# Looking at Rafa's W% on clay

##### Hall of Fame
Rafa's winning percentage on clay is often shown in a way that doesn't give it as much praise as it should receive. To show you what I mean I've used the formula y = x/(1-x)
What this does is show how win percentage (x-axis) correlates with how many average wins/loss (y-axis)

What this shows is that when you approach 100% on the x-axis, the steeper the climb on the y-axis. Each loss weighs more than each win, assuming a player's win percentage is above 50%. A loss to a great player means they are rubberbanded back towards the 50% mark, a win for a terrible player means they are rubberbanded towards the 50% mark. This makes sense, since it is literally impossible to attain a 100% win/loss ratio after just one loss, and also impossible to attain a 0% win/loss ratio after one win. There is a monumental difference between just 95% and 99%, whereas there's practically no difference between 50% and 54%. That is why it's a mistake to look at two players and say x player is only 4% better than y player. Instead, it's better to look at the average amount of wins/loss each player has and find the difference. Bearing this in mind, let's compare Rafa on clay, to Roger on grass.

X-mark represents Career W%
O-mark represents Peak Career W%
Green = Roger
Orange = Rafa

Rafa's Career W% on clay = 11.527778 wins/loss
Roger's Career W% on grass = 6.833333 wins/loss
The difference between each player's wins/loss for their entire career: 11.527778 - 6.833333 = 4.694445 wins/loss

Rafa's Peak Career W% on clay = 14.285714
Roger's Peak Career W% on grass = 7.666667
The difference for when both player's career's hit their peak: 14.285714 - 7.666667 = 6.619047 wins/loss

The second greatest peak winning percentage of all time (In the open era) on any single surface belongs to Rod Laver on hardcourts: 91.00% = 10.111111 wins/loss

14.285714 - 10.111111 = 4.174603 wins/loss

Discuss

Last edited:

#### AnOctorokForDinner

##### G.O.A.T.
Don't use ATP stats for part OE players like Laver, seriously.

##### Hall of Fame
Don't use ATP stats for part OE players like Laver, seriously.
Soz

#### titoelcolombiano

##### Hall of Fame
I've said it in a recent thread - beating Rafa at Roland Garros is one of the toughest tasks in world sport

#### Firstservingman

##### Talk Tennis Guru
Nice.

Fed has also played several more years and reached his potential longer after he started playing though. So there's that to consider. But nothing stated here is wrong.

##### Hall of Fame
Nice.

Fed has also played several more years and reached his potential longer after he started playing though. So there's that to consider. But nothing stated here is wrong.
That's true, I was thinking of comparing both their best stretch of winning matches with only one loss. That would close the window so to speak, instead of having to start from the beginning of their careers if that makes sense. Might do that later. Roger's peak level of play does not exactly correlate with these graphs, but in terms of career averages or domination over their entire career I think it's a pretty good measure.

#### van_Loederen

##### Professional
what a versed duck.

props also to Nadal.

D

#### Deleted member 756486

##### Guest
This just confirms what I’ve always suspected.

Ducks are smarter than humans.

#### Zoolander

##### Hall of Fame
No! I dont come to a friggin tennis forum to have discussions about statistical maths!

Just tell me if your a Roger or Rafa fan, and then start b1tching about the other guy. Thats so much easier!

##### Bionic Poster
I've said it in a recent thread - beating Rafa at Roland Garros is one of the toughest tasks in world sport
The most epic certainly

#### Shaolin

##### G.O.A.T.
We get it already. He never loses on clay, as if we hadn't noticed.

Excellent post

##### Hall of Fame
One thing to add: because of the principle you pointed out (i.e., that the better your winning % the more a loss will penalize it), the number of matches played on any given surface becomes important. The more a player plays on a particular surface, the more chances for an injury or a lower level of play or a bad matchup, etc. In essence, it becomes harder and harder to continue to win at a high rate over a greater number of matches.

So considering the stats OP set forth above, it's especially impressive that not only has Nadal demonstrated a markedly higher # of wins per loss on clay vs Federer on grass, but he has also done so over 451 matches, whereas Roger has only played 188 matches on grass. That means that in order for Roger to match Nadal's feat on grass, Roger would have to play another 263 matches on grass (which is 139.9% of the number of matches Roger has played as of today), and he'd have to win a staggering 251 of them.

