point in trying to win every single set 6-0, this will burn you out. It's mentally extremely demanding to keep up the concentration that high. I guess everyone of us who plays tennis knows it from own experience: Once you're up against a lesser talented opponent you tend to lean back a little.

Two weeks ago in the Swiss TV, Federer was interviewed about 20mins. He sayed that one adaption he had to make the last 5-10 years was to play a lot more "percentage tennis". This may sound ridiculous since my image of percentage tennis is more something like Ferrer, not Rogers tennis. Imo, this change allows him to still beat the vast majority reliably.

(It's also every coaches dream - my coach often is close to killing me for my rather low-percentage style ;) )

interesting, do you have a link to the interview? And I think it's 100 % true. Fed is no longer quite the highlight reel, he used to be and part of that is him going for the safer shots more often.
You're also right in terms of why win 6-0, when you can win 6-3 without trying too hard
 

kOaMaster

Hall of Fame
sayed...iieks, how on earth did that mistake happen?

link to the interview:
http://www.srf.ch/sendungen/sportpanorama/motorrad-eishockey-fussball-und-studiogast-roger-federer
starts @ about 38:30. It's in Swiss German, I don't think I'll do a translation (did that once already in 2009, was a lot of work).
Can't tell the exact moment or words, would have to watch it again.

They are also showing some of his "best" points this year, horrible selection.

Also interesting at the beginning when asked about him not being in Swiss TV for suchs interview more often (last visit in the show was 2010): He tries to cut back the presence in Switzerland, as he naturally is already a main topic due to his results and doesn't want to "flood" Switzerland with his media omnipresence. Also he said it's still the place where he can come at rest better than anywhere else, people are holding off mostly. He'd be afraid this could change if he would be more in TV etc.
 

falstaff78

Hall of Fame
what a service! thanks
Does this bit
overall (2006 / 2014)
Win loss: 92-5 / 66-10
Dominance ratio: 1.40 / 1.39
mean that he has beaten the players he's beaten more comprehensively this year given it's more or less the same despite a vastly worse W/L-record?

That's a great insight and borne out by the data.

Dominance Ratio in wins:
2006: 1.45
2014: 1.56

Dominance Ratio in losses::
2006: 0.95 (range 0.83 - 1.25)
2014: 0.86 (range 0.58 - 1.02)

These numbers are telling us is that in 2006 when Federer lost it was either close (eg Rome) or he was just plain, stupid unlucky (eg Dubai where he significantly outperformed his opponent with a DR of 1.25). Unlike 2014 it simply wasn't possible for anyone in any environment to blow him away.

They are also telling us that he had a (slightly) lower margin for error when winning. The loss numbers tell us this wasn't because he is better in 2014. (As does the evidence of our eyes, his performances in finals and therefore his overall achievements for the year and age-related data for every professional in the open era. This must mean he was either holding back when winning in 2006 and/or over time has figured out how to beat easier opponents more efficiently.

Another curious observation from the 2006 vs 2014 comparison is that vs opponents outside the top 10 Fed had the same DR in 2006 and 2014 (1.47) but a much better winning record in 2006 (73-1 vs 53-6). The hypothesis that he was holding back in wins partially explains this. Indeed in those 73 wins in 2006 his DR was 1.49 whereas in those 53 wins in 2014 his DR is 1.62

Many thanks to kOaMaster for his/her extremely valuable posts with great insights from the interview of the maestro about the points we are discussing in this post, as well as for the comments about Feds relation with Switzerland. Great insights!
 
Last edited:

Gary Duane

Talk Tennis Guru
You're also right in terms of why win 6-0, when you can win 6-3 without trying too hard
I have an answer to that. 6/3 when you are serving first is only one break. So that's one break away from Isner-like tennis, where so often everything comes down to TBs. I think depending on winning TBs is very risky.

Think about it. Normally 6/3 means you are already at 3/3, again if you are serving first, you get to 4/3 and the set is still even. So you have to be very sure you will continue winning your serve and that you can get that break. You are running out of time.

