Dominance Ratio -- What Value do you give to this and other Predictive Analytics?

[RaulRamirez]

I see DR (Dominance Ratio) cited fairly often in threads, and I wanted to start a post about it, and to ask what value this stat has, if any, for you.
Then, I saw a very good, analytical thread on DR that was started 6 years ago, slightly before my time here.
(I'm not a statistician, and only recall about .013% of my one college statistics class from decades ago, so I can't offer a deep numerical drive.)

My questions are more general, and perhaps, more common-sense driven, although there may be some holes in my own understanding:

Do you find that DR has great (or any) value as a predictor - as a predictive analytic?
If it does have value, does it have value to you in analyzing actual results -- whether for a match, a tourney or for a longer duration?
Are there other stats that you prefer as predictive analytics, and do you apply them to actual results?

To generalize - and drawing from mostly team sports like baseball - I see the value of some stats as predictors of future success (which could help scouts and GMs, etc.) but I find them to not be helpful when it comes to analyzing actual results, including comparing/ranking players for various awards. In tennis, my sense is that the predictors aren't as well-developed [compared to baseball] let alone of much use in analyzing actual results.

 

[Neptune]

Maybe similar to H2H, could be very useful in rigging draws?

 

[nolefam_2024]

It matters a bit. Tennis scoring is such that only important points are counted. DR is such criteria. If you don't take extreme examples like Wimbledon 2019 final, mostly the one with higher DR is the winner of the match.

 

[nolefam_2024]

Ex

Djokovic has only won 1 match with point dominance under 0.8. It was his match vs Monfils in 2005 USOpen

While he has lost only 2 match with point dominance over 1.2
Vs rune in Paris 2022 and PCB in USopen 2020.

I don't see DR as one of the stats on uts but if someone finds it you will be able to filter matches won with Dr < 0.9 and lost with Dr > 1.1

It's highly effective way of looking at things that simple set scores and even points dominance. And if you see DR in a rivalry getting closer and closer especially a pattern, you can use it to extrapolate what may happen in a match which is always a possibility. Not certainity.

 

[Rafa4LifeEver]

It matters a bit. Tennis scoring is such that only important points are counted. DR is such criteria. If you don't take extreme examples like Wimbledon 2019 final, mostly the one with higher DR is the winner of the match.

Djokovic lost 2 matches to the same player in the last 3 weeks despite having a higher DR. It rarely happens with him though.

 

[nolefam_2024]

Djokovic lost 2 matches to the same player in the last 3 weeks despite having a higher DR. It rarely happens with him though.

Yes. I think with age it will happen more. As Djokovic couldn't capitalize on some very big pts (3 mp in a row)

 

[Rafa4LifeEver]

Yes. I think with age it will happen more. As Djokovic couldn't capitalize on some very big pts (3 mp in a row)

It won't happen in a slam match or in any other big match. He won in straights where he needed to, Wimbledon SF & ATP Finals F.

 

[RaulRamirez]

...was hoping to get more (not player-specific) replies.

 

[NatF]

IIRC D/R favours servers and its susceptible to being slanted by results versus lower ranked players e.g. high dominance in early rounds will push D/R which can flatter to deceive.

For me it's an interesting stat that should be evaluated along with a host of other things.

 

[nolefam_2024]

IIRC D/R favours servers and its susceptible to being slanted by results versus lower ranked players e.g. high dominance in early rounds will push D/R which can flatter to deceive.

For me it's an interesting stat that should be evaluated along with a host of other things.

Yeah it's possible.

Tsitsipas DR is at 1.15, above Rublev who is at 1.14.

I am not sure who is better between these two though I know tsitsipas return is much poorer than Rublev.

 

[NatF]

As an addendum I would say if a player loses with a clearly higher D/R than his opponent it was probably a pretty unclutch performance. The poster match for this is the 2004 USO QF where Roddick lost to Johansson despite a massive edge in D/R.

 

[RaulRamirez]

Yeah it's possible.

Tsitsipas DR is at 1.15, above Rublev who is at 1.14.

I am not sure who is better between these two though I know tsitsipas return is much poorer than Rublev.

I still regard Tsit as the better player, though of late?
And I also think, to generalize, that DR rewards more dominant serving.

But I question, given the nature of tennis - and not all points being of equal importance - anything that breaks down at the percentage of points won - whether by pure percentage or DR.

 

[RaulRamirez]

As an addendum I would say if a player loses with a clearly higher D/R than his opponent it was probably a pretty unclutch performance. The poster match for this is the 2004 USO QF where Roddick lost to Johansson despite a massive edge in D/R.

I don't recall that match, per se, but I would certainly think that, generally, there is a very high correlation between a better percentage of points won, games won and DR with winning. So, I don't dismiss it.

I just don't think that any of these stats is so important that a fan or pundit should assume that those with the better DR (or points won, or even games won) deserved to win a given match.

Winning a service game at love or at deuce has the same value (although, the second way may not augur as well.)