One more thing. I think it's also worth pointing out that nobody regularly plays, trains or grows up on grass, whereas they do on clay in spades. So today you almost never get a matchup against a player who is super intimately familiar with the surface as a specialist, whereas there are a number of players who have total comfort and familiarity on clay, have honed their games for clay, and indeed may have even played Nadal regularly on clay growing up.

When you think about it that way, it's almost impossible to understate just how phenomenal Nadal's record is on clay. It's likely that we will never see this type of performance again statistically and I'd bet it's an outlier on par with Dimaggio's 56-game hitting streak.

#### tennis_pro

##### Bionic Poster
Can you please do Nadal on grass since 2011 too?

##### Hall of Fame
We get it already. He never loses on clay, as if we hadn't noticed.
Way to entirely miss the point

#### Fedeonic

##### Hall of Fame
To show you what I mean I've used the formula y = x/(1-x)
Don't you mean f(x)=x/(x+1)?

#### Tennisanity

##### Legend
Grass has much slimmer margins, much harder to win on grass compared to clay. Ask Rafa about that from 2012 till now.

#### GabeT

##### Legend
And yet when I posted sometime ago that Nadal was better at clay than Fed in grass I was told I was a troll. Or something like that.

#### Otacon

##### Hall of Fame
The most impressive stat to me is 18-0 in Bo5 clay finals.

#### Shaolin

##### G.O.A.T.
Way to entirely miss the point
I get the point. OP had a spare 12 minutes on his hands and made a graph to brag on his boy Rafa.

##### Hall of Fame
One thing to add: because of the principle you pointed out (i.e., that the better your winning % the more a loss will penalize it), the number of matches played on any given surface becomes important. The more a player plays on a particular surface, the more chances for an injury or a lower level of play or a bad matchup, etc. In essence, it becomes harder and harder to continue to win at a high rate over a greater number of matches.

So considering the stats OP set forth above, it's especially impressive that not only has Nadal demonstrated a markedly higher # of wins per loss on clay vs Federer on grass, but he has also done so over 451 matches, whereas Roger has only played 188 matches on grass. That means that in order for Roger to match Nadal's feat on grass, Roger would have to play another 263 matches on grass (which is 139.9% of the number of matches Roger has played as of today), and he'd have to win a staggering 251 of them.

One more thing. I think it's also worth pointing out that nobody regularly plays, trains or grows up on grass, whereas they do on clay in spades. So today you almost never get a matchup against a player who is super intimately familiar with the surface as a specialist, whereas there are a number of players who have total comfort and familiarity on clay, have honed their games for clay, and indeed may have even played Nadal regularly on clay growing up.

When you think about it that way, it's almost impossible to understate just how phenomenal Nadal's record is on clay. It's likely that we will never see this type of performance again statistically and I'd bet it's an outlier on par with Dimaggio's 56-game hitting streak.
Yep, we can only speculate how either player would've performed had the conditions been the same or flipped, but another thing that matters is the difference between BO3 and BO5 which I didn't even get into here. Once matches use the BO5 format, the chances of upset decrease drastically, so what's impressive is that Rafa has been able to keep his wins/loss ratio so high when he plays an inordinate amount of BO3 matches every season. Most people expect him to lose just once, but only during the BO3 format.

##### Hall of Fame
Is that true? The top winning percentages by court type seem to suggest otherwise

https://en.wikipedia.org/wiki/Open_Era_tennis_records_–_men's_singles#Per_court_type_career_totals_2

And looking at multiple champions at each of RG and W in the open era:

RG: 2 at 5+ titles, 5 at 3+

W: 3 at 5+ titles, 6 at 3+
This also confused me. Originally I had the same assumptions as this guy (That because the margins are slimmer on grass courts, the chances of upset were higher) but the real-life evidence seems to suggest the exact opposite conclusion. The best players on hard/grass courts generally have a better wins/loss ratio than the best players on clay. Have not figured out why that is the case though.