But if you can get to 2/0 quickly, or 4/2, you have a break already. You can still cruise, but you also have time to get another break, and that gives you a huge amount of confidence.

Also, taking an early lead will make a lot of players try less. Not all of them, and not the great ones, but a lot of players will start to think about turning things around in the 2nd set.

To me percentage tennis is about hitting the smart shot AND about applying the most effort at time when odds favor you. That means that as a receiver you're probably going to put a lot more effort into a 30/30 game than a 40/love game where you are going to have to win at least 6 points for the break.

In my opinion Fed has to pace himself better now. So by upping his % of service games won - which is always connected to the serve itself - he gives himself more margin for losing more return games. And of everything we can consider that would be the most dangerous against the very best players.

Today's Fed can't go for broke as often on returns, in my opinion.
 

Chanwan

G.O.A.T.
sayed...iieks, how on earth did that mistake happen?

link to the interview:
http://www.srf.ch/sendungen/sportpanorama/motorrad-eishockey-fussball-und-studiogast-roger-federer
starts @ about 38:30. It's in Swiss German, I don't think I'll do a translation (did that once already in 2009, was a lot of work).
Can't tell the exact moment or words, would have to watch it again.

They are also showing some of his "best" points this year, horrible selection.

Also interesting at the beginning when asked about him not being in Swiss TV for suchs interview more often (last visit in the show was 2010): He tries to cut back the presence in Switzerland, as he naturally is already a main topic due to his results and doesn't want to "flood" Switzerland with his media omnipresence. Also he said it's still the place where he can come at rest better than anywhere else, people are holding off mostly. He'd be afraid this could change if he would be more in TV etc.
Cheers, I can get a bit of Swiss German, but boy that's a difficult language!

That's a great insight and borne out by the data.

Dominance Ratio in wins:
2006: 1.45
2014: 1.56

Dominance Ratio in losses::
2006: 0.95 (range 0.83 - 1.25)
2014: 0.86 (range 0.58 - 1.02)

These numbers are telling us is that in 2006 when Federer lost it was either close (eg Rome) or he was just plain, stupid unlucky (eg Dubai where he significantly outperformed his opponent with a DR of 1.25). Unlike 2014 it simply wasn't possible for anyone in any environment to blow him away.

They are also telling us that he had a (slightly) lower margin for error when winning. The loss numbers tell us this wasn't because he is better in 2014. (As does the evidence of our eyes, his performances in finals and therefore his overall achievements for the year and age-related data for every professional in the open era. This must mean he was either holding back when winning in 2006 and/or over time has figured out how to beat easier opponents more efficiently.

Another curious observation from the 2006 vs 2014 comparison is that vs opponents outside the top 10 Fed had the same DR in 2006 and 2014 (1.47) but a much better winning record in 2006 (73-1 vs 53-6). The hypothesis that he was holding back in wins partially explains this. Indeed in those 73 wins in 2006 his DR was 1.49 whereas in those 53 wins in 2014 his DR is 1.62

Many thanks to kOaMaster for his/her extremely valuable posts with great insights from the interview of the maestro about the points we are discussing in this post, as well as for the comments about Feds relation with Switzerland. Great insights!
I really should take a look at tennisabstract and figure out how it works. Out of curiosity, how long does it take for you to find numbers like the above? 2 minutes? 15? Interesting with the holding back thesis.

@Gary,
I'm sorry but I'm a bit unsure of what to take away from your post? That Fed is pacing himself better in return games now because he has to physically and his ever so slightly better hold percentage (92 % in 2004 and 90 % in 2006 btw) now compared to 2005 is helping him with that?
Also, his TB-record back then was better than now. Plus he had an extra gear, he could switch on - if he needed a late break or if the opponent suddenly won a set. So I don't think not getting more than one break per set was risky as is also evidenced by his 73-1 record vs. players outside the top-10.

btw - no trolls!!!! (no jinx)
 

Gary Duane

Talk Tennis Guru
That's a great insight and borne out by the data.