 

[Kralingen]

I find it most useful over large swathes of time for determining who the best, most anti-fragile players in the world are. If you are better at tennis, you usually win by more, ergo, DR can capture how much you’re winning by, and it can also serve as an “underlying statistic” to predict a player set to break out.

Pretty simple basic logic there.

It also can be interesting on a one-match basis, to highlight clutch or unclutch performances.

However, I find it somewhat inconvenient to use in head to head comparisons between players, as it does not control a) for opponent quality, b) for surface (and average service/return points won on a surface), and c) does not make any note of the context of a match.

In DR world every point in every game vs every opponent in every round is exactly equal. This is its main limitation.

 

[ghostofMecir]

I see DR (Dominance Ratio) cited fairly often in threads, and I wanted to start a post about it, and to ask what value this stat has, if any, for you.
Then, I saw a very good, analytical thread on DR that was started 6 years ago, slightly before my time here.
(I'm not a statistician, and only recall about .013% of my one college statistics class from decades ago, so I can't offer a deep numerical drive.)

My questions are more general, and perhaps, more common-sense driven, although there may be some holes in my own understanding:

Do you find that DR has great (or any) value as a predictor - as a predictive analytic?
If it does have value, does it have value to you in analyzing actual results -- whether for a match, a tourney or for a longer duration?
Are there other stats that you prefer as predictive analytics, and do you apply them to actual results?

To generalize - and drawing from mostly team sports like baseball - I see the value of some stats as predictors of future success (which could help scouts and GMs, etc.) but I find them to not be helpful when it comes to analyzing actual results, including comparing/ranking players for various awards. In tennis, my sense is that the predictors aren't as well-developed [compared to baseball] let alone of much use in analyzing actual results.

I don’t care about it as a predictor, but it is good as an evaluative tool outside of extreme circumstances.

 

[Rovesciarete]

The most dominant players according to this ratio are ranked in a predictable pattern. Big servers slightly outperform their total points won. Interestingly we see once again the underperformance in sets won by Sinner and the overperformance by Alcaraz.

Overall it helps to understand single matches better as mentioned above.

Rk	Player	M​	DR​	TPW%​	TB W%​	TB/S​	S W%​	G W%​
1	Novak Djokovic [SRB]	63​	1.36​	54.9%​	79.5%​	22.2%​	81.8%​	59.6%​
4	Jannik Sinner [ITA]	79​	1.27​	53.9%​	58.1%​	15.0%​	72.5%​	58.0%​
2	Carlos Alcaraz [ESP]	77​	1.26​	53.9%​	67.9%​	14.2%​	78.2%​	58.2%​
10	Taylor Fritz [USA]	76​	1.21​	53.0%​	59.5%​	21.5%​	66.7%​	55.0%​
3	Daniil Medvedev [RUS]	84​	1.20​	53.4%​	53.1%​	15.2%​	71.9%​	57.6%​
14	Grigor Dimitrov [BUL]	64​	1.15​	52.3%​	50.0%​	22.1%​	63.8%​	54.5%​
6	Stefanos Tsitsipas [GRE]	73​	1.15​	52.2%​	54.0%​	26.0%​	63.5%​	54.2%​
5	Andrey Rublev [RUS]	81​	1.14​	52.1%​	52.8%​	16.6%​	64.1%​	54.6%​
15	Karen Khachanov [RUS]	52​	1.14​	52.1%​	63.2%​	14.1%​	62.2%​	53.8%​
12	Alex De Minaur [AUS]	72​	1.13​	52.0%​	60.9%​	12.6%​	58.8%​	54.0%​
7	Alexander Zverev [GER]	82​	1.13​	51.7%​	46.2%​	18.6%​	60.5%​	53.5%​
19	Nicolas Jarry [CHI]	63​	1.13​	51.7%​	61.1%​	21.7%​	62.0%​	52.9%​
9	Hubert Hurkacz [POL]	67​	1.12​	51.4%​	53.3%​	30.9%​	59.8%​	52.5%​
39	Roman Safiullin [RUS]	59​	1.11​	51.8%​	59.1%​	14.8%​	59.1%​	53.0%​
16	Frances Tiafoe [USA]	60​	1.11​	51.6%​	57.1%​	23.2%​	61.6%​	53.7%​
13	Tommy Paul [USA]	70​	1.10​	51.6%​	48.6%​	18.7%​	58.3%​	53.4%​
24	Sebastian Korda [USA]	42​	1.08​	51.2%​	54.5%​	29.7%​	59.5%​	52.6%​
11	Casper Ruud [NOR]	59​	1.08​	51.4%​	58.1%​	19.6%​	60.1%​	53.4%​
18	Cameron Norrie [GBR]	61​	1.08​	51.3%​	45.5%​	13.7%​	55.9%​	52.9%​
27	Lorenzo Musetti [ITA]	61​	1.07​	51.1%​	66.7%​	10.1%​	52.3%​	52.5%​
30	Tomas Martin Etcheverry [ARG]	59​	1.06​	51.0%​	38.7%​	19.9%​	53.8%​	51.9%​
8	Holger Rune [DEN]	68​	1.05​	50.8%​	67.6%​	19.0%​	62.6%​	52.4%​

 

[Rovesciarete]

As we mentioned despite his great late run Sinner once again underperformed considerably his total points won while Rune did the opposite, winning far more tiebreaks and sets than he should have. The highest percentage amount of tiebreaks were played quite obviously by Hurkacz while weak servers like Musetti and de Minaur are almost naturally at the bottom.