Perhaps it could have to do with the ratio of matches played using BO3:BO5

#### AnOctorokForDinner

##### G.O.A.T.
This also confused me. Originally I had the same assumptions as this guy (That because the margins are slimmer on grass courts, the chances of upset were higher) but the real-life evidence seems to suggest the exact opposite conclusion. The best players on hard/grass courts generally have a better wins/loss ratio than the best players on clay. Have not figured out why that is the case though.

Perhaps it could have to do with the ratio of matches played using BO3:BO5
More matches -> tougher to keep up the level throughout the season -> a likely loss at a lesser tournament. But in BO5, the tournament(s) that all top players on that surface seek to time their peak for, it's different and peak to peak grass should be more upset-friendly, but harsher physicality on clay means shorter prime for most claycourters, hence still lower winning % career-wise.

#### Tennisanity

##### Legend
This also confused me. Originally I had the same assumptions as this guy (That because the margins are slimmer on grass courts, the chances of upset were higher) but the real-life evidence seems to suggest the exact opposite conclusion. The best players on hard/grass courts generally have a better wins/loss ratio than the best players on clay. Have not figured out why that is the case though.

Perhaps it could have to do with the ratio of matches played using BO3:BO5
Probably because serve is a big factor. The top players have great serves. A serve is a stroke completely under your control.

##### Hall of Fame
More matches -> tougher to keep up the level throughout the season -> a likely loss at a lesser tournament. But in BO5, the tournament(s) that all top players on that surface seek to time their peak for, it's different and peak to peak grass should be more upset-friendly, but harsher physicality on clay means shorter prime for most claycourters, hence still lower winning % career-wise.
Agree with your points, except this one. Even Borg falls short of many hard and grass court greats when it comes to his Peak Career Win Rate on clay. Still, all the more impressive that Nadal has such a high win rate even now (granted it has dropped like 3 levels). Basically FrontHeadlock's point about number of matches.

#### AnOctorokForDinner

##### G.O.A.T.
Agree with your points, except this one. Even Borg falls short of many hard and grass court greats when it comes to his Peak Career Win Rate on clay. Still, all the more impressive that Nadal has such a high win rate even now (granted it has dropped like 3 levels). Basically FrontHeadlock's point about number of matches.
I meant lower career % for clay because of the physical effort. Didn't realise the wording wasn't clear, sorry.

#### True Fanerer

##### G.O.A.T.
Clay is over. Time to move on. I know it's hard for you, but you have to.

##### Hall of Fame
Grass is a much faster surface. If you're slightly off an upset is likely. On clay you can grind it out.
Intuitively I understand what you're saying, but I think we'd need some data to prove it.

##### Hall of Fame
Yep, we can only speculate how either player would've performed had the conditions been the same or flipped, but another thing that matters is the difference between BO3 and BO5 which I didn't even get into here. Once matches use the BO5 format, the chances of upset decrease drastically, so what's impressive is that Rafa has been able to keep his wins/loss ratio so high when he plays an inordinate amount of BO3 matches every season. Most people expect him to lose just once, but only during the BO3 format.
That's right.

And even more impressively, I'll bet Nadal has played a higher % of matches on clay as BO3 than Federer has on grass.

#### Tennisanity

##### Legend
Intuitively I understand what you're saying, but I think we'd need some data to prove it.
I don't think data is really applicable here. Imagine an era with Sampras and no good clay court players. Sampras will every Wimbledon, but FO will be spread out over a number of players just because there doesn't happen to be any really good clay courters that can dominate. The data would say it's easier to dominate on grass, but it's not really the case because of the specific players involved. That's an extreme example, but that kind of nuance will render the data useless.

#### ak24alive

##### Legend
Rafa's winning percentage on clay is often shown in a way that doesn't give it as much praise as it should receive. To show you what I mean I've used the formula y = x/(1-x)
What this does is show how win percentage (x-axis) correlates with how many average wins/loss (y-axis)

What this shows is that when you approach 100% on the x-axis, the steeper the climb on the y-axis. Each loss weighs more than each win, assuming a player's win percentage is above 50%. A loss to a great player means they are rubberbanded back towards the 50% mark, a win for a terrible player means they are rubberbanded towards the 50% mark. This makes sense, since it is literally impossible to attain a 100% win/loss ratio after just one loss, and also impossible to attain a 0% win/loss ratio after one win. There is a monumental difference between just 95% and 99%, whereas there's practically no difference between 50% and 54%. That is why it's a mistake to look at two players and say x player is only 4% better than y player. Instead, it's better to look at the average amount of wins/loss each player has and find the difference. Bearing this in mind, let's compare Rafa on clay, to Roger on grass.