Dominance Ratio in wins:
2006: 1.45
2014: 1.56

Dominance Ratio in losses::
2006: 0.95 (range 0.83 - 1.25)
2014: 0.86 (range 0.58 - 1.02)

These numbers are telling us is that in 2006 when Federer lost it was either close (eg Rome) or he was just plain, stupid unlucky (eg Dubai where he significantly outperformed his opponent with a DR of 1.25). Unlike 2014 it simply wasn't possible for anyone in any environment to blow him away.

They are also telling us that he had a (slightly) lower margin for error when winning. The loss numbers tell us this wasn't because he is better in 2014. (As does the evidence of our eyes, his performances in finals and therefore his overall achievements for the year and age-related data for every professional in the open era. This must mean he was either holding back when winning in 2006 and/or over time has figured out how to beat easier opponents more efficiently.

Another curious observation from the 2006 vs 2014 comparison is that vs opponents outside the top 10 Fed had the same DR in 2006 and 2014 (1.47) but a much better winning record in 2006 (73-1 vs 53-6). The hypothesis that he was holding back in wins partially explains this. Indeed in those 73 wins in 2006 his DR was 1.49 whereas in those 53 wins in 2014 his DR is 1.62

Many thanks to kOaMaster for his/her extremely valuable posts with great insights from the interview of the maestro about the points we are discussing in this post, as well as for the comments about Feds relation with Switzerland. Great insights!
These are great insights.

Here is another slant on comparing those years:

Fed 2006: Grass 94/30 124 (or 62% of all games won) WIMBLY
Fed 2014: Grass: 95/22 117
Novak 2014: Grass 92/23 115

This suggests Novak and Fed were very close this year, which I think is born out by both reaching the final and a VERY close match.

Fed 2006: HC 91/32 123 - AO and USO
Fed 2014: HC: 89/28 117
Novak 2014: HC: 88/32 120
Cilic 2014: HC: 88/23 111
Wawrinka 2014: HC: 86/23 109

I can't explain how Cilic won the USO, or how Stan won the AO.

None of this shows WHY Fed won more % of games in 2006, but we can see from the stats that the problem is in returning, not in serving, on grass. On HCs this year his serving is down just a bit, his returning more so.

The problem is in returning, and that indicates problems not related to serve, obviously. We have all seen more shanks, less this year than last year. The BH is POSSIBLY very close to his peak, possibly because the new racket allow him to step in closer, and he's hit some wicket BHDL shots lately, but his FH is simply not as good as it was. I think that's the biggest issue, not his movement or foot-speed.

We can't compare UEs and winners from year to year because those scores are too subjective. But I'm sure he has more UEs now from errant groundstrokes.

My biggest question: How much is the racket helping? If it is really helping, can he become more comfortable with it in his second year? I think it has helped his serve and his BH, but more than anything else I think he is not as comfortable with the old FH stroke, using the new racket.
 

BreakPoint

Bionic Poster
How did you figure out the "per game" numbers? Did you manually count every single game that he played that year or is there an easier way?

BTW, you should check out his stats for 2001. He had a higher aces per match figure using his PS 6.0 85 than he does in 2014 using his RF97A. :shock:
 

Chanwan

G.O.A.T.
Fed 2006: HC 91/32 123 - AO and USO
Fed 2014: HC: 89/28 117
Novak 2014: HC: 88/32 120
Cilic 2014: HC: 88/23 111
Wawrinka 2014: HC: 86/23 109

I can't explain how Cilic won the USO, or how Stan won the AO.
Try narrowing it down to the AO and US respectively and I'm sure you'll find your answer. They did what players have done for generations: they caught fire/got hot or whatever you want to call it
 

above bored

Semi-Pro
interesting, do you have a link to the interview? And I think it's 100 % true. Fed is no longer quite the highlight reel, he used to be and part of that is him going for the safer shots more often.
You're also right in terms of why win 6-0, when you can win 6-3 without trying too hard
He said a similar thing in the press conference after he won Wimbledon in 2012.

Q. I imagine when you were 22 that you felt like a better tennis player than you were at 18. I'm curious, how you feel about that now? Do you feel like you are a better tennis player now than you were than five years ago?

