The Economist welcomes your views. Please stay on topic and be respectful of other readers. Review our comments policy.
You must be logged in to post a comment. Log in to your account.Don't have an account? Register
This method of calculating opposition strength is faulty in an important way. For instance, Nadal has won 10 of his titles on clay but his opposition's record is derived from their performance on all surfaces. It is still the case that most players probably find clay a more difficult surface to play on, i.e., their abilities on clay are weaker than their overall ability. At the end of the day, if more than 60 percent of your GS wins come from one surface, it is hard to get away from a conclusion of relative one-dimensionality. This, of course, is only in comparison to Federer. By any yardstick of performance across his career Nadal is one of the greatest champions the sport has seen.
Oh I love these arguments
2 reasons Nadal has my vote
1. Nadal achieved the career slam at the tender age of 24 to Federers achievement at 28
2. Rafa beat Federer in an epic grass court 5 setter in the 2008 Wimbledon final. Federer never beat Rafa in Paris. In fact he has been humiliated in the 2008 Roland Garros final by Nadal.
So both in their prime Nadal on grass and Federer with revenge on his mind and Rafa still dug out the win against the greatest grass court player ever.
This is all very interesting, but tennis isn't baseball. Stats matter, sure...but not to the extent that a player, a player's history, his/her impact on the sport, his/her stature in the canon of the sport, can be reduced to stats. In that sense (and others) tennis is more like boxing.
You can skew results to "prove" whatever you want. I guess we should also negate when Roger didn't beat Rafa during his Slam wins at Wimbledons 2012 and 2005, US Open wins in 2004- 2007 and 2009 and Australian Open wins in 2004, 2007 and 2010 because- well, you know, Rafa lost earlier. And has stunk at Wimbledon since 2012.
Look, clearly Rafa had Roger's number for a number of crucial years. BUT, their record is also somewhat skewed because Roger kept making later rounds at Roland Garros and clay tournaments and losing to Rafa, where obviously Rafa has always been far superior. AND- it is very arguable that Roger's draws during his Slam wins were "weaker" because Rafa, Novak and Murray were losing during a number of Roger's Slam title wins. It's his fault that he couldn't face them because they lost? Most certainly, throughout their careers, Roger has the most consistent Slam records, and fewer "bad" losses. (Especially if we burn Robredo out of our memories.) AND all of this is not meant to negate Rafa's greatness. It's also obviously not Rafa's fault that he hasn't been able to play Roger at the US Open during Rafa's wins this year, 2013 and 2010 because Roger lost earlier. Spoiler Alert: Their careers are not over. Roger and Rafa arguably are the two greatest players ever already. Why not just let this all play out already>
A simpler way would be to check each one's dominance by deleting their favourite surface - i.e. where each has won maximum Grand Slams. By this method, Federer has 19 - 8 Wimbledon = 11. Nadal on the other hand, has 16 - 10 French = 6 and is at the same level as Djokovic 12 - 6 Australian Open = 6. Clearly the King of Clay is only just that. Federer is way ahead in overall dominance.
How do you factor in gs winning percentage? I.e. If Federer had won every gs he entered (say 19/19) this would have to weigh in his favor. In this case you take all the gs's each entered and assign a score of '1' (which is what u've done by normalizing the 1.23 avg).
Total normalized score equals the number of gs's entered. Then divide ur adjusted total score by the total possible score to get an adjusted 'gs winning ratio'.
So if Rafa has entered 40 gs's over his career, he has 40 possible gs points. He's won 18.8 points so that's a 47pct adjusted win percentage.
Another way of comparing both is to remember that 50% of Grand Slams are in hard courts, and that factor favors Federer. Had tradition be that two Grand Slams are played in clay (instead of one) Nadal, following my calculations, would have 24 to 15 for Federer. With two grass completions a year, Federer would have got 22, to 17 for Nadal.
To be fairer, if clay, grass and hard courts would have a 33% value each (instead of 50% hard courts and 25% de other two), Nadal would have had 19 Grand Slams vs 18,7 for Federer. Nadal is GOAT.
Silly question really. Laver won 2 calendar year Grand Slams 5 years out of action being professional would have won probably a good 10 more at least - give Rog and Nadal a wooden racket with natural gut strings I'd have money on the Swiss and Aussie to dominate. Rog raised the game hugely in his early years - setting a new benchmark for Djoko and Nadal and co - playing the likes of Safin, Agassi and Safin who were no slouches. He may well be the 2nd best clay court player ever having the misfortune of playing in the era of Nadal whose game matches up well to Fed's. Most pros and tennis players would agree Nadal is the best clay courter ever, Rog the best on fast courts and the most elegant player they have seen. Rest is just media noise.