X-mark represents Career W%
O-mark represents Peak Career W%
Green = Roger
Orange = Rafa

Rafa's Career W% on clay = 11.527778 wins/loss
Roger's Career W% on grass = 6.833333 wins/loss
The difference between each player's wins/loss for their entire career: 11.527778 - 6.833333 = 4.694445 wins/loss

Rafa's Peak Career W% on clay = 14.285714
Roger's Peak Career W% on grass = 7.666667
The difference for when both player's career's hit their peak: 14.285714 - 7.666667 = 6.619047 wins/loss

The second greatest peak winning percentage of all time (In the open era) on any single surface belongs to Rod Laver on hardcourts: 91.00% = 10.111111 wins/loss

14.285714 - 10.111111 = 4.174603 wins/loss

Discuss
Absolutely brilliant stuff OP. I made calculations about how much winning and losing a match when you are at 90% or 80% costs. The obvious conclusion is that when you are at a higher percentage of wins winning a match adds very less to your percentage and losing a match subtracts very much from your percentage and when you are at a lower percentage a loss costs you less percentage than a loss being already at high percentage and a win being already at a lower percentage gains you more percentage than a win being already at a high percentage would have.
I took five examples to see the hike in percentage when you win being already at a lower percentage vs the hike in percentage when you win being at a higher percentage.

Your analysis is brilliant but the only problem is the graph and the ratio(the y axis) you used.
Look we all know what Win ratio means(no of wins/no of matches) but what does your ratio(no of wins/no of losses) even mean? All the conclusions you made are right as I said above but my friend a win/loss ratio basically skews the numbers beyond the conclusion we needed.
I'll give you an example.
We will call your ratio "duck" for the sake of ease of conversation from now on.
Now suppose all the sh*t I am about to write.
So a 92% win percentage ATG has a duck value of 11.5.
93% has a duck value of 13.285 .
94% has 15.67.
95% has 19.
96% has 24.
97% has 32.33.
See? I don't think there's a lot of difference between the greatness of Mr. 94 and Mr. 95 but your ratio kills it with the difference.
Same with 95 and 96.
I get the point that there's more difference if you go from 92 to 94 than there is if you go from 50 to 60 but the question is how much more is that difference?
Now as I have clearly explained above your function(f(x)=x/1-x) is just giving too much disparity when we go to higher percentage so by this ratio we will end up relatively overhyping an ATG when compared to another ATG just behind him in win percentage (in this case Nadal).
I hope you see my point

PS: pardon my bad english and explanation. many thanks for the great post again

#### Deon Sanders

##### Legend
Is there a good way to make a chart showing the French Open having zero relevance

##### Hall of Fame
Is there a good way to make a chart showing the French Open having zero relevance
That's a logical impossibility.

##### Hall of Fame
Now suppose all the sh*t I am about to write.
So a 92% win percentage ATG has a duck value of 11.5.
93% has a duck value of 13.285 .
94% has 15.67.
95% has 19.
96% has 24.
97% has 32.33.
See? I don't think there's a lot of difference between the greatness of Mr. 94 and Mr. 95 but your ratio kills it with the difference.
Same with 95 and 96.
Here's another way to see how crazy this effect is.

The difference in number of wins per loss moving from 94 to 96%, a mere two percentage points, is 8.33, meaning you need an additional 8.33 wins for every loss to take your win % up those two points.

Now assume you're sitting at 50% which effectively means that you lose in the second round of every tournament because you have 1 win per loss. Well, if you increase that by 8.33 wins per loss, you would sit at a win % of 90.32%, an increase of over 40% and one that would give you the second best win percentage of all time on any surface in the open era.