ROGER FEDERER: I hope so. God, I've practiced so much that I‑‑ you don't want to be worse five years later, you know. (Laughter.)
I feel I have, you know, a great game today. But then again, maybe there were times I had such incredible confidence that you do pull triggers and you pull off shots that maybe today I don't because I maybe do play a bit more the percentages.
I know how hard it is, you know, to pull off those great shots and I know how easy it is to miss, so I'm more aware of these things.
But I'm so happy I'm at the age I am right now, because I had such a great run and I know there's still more possible. You know, to enjoy it right now, it's very different than when I was 20 or 25. I'm at a much more stable place in my life. Yeah, I wouldn't want anything to change. So this is very, very special right now.

http://www.asapsports.com/show_interview.php?id=81429
 

Gary Duane

Talk Tennis Guru
@Gary,
I'm sorry but I'm a bit unsure of what to take away from your post? That Fed is pacing himself better in return games now because he has to physically and his ever so slightly better hold percentage (92 % in 2004 and 90 % in 2006 btw) now compared to 2005 is helping him with that?
No. The opposite. He is pacing himself more now because he HAS to - age. His better serving record takes a bit of stress of returning, but not enough to cover his drop in return percentage. The returning is what makes him less dominating. To me this means he is less relentless when returning.
Also, his TB-record back then was better than now. Plus he had an extra gear, he could switch on - if he needed a late break or if the opponent suddenly won a set.
That's what I am saying. Exactly that. Also the difference between serving and receiving is never more critical than in TBs. The best server in the world still has to get min-breaks. It's harder for Fed to get them now.
So I don't think not getting more than one break per set was risky as is also evidenced by his 73-1 record vs. players outside the top-10.
Let's stay away from one break per set, because that is too general.

Let's say that most likely his wins against players outside the top 10 were at a higher margin.

So he is winning with LESS than 1 break per set against the best.

Consider this: 7/5 5/7 7/6 (TB) For that match it is 1/3 of a break on average per set.

Bottom line: Any drop in either % serve game OR % return game will lower results for the year if the other score does go up as much.
 

Carsomyr

Legend
BTW, you should check out his stats for 2001. He had a higher aces per match figure using his PS 6.0 85 than he does in 2014 using his RF97A. :shock:

He has 589 aces in 71 matches this year compared to 527 aces in 70 matches in 2001...so how do you figure?
 

Gary Duane

Talk Tennis Guru
Try narrowing it down to the AO and US respectively and I'm sure you'll find your answer. They did what players have done for generations: they caught fire/got hot or whatever you want to call it
I was not being literal there. I get that. ;)

But it really does not happen often, because it means that very top players in the world are not at peak level AND some other player gets really hot - and this year it happened twice.

Outliers would be Safin at the AO in 2005, Gaudio at the FO in 2004, and Sampras at the USO in 2002 (because he was not even top 10). With Sampras it proves that you can NEVER count out a former champion at a slam, which is why I think Fed can still win a slam. ;)
 

Chanwan

G.O.A.T.
He has 589 aces in 71 matches this year compared to 527 aces in 70 matches in 2001...so how do you figure?

looking at number of aces per first serve is a better metric imo:
589/3592 = 16,4 % of his first serves were aces in 2014*
527/3425 = 15,4 % of his first serves were aces in 2001*

*discounting any 2nd serve aces in both years - or rather, counting them in the first serve statistics. If you want the more precise number, aces per serve (first and second), simply ad the number of second serves for both years and the percentage will be around 10 % of aces per serve. He may have had more 2nd serve aces in 2001, since he made quite a few more DF's, which could indicate he went for it more often (or wasn't as experienced in terms of hitting his spots with it)
 
Last edited:

Chanwan

G.O.A.T.
I was not being literal there. I get that. ;)

But it really does not happen often, because it means that very top players in the world are not at peak level AND some other player gets really hot - and this year it happened twice.