How about going with simpler math by looking at GS finals winning percentages? Fed: 19/29 = 66%, Nadal: 16/23 = 69% & Djoker: 12/21 = 57%. Also, not only does Rafa own Fed at RG (5-0), but also at the AO (3-1) and is still a respectable 1-2 at Wimby.
Percentages are a nonsense. So if someone won 3 out of 4... 75%? You have to consider numbers of tournaments played and the longer someone plays the lower the percentage. Also you to consider all gs played, not just finals.
Since turning 30 Roger has played 22 grand slams and won only 3 = 13.6%. Before 30 his record was 16 out of 51 played = 31.4%. Rafa's pre 30 record was 14 out of 44 = 31.8%. So on an age-equivalent basis Roger wins in total majors but they equal on percentages won pre-30. How well Rafa does in the coming years is critical to both indicators
Stop applying the titles Mr. It seems weird and unnecessarily formal.
How can Federer be considered better than Nadal if Federer has lost 62% of the games played vs. Nadal?
how can nadal be considered better than davydenko when he lost 55% of the time
Not a good comparison since Davydenko's accomplishments are far inferior to Nadal's (titles, ranking, Grand Slams, etc.); it's also important that if Nadal wins 1 more game vs. Davydenko then they'd be at 50%.
In comparison, when looking at Federer vs. Nadal their accomplishments are very close to identical but the head-to-head results are very skewed to Nadal; it'd also take 9 straight Federer wins to get to 50%.
In the end this is about who you like... clearly!
One of the big drawbacks of this method is that it uses the Elo rating attained on all surfaces to assign probabilities on clay. This probably grossly underestimated Nadal's chance of winning Roland Garros, given that the Elo rating is heavily skewed by matches on hard court, as it is the most common surface. As a result, his wins were assigned more weight by this methodology.
A similar issue arises when looking at the head to head count between Federer and Nadal (14-23). Counterintuitively, this domination by Nadal is largely due to the fact that Federer was so good on Clay: had Federer not been as good on clay, he would not have reached the final against Nadal with such consistency. Less confrontations on clay would thus have happened, and the overall matchup would be influenced more heavily by confrontations on grass/hard, on which Federer has a slight edge (though not as big as Nadal has on clay. Nobody disputes the fact that Nadal is the GOAT on clay).
False narrative! You either don't know what youre talking about OR are so ill-informed you just made a point FOR Nadal's favor and against Roger... You said "Federer was so good on Clay: had Federer not been as good on clay, he would not have reached the final against Nadal with such consistency." Well let's see Federer has reached 5 total French open finals in his career and Roger won only once again Robin Soderling in that lone FO title. HOWEVER, Nadal has reached FIVE total Wimbledon finals, BUT he has won 2 of those! And as we all know, Rafa even BEAT OUT Federer on grass in Wimbledon to take a title, whereas Roger has never sniffed a French Open win against Rafa! Therefore, Roger's weakest surface(clay) is definitely weaker than Rafa's(grass), no question at all. AND additionally, Rafa's STRONGEST SURFACE (Clay) is definitively waaaay better than Roger's(grass) and futhermore Rafa's best is better than anyone in the history of the sport. It is indisputable dominance never before seen, and never again to be seen. Nice try, and thanks for the help though!
Elo ratings are fascinating for tennis nuts (like me) but are boring and distant for most of the public who would like to keep it simple and rate GOAT or greats on number of Grand Slams won.
As for Elo rating their practicality is not very high as Tennis is physical game unlike Chess where Brains matter only. And physical endurance and capability degrade with advancing age hence 2 Grand Slams by Mr. Federer at ripe age of 35 should count more than Mr. Nadal's (who is spry 31) Two Grand Slams this year. Another thing which needs to built in Elo would be is player coming off from injury or long lay off which may decrease his/her seeding hence the wins should count more.
Anyways whatever is the case, Present generation of Male Tennis Players is Golden generation, period.
As for rivalry between Mr. Nadal and Mr. Federer is concerned Mr Federer is GOAT but Mr. Nadal gets better off Mr. Federer when they play.
PS: Mr. Murray is very good tennis player but his being British should be not be the reason for 'The Economist' to put him in league of Mr. Federer, Mr. Nadal and Mr. Djokovic. After all, come on 3 slams vs. 19,16 and 12, you gotta be kidding me.