##### Hall of Fame
Absolutely brilliant stuff OP. I made calculations about how much winning and losing a match when you are at 90% or 80% costs. The obvious conclusion is that when you are at a higher percentage of wins winning a match adds very less to your percentage and losing a match subtracts very much from your percentage and when you are at a lower percentage a loss costs you less percentage than a loss being already at high percentage and a win being already at a lower percentage gains you more percentage than a win being already at a high percentage would have.
I took five examples to see the hike in percentage when you win being already at a lower percentage vs the hike in percentage when you win being at a higher percentage.

Your analysis is brilliant but the only problem is the graph and the ratio(the y axis) you used.
Look we all know what Win ratio means(no of wins/no of matches) but what does your ratio(no of wins/no of losses) even mean? All the conclusions you made are right as I said above but my friend a win/loss ratio basically skews the numbers beyond the conclusion we needed.
I'll give you an example.
We will call your ratio "duck" for the sake of ease of conversation from now on.
Now suppose all the sh*t I am about to write.
So a 92% win percentage ATG has a duck value of 11.5.
93% has a duck value of 13.285 .
94% has 15.67.
95% has 19.
96% has 24.
97% has 32.33.
See? I don't think there's a lot of difference between the greatness of Mr. 94 and Mr. 95 but your ratio kills it with the difference.
Same with 95 and 96.
I get the point that there's more difference if you go from 92 to 94 than there is if you go from 50 to 60 but the question is how much more is that difference?
Now as I have clearly explained above your function(f(x)=x/1-x) is just giving too much disparity when we go to higher percentage so by this ratio we will end up relatively overhyping an ATG when compared to another ATG just behind him in win percentage (in this case Nadal).
I hope you see my point

PS: pardon my bad english and explanation. many thanks for the great post again
I understand what you're saying, but perhaps that's the reason why nobody is able to achieve that in the first place? You might be able to achieve such high numbers when the number of matches played is too low, for example when Nadal/Federer won so many matches in a row on their favorite surface, we knew overall their level wasn't at infinity (lol), but the less data (No. of matches played), the less accurate the results. The more data, the more accurate the results. So my answer to you is this graph is not accurate when looking at the 96-99% percentile range IF we're looking at players who have won low number of matches throughout their career. The higher the percentage, the less accurate it's going to be if you're looking at a short span of matches comparatively to the percentage. Basically, anything from 97-99.99% range we'd need thousands of matches to be sure it's accurate. But you're free to treat wins/loss however you want tbh. All it is, is a fraction.

#### ak24alive

##### Legend
Here's another way to see how crazy this effect is.

The difference in number of wins per loss moving from 94 to 96%, a mere two percentage points, is 8.33, meaning you need an additional 8.33 wins for every loss to take your win % up those two points.

Now assume you're sitting at 50% which effectively means that you lose in the second round of every tournament because you have 1 win per loss. Well, if you increase that by 8.33 wins per loss, you would sit at a win % of 90.32%, an increase of over 40% and one that would give you the second best win percentage of all time on any surface in the open era.
That is great analysis. I like this one a lot.

#### ak24alive

##### Legend
I understand what you're saying, but perhaps that's the reason why nobody is able to achieve that in the first place? You might be able to achieve such high numbers when the number of matches played is too low, for example when Nadal/Federer won so many matches in a row on their favorite surface, we knew overall their level wasn't at infinity (lol), but the less data (No. of matches played), the less accurate the results. The more data, the more accurate the results. So my answer to you is this graph is not accurate when looking at the 96-99% percentile range IF we're looking at players who have won low number of matches throughout their career. The higher the percentage, the less accurate it's going to be if you're looking at a short span of matches comparatively to the percentage. Basically, anything from 97-99.99% range we'd need thousands of matches to be sure it's accurate. But you're free to treat wins/loss however you want tbh. All it is, is a fraction.
That is exactly the answer I was expecting. It is ridiculous to talk about this ratio when it makes the difference between f(x1) and f(x2) where x1 and x2 are 2 consecutive x(win%) so high. So maybe this ratio gives a more realistic picture if we stay inside the 85% mark and that is primarily why I was saying that Nadal and Federer difference in the graph looks much more than it is because Nadal is 91 and Fed is 75. These are till 2017 figures.
But that 91 speaks for itself and moreover the fact you uncovered that keeping it at 91 is much more difficult says it all.