Outliers would be Safin at the AO in 2005, Gaudio at the FO in 2004, and Sampras at the USO in 2002 (because he was not even top 10). With Sampras it proves that you can NEVER count out a former champion at a slam, which is why I think Fed can still win a slam. ;)

Ah, of course you got it.
Very few outliers in the Fedalovic era, but they were common before. Not sure I agree with you on the part of the top-players not peaking. Djoko played a great match vs. Wawa at the AO and lost. Fed played a great AO for his current standards before running into a Rafa who played very well vs. Fed.
Cilic did get a break with Fed not peaking and Novak out, but I think his level was good enough to beat anything but their very best form, which - let's face it - is not always required for the top-players either. Those last 3 matches from Cilic were everybit as high a level as we've gotten used to from Fedalovic
 

BreakPoint

Bionic Poster
He has 589 aces in 71 matches this year compared to 527 aces in 70 matches in 2001...so how do you figure?

I was going by the OP's number of 552 aces in 2014, and he's played 76 matches this year, not 71.

2014: 552/76 = 7.26 aces per match
2001: 527/70 = 7.53 aces per match.


Here are stats from the ATP site. I'm comparing Fed in 2004 and 2005.

Service Record Year-to-Date: 2005 and 2014

599 Aces 2005: .511 per game
552 Aces 2014: .628 per game
 

Gary Duane

Talk Tennis Guru
Ah, of course you got it.
Very few outliers in the Fedalovic era, but they were common before. Not sure I agree with you on the part of the top-players not peaking. Djoko played a great match vs. Wawa at the AO and lost. Fed played a great AO for his current standards before running into a Rafa who played very well vs. Fed.
Cilic did get a break with Fed not peaking and Novak out, but I think his level was good enough to beat anything but their very best form, which - let's face it - is not always required for the top-players either. Those last 3 matches from Cilic were everybit as high a level as we've gotten used to from Fedalovic
I don't personally believe he would have beaten a healthy Nadal. My God, Wawrinka never before one a SET from Nadal.

Please don't think that I have anything against Wawrinka, because I don't. If he had gone from the AO win to doing really well this year I might have become a fan. ;)
 

Chanwan

G.O.A.T.
I don't personally believe he would have beaten a healthy Nadal. My God, Wawrinka never before one a SET from Nadal.

Please don't think that I have anything against Wawrinka, because I don't. If he had gone from the AO win to doing really well this year I might have become a fan. ;)
He won more points than Rafa at their match at the WTF two months prior and he was up a set and a break before Rafa got injured - beating Novak in the quarters was the last piece he needed to believe. The Wawa Rafa played at the AO and the Wawa who lost 26 straights to Rafa prior to that was by no means the same player.
Of course there's a chance that Rafa could have come back from a set and a break down, but Wawa was playing the kind of tennis that has beaten Rafa time and time again - bold, attacking tennis from a redlining opponent.
Unless Stan's level would have dropped, I don't see Rafa coming back in that match.
So I give full credit to Stan, beating Djoko and Rafa in the same slam ain't easy
 

Chanwan

G.O.A.T.
Maybe they haven't updated the number of matches, but it lists him as having 589 aces, not 552.

So 589/76 = 7.75 aces per match.

@breakpoint,
this should be more significant:
looking at number of aces per first serve is a better metric imo:
589/3592 = 16,4 % of his first serves were aces in 2014*
527/3425 = 15,4 % of his first serves were aces in 2001*

*discounting any 2nd serve aces in both years - or rather, counting them in the first serve statistics. If you want the more precise number, aces per serve (first and second), simply ad the number of second serves for both years and the percentage will be around 10 % of aces per serve. He may have had more 2nd serve aces in 2001, since he made quite a few more DF's, which could indicate he went for it more often (or wasn't as experienced in terms of hitting his spots with it)
 