The fact that the 2013 French Open was the highest rated of Rafa's wins shows how flawed this methodology is. Stan Wawrinka in 2013 wasn't as good as he later became and had zero shot of beating Nadal. Djokovic meanwhile wasn't 'prime' during that summer either. In 2013 he failed to win Indian Wells and Miami, and failed to reach the final of Rome and Madrid. Moreover many of Nadal's French Open wins would carry bigger weights because he beat Federer in them. This is also silly as Nadal had a huge advantage over Federer on the surface in particular on the Chatrier court (relative to other clay courts).
It seems as though the author doesn't actually follow tennis that closely, and passes off a deeply flawed statistical analysis as objective evidence.
2013 French open Nadal was fourth seed and played Djokovic in the semi final. Much tougher route to final. Djokovic also beat Nadal in straight sets in Monte Carlo final when Rafa was the 8 time defending champion so it's hard to argue that beating Djokovic was going to be easier than normal for Rafa.
Nadal beat Federer in 4 finals in Roland Garros. The same argument you make about Elo understanding the probability of him beating Federer would apply to Federer's two wins over Nadal at Wimbledon in 2006-2007. And unlike Federer, Rafa actually beat Federer at Wimbledon in 2008 despite his natural handicap on the surface.
The author's statistical take may be imperfect but I would not say that he doesn't follow tennis closely. Most of his previous posts on the sport have been pretty decent.
Rafa was only 4th seed as he took 2012 off after Wimbledon. After beating Rafa at Monte Carlo, Novak lost to a cramping Dimitrov in the 2nd round of Madrid, and to Berdych in the Quarters of Madrid. He was slumping (relatively) for much of the year so certainly wasn't "peak." I agree he was a tough out though. Don't agree that Wawrinka in QF and Ferrer in F were anything beyond a walk in the park.
It's funny that the title of the piece says "Not just the king of clay" but the methodology used basically gives Rafa extra credit for his french open wins because of his dominance on clay (as it gives extra weight to wins against wawrinka, federer and djokovic even though they were all massive underdogs in almost all of their french open encounters.
Also rafa's win over roger in '08 encpsulates the discussion pretty well. That is that Rafa is further ahead of Roger on clay than Roger is of him on all the other surfaces. However given that clay is only 25% of majors and 33% of masters, all the others have a greater weighting, and Rafa hasn't done enough "YET" to overtake Roger either by way of winning even more clay events or by winning more non-clay events.
Can you do similar analysis of Nicklaus vs. Woods in terms of major victories?
Sorry. This logic does not make sense. So, assume a player is truly a GoAT, likely winning consistently most of the matches against all of his/her peers. Then the likely chance of winning across all the matches in a slam would be significantly high. This logic would then say, the player is worthy of only much lesser slams. It is not a players fault that others are comparatively inconsistent. So would USAIN BOLT be worth less than or more than 19 golds(Olympics/World Championships) that he had won? You don't need to calculate it. Based on what the article states we all know what the result will be.
Good attempt, but I couldn't actually understand the method you have applied as the description is a bit opaque. How exactly do you estimate the probabilities and what is the sense in deducting them from one? Why is this better than simply looking at their Elo ratings? Also, as others have said below - allowances need to be made for versatility across surfaces; Rafa is a bit of a specialist, whereas Fed performs well across the board. Finally, the statistics somewhat miss the point about what makes the Fed the greatest tennis player of all time in the eyes of his fans. David Foster Wallace's essay probably does this argument better justice.
Quite. DFW essay extrapolates the difference between the two far better than any data analysis can. Nadal, though a great player, is interesting primarily a means of describing the limits of Federer's genius.
Quite. DFW essay explains the difference between the two far better than any data analysis can. Nadal, though a great player, is interesting primarily a means of describing the limits of Federer's genius.
Federer's head to head losing record vs Nadal is proof enough that the term "genius" simply doesn't apply. He's a genius despite his 14-23 record against Nadal? He's a genius despite his 3-9 record vs Nadal in Grand Slam matches? This is how genius is defined? And please don't mention the clay court encounters. If Federer were truly a tennis genius, he would win on any surface.
I enjoyed the article and especially the analytical aspect however, I do believe the "elo" score is flawed for the simple reason that Roger Federer occupied the top seed for pretty much the entirety of the period the analysis was conducted on. By virtue of being the highest seed he would have been granted an easier route to the final of that grand slam compared to his competitors, therefore reducing the adjusted grand slam points he accumulates. The analysis also doesn't factor in the advantage some players have on different surfaces, namely Rafa's 10 Grand Slam wins on clay - a phenomenal feat. But it means 62.5% of his grand slam wins have come from one surface which is evidence to suggest that the elo score cant be universally applied to every player and on every surface.