##### Hall of Fame
That is exactly the answer I was expecting. It is ridiculous to talk about this ratio when it makes the difference between f(x1) and f(x2) where x1 and x2 are 2 consecutive x(win%) so high. So maybe this ratio gives a more realistic picture if we stay inside the 85% mark and that is primarily why I was saying that Nadal and Federer difference in the graph looks much more than it is because Nadal is 91 and Fed is 75. These are till 2017 figures.
But that 91 speaks for itself and moreover the fact you uncovered that keeping it at 91 is much more difficult says it all.
Hmm, that isn't exactly what I was trying to say, because I would still back Nadal's and Federer's numbers because there is enough data to support it. I would also 100% back anything in the very high percentages of this graph if there was enough data to prove it.

Here's another way to see how crazy this effect is.

The difference in number of wins per loss moving from 94 to 96%, a mere two percentage points, is 8.33, meaning you need an additional 8.33 wins for every loss to take your win % up those two points.

Now assume you're sitting at 50% which effectively means that you lose in the second round of every tournament because you have 1 win per loss. Well, if you increase that by 8.33 wins per loss, you would sit at a win % of 90.32%, an increase of over 40% and one that would give you the second best win percentage of all time on any surface in the open era.
The effect seems crazy, but I 100% back those numbers. The only limiting factor to the accuracy of those numbers is the number of matches that are played. If we were able to simulate ageless players playing hundreds of thousands of matches and that turned out to be their percentage, then those figures would be entirely accurate. For example if player x played at 95.24%(20 wins/loss) his career win rate is still twice as good as 90.91% (10 wins/loss). Whether you want to say he is twice as good a player is up for debate though, but there's no denying the percentages.

Edit: Another example. 98.8% (80 wins/loss) is still twice as good as 97.6% (40 wins/loss)

I know it sounds insane, but really the only thing you have to agree on to agree with me is that 100% is an infinitely better percentage than anything else. You would need an infinite amount of data to 100% prove someone had a 100% win rate though, that's the catch.

Last edited:

#### ak24alive

##### Legend
Hmm, that isn't exactly what I was trying to say, because I would still back Nadal's and Federer's numbers because there is enough data to support it. I would also 100% back anything in the very high percentages of this graph if there was enough data to prove it.

The effect seems crazy, but I 100% back those numbers. The only limiting factor to the accuracy of those numbers is the number of matches that are played. If we were able to simulate ageless players playing hundreds of thousands of matches and that turned out to be their percentage, then those figures would be entirely accurate. For example if player x played at 95.24%(20 wins/loss) his career win rate is still twice as good as 90.91% (10 wins/loss). Whether you want to say he is twice as good a player is up for debate though, but there's no denying the percentages.

Edit: Another example. 98.8% (80 wins/loss) is still twice as good as 97.6% (40 wins/loss)
I was about to post again before you posted.
Maybe you misunderstood me when I said that this graph doesn't put a realistic image when talking above 85%. What I meant is that it takes a very sharp turn after some value of x and although numbers will back it as you said but we can't conclude that Mr 95.24% guy is twice as greater as Mr 90.91% .
So basically my point was that this graph although is great but doesn't show the level of the players as then it would mean Nadal is approximately thrice as great as Roger on clay. I hope it is clear what I meant.

#### Tennis_Hands

##### Bionic Poster
Can I get the number of RG titles (ok, ok, number of clay matches), where he didn't go over the time limit on the most important points even once?

That would be epic.

Thank you!

#### ak24alive

##### Legend
Can I get the number of RG titles (ok, ok, number of clay matches), where he didn't go over the time limit on the most important points even once?

That would be epic.

Thank you!

You were missed here. There was too much logical and sensible talk going on and to bring back the balance of ttw someone like you was needed. Thanks to bestow your saltiness on this thread.