Gary Duane

Talk Tennis Guru
He won more points than Rafa at their match at the WTF two months prior and he was up a set and a break before Rafa got injured - beating Novak in the quarters was the last piece he needed to believe. The Wawa Rafa played at the AO and the Wawa who lost 26 straights to Rafa prior to that was by no means the same player.
Of course there's a chance that Rafa could have come back from a set and a break down, but Wawa was playing the kind of tennis that has beaten Rafa time and time again - bold, attacking tennis from a redlining opponent.
Unless Stan's level would have dropped, I don't see Rafa coming back in that match.
So I give full credit to Stan, beating Djoko and Rafa in the same slam ain't easy
I remember that AO better than any other because I was thinking how pivotal it would be for the slam race. I remember thinking that this was Rafa's best shot in some time, and coming off the year he had in 2013 I thought it was more or less up for grabs, with Novak the favorite, and it looked like Federer had an outside chance. So when Nadal beat Fed, I figured Nadal just HAD to beat Waw. But what worried me was that awful blister Nadal had, and he started having big problems with his serve. Nadal was simply not fully healthy, even before the back thing.

That said, we can list all the TIMES Nadal was not fully healthy. ;)

But as I said, I figured Novak, Rafa or Fed would be picking up another slam. And I felt the same way at the USO. In my mind this has been a very strange year. I'm eager to see what happens in 2015.
 

Chanwan

G.O.A.T.
I remember that AO better than any other because I was thinking how pivotal it would be for the slam race. I remember thinking that this was Rafa's best shot in some time, and coming off the year he had in 2013 I thought it was more or less up for grabs, with Novak the favorite, and it looked like Federer had an outside chance. So when Nadal beat Fed, I figured Nadal just HAD to beat Waw. But what worried me was that awful blister Nadal had, and he started having big problems with his serve. Nadal was simply not fully healthy, even before the back thing.

That said, we can list all the TIMES Nadal was not fully healthy. ;)

But as I said, I figured Novak, Rafa or Fed would be picking up another slam. And I felt the same way at the USO. In my mind this has been a very strange year. I'm eager to see what happens in 2015.

it has been a strange year and Rafa was favored over Wawa (I remember it very well too and was arguing a lot about it here afterwards :) ). But I disagree that he wasn't healthy when he took the court - the blisters were obviously done bothering from the semi onwards and the serve didn't go until he took the MTO down a break in the second - and it almost came back to his usual serving level in the 4th.
 

falstaff78

Hall of Fame
I really should take a look at tennisabstract and figure out how it works. Out of curiosity, how long does it take for you to find numbers like the above? 2 minutes? 15?

Not even 2 minutes....More like 30 seconds...it's super easy!!! Just google "tennis abstract player name", go to the page, click a couple of buttons on the left to set the filters, and voila!! You can see summary stats on the top and the list of matches corresponding to your filter at the bottom.

The links below will take you straight to all the 2006 numbers I posted.

Fed 2006 overall

Fed 2006 wins
Fed 2006 losses

Fed 2006 vs. top 10
Fed 2006 vs. 11+
Fed 2006 wins vs. 11+

Fed 2006 pre-semis
Fed 2006 semis
Fed 2006 finals
 
Last edited:

Gary Duane

Talk Tennis Guru
It seems to me that saw a lot of 2013 Fed today against Chardy. His serve stats were not the best, and after losing the first game of the first set, he pulled even and had a chance to serve for the set but got broken.

His ground strokes were very uneven, with some really good shots but also a few shanks. And I believe his % of break points was not too good.

I'm wondering if 2014 Fed will come back for the rest of the week?
 

sunny_cali

Semi-Pro
It seems to me that saw a lot of 2013 Fed today against Chardy. His serve stats were not the best, and after losing the first game of the first set, he pulled even and had a chance to serve for the set but got broken.

His ground strokes were very uneven, with some really good shots but also a few shanks. And I believe his % of break points was not too good.

I'm wondering if 2014 Fed will come back for the rest of the week?

Slower court, higher bounce and a tired Fed. I'd be very surprised if Fed moved past the SF in Paris. In fact, I hope he loses earlier and is reasonably fresh for the WTF/Davis Cup.

Regarding 2014 Fed - I'd wait till March 2015 before claiming that it was on par with 2012. Lot of things fell in place for Fed this year..
 
D

Deleted member 307496

Guest
This just proves that Fed hasn't really declined at all. It's just that the competition is much tougher now that prevents him from winning slam. Thank you for solidifying Fed's extremely weak era.
What a joke.
 
Last edited by a moderator:
Top