#### van_Loederen

##### Professional
Is that true? The top winning percentages by court type seem to suggest otherwise

https://en.wikipedia.org/wiki/Open_Era_tennis_records_–_men's_singles#Per_court_type_career_totals_2

And looking at multiple champions at each of RG and W in the open era:

RG: 2 at 5+ titles, 5 at 3+

W: 3 at 5+ titles, 6 at 3+
actually it's rather the upset rate on faster surfaces that is higher. on grass however do specialists get some edge.

and what AnOctorokForDinner said ("harsher physicality on clay means shorter prime for most claycourters, hence still lower winning % career-wise") matches with my own observations and might explain those stats.

Once matches use the BO5 format, the chances of upset decrease drastically
drastically? i really doubt it. i think it's rather because the Majors are the main events that most players focus on.

f.i. at the YEC do the favorites get through pretty regularly too. well, it's only top 8 plus RR though.

but there's also the Sunshine double. sometimes players would be out of form after their holidays or tired from Dubai or so, but it rarely looks to me like they are directly upset. especially thanks to Miami right afterwards (and Dubai/Acapulco prior) one can usually notice these things very well.

the other Masters events are either scheduled as preparation for the Majors (to an extent) or during more or less holiday periods for the top players (MC, Canada, (Shanghai)).

and while Bo3 is indeed a bit short (i remember f.i. Isner doing some serious menace), Bo5 tends to favor physicality over technique during the final rounds.

##### Hall of Fame
actually it's rather the upset rate on faster surfaces that is higher. on grass however do specialists get some edge.
again, intuitively that makes sense to me, but you're both merely asserting that without backup. i wouldn't be surprised if it were true, but at least the limited data we have regarding the top players doesn't really bear that out.

#### van_Loederen

##### Professional
again, intuitively that makes sense to me, but you're both merely asserting that without backup. i wouldn't be surprised if it were true, but at least the limited data we have regarding the top players doesn't really bear that out.
nowadays there are no fast hardcourts anymore on the ATP tour
and i was mostly (not entirely) disagreeing with that other poster.

as for backing it with numbers of past periods, it might be somewhat difficult. during the 90s existed more quicker courts, but the greater differentiation of surfaces also rendered the rankings less usable for compiling such stats.
but one often saw how important serve and form on the day was in faster hardcourts (and back then was also grass still not such an alien).

nowadays one can still get an impression during the first week of Wimbly.
a better example was the speeding-up of Bercy in 2017, that promptly lead to Isner and even Karlovic overperforming.

#### Rago

##### Hall of Fame
Good post.

There's no doubt that beating Nadal on dirt (esp. Philippe Chatrier) is the most daunting task in tennis.

Last edited:

#### Rafa24

##### Hall of Fame
Can you please do Nadal on grass since 2011 too?
And fed on clay in 17, 18.

##### Hall of Fame
drastically? i really doubt it. i think it's rather because the Majors are the main events that most players focus on.

f.i. at the YEC do the favorites get through pretty regularly too. well, it's only top 8 plus RR though.

but there's also the Sunshine double. sometimes players would be out of form after their holidays or tired from Dubai or so, but it rarely looks to me like they are directly upset. especially thanks to Miami right afterwards (and Dubai/Acapulco prior) one can usually notice these things very well.

the other Masters events are either scheduled as preparation for the Majors (to an extent) or during more or less holiday periods for the top players (MC, Canada, (Shanghai)).

and while Bo3 is indeed a bit short (i remember f.i. Isner doing some serious menace), Bo5 tends to favor physicality over technique during the final rounds.
I know that from on odds on perspective, the BO5 format will favour the better player, because statistically it's going to be harder to beat someone better than you for 3 out of 5 sets, as opposed to 2 out of 3 sets. How much we can attribute it to that, I don't know. There's other things we need to consider like what you've mentioned, but you realize that you're only adding arguments to why it drastically favours the better player right? First you say you doubt it 'drastically' favours the better player, but then list reasons why it might, so I'm guessing you're just doubting how much we can actually attribute it to being from an odds on perspective?

Another thing to touch on is you say it favours 'physicality,' when I believe the word you mean is endurance. You make the mistake of assuming that endurance or physicality should be separated from what can be attributed to a 'better' player. Perhaps it's my mistake for using the word 'better' when that word can be entirely subjective though. The definition of 'better' I would be most interested in is the player who can win the most matches against the most opposition.