So in the long run DR and TPW will converge but in comparison DR values big servers slightly more.

 

[Red Rick]

Low predictive value, but great for assessing matches afrer the fact in terms of who was actually the better player most of the match, qnd if a certain outcome was actually an outlier or not

 

[RaulRamirez]

I appreciate the replies, and well, maybe, I'm stubborn, but here are some of my present thoughts:

I'm not anti-analytics (whether predictive or post-match), although I think that they are best utilized for coaches to devise strategy -- numbers relating to court positioning (though sometimes simplistic), patterns of play, etc. If you have astute pundits and analysts, they can break these analytics down further to offer more insight to viewers.

If DR is slightly skewed to bigger/better servers, then it's not even as reliable as percentage of points won, right? Or, maybe TPW % is similarly skewed?
And while players (obviously - and generally speaking) try to win all points, winning tennis matches isn't simply about winning more points or even winning more games.

So, DR strikes me as a better predictor over a larger range of number, but I just don't see it as valuable in analyzing a match or even a tourney.

 

[Rovesciarete]

I appreciate the replies, and well, maybe, I'm stubborn, but here are some of my present thoughts:

I'm not anti-analytics (whether predictive or post-match), although I think that they are best utilized for coaches to devise strategy -- numbers relating to court positioning (though sometimes simplistic), patterns of play, etc. If you have astute pundits and analysts, they can break these analytics down further to offer more insight to viewers.

At the most basic level analytics can help you to come up with some sort of game plan against somebody but maybe more importantly to look at your own game. Sometimes analytics can also show that losing was more about the other guy winning it. Sinner got smashed but it was key for me to see that he understood that the best player of the year performed extremely well under those specific circumstances.

He was very relaxed against Novak in the Davis Cup and did win despite a negative DR turning that around.

If DR is slightly skewed to bigger/better servers, then it's not even as reliable as percentage of points won, right? Or, maybe TPW % is similarly skewed?
And while players (obviously - and generally speaking) try to win all points, winning tennis matches isn't simply about winning more points or even winning more games.

So, DR strikes me as a better predictor over a larger range of number, but I just don't see it as valuable in analyzing a match or even a tourney.

TPW and DR are over a long timeframe, just like ELO, better predictors of future success than almost all. However you need still a tennis brain and eye. Take for example Rune. His dominance ration is poor and he won just a little over half of his points. This should anchor your expectations somewhat. However he can hit big, has good movement and is suffering from poor shot selection and concentration lapses. You can fix that more easily than the lack of big weapons.

This older threat got a more scientific angle and also shows that better servers achiever higher dominance ratios.

P.S: More math with very good insight.

 

[Rovesciarete]

To frame it a bit differently. At quick glance at DR and TPW allows us to show how well overall a player is roughly doing and how much it has to do with his serve and return qualities. If you look at the table bigger servers have a slightly higher DR relative to the TPW. It also helps to track the development of players in a very quick and easy way.

First Djokovic had a far higher, incredible 55.4% TPW but was less dominant. This suggests a big serve+ improvement as the return game is much more sticky. Alcaraz with a DR of only 1.09 despite 51.6% TPW was greatly hampered by his weak serve. FAA is easy to spot as a very serve-dependent player, just like Hurkacz.

Sinner was earning more or less as many sets as expected, supporting my idea that a return to mean in 2024 should result in a higher win percentage...

Rk	Player	M​	DR​	TPW%​	TB W%​	TB/S​	S W%​	G W%​
1	Novak Djokovic [SRB]	62​	1.38​	55.4%​	50.0%​	13.1%​	77.6%​	61.0%​
3	Daniil Medvedev [RUS]	76​	1.33​	54.5%​	52.2%​	11.5%​	76.0%​	59.2%​
7	Alexander Zverev [GER]	73​	1.27​	53.4%​	67.6%​	17.3%​	74.6%​	56.8%​
6	Stefanos Tsitsipas [GRE]	74​	1.24​	53.5%​	55.2%​	14.7%​	68.5%​	57.0%​
11	Casper Ruud [NOR]	74​	1.21​	53.1%​	60.9%​	12.7%​	69.6%​	56.8%​
5	Andrey Rublev [RUS]	75​	1.20​	52.9%​	53.1%​	16.7%​	65.6%​	55.7%​
29	Felix Auger Aliassime [CAN]	61​	1.13​	51.7%​	55.6%​	17.0%​	58.5%​	52.8%​
4	Jannik Sinner [ITA]	71​	1.12​	52.0%​	61.1%​	20.5%​	63.6%​	54.0%​
14	Grigor Dimitrov [BUL]	42​	1.12​	51.9%​	33.3%​	17.0%​	55.7%​	53.2%​
18	Cameron Norrie [GBR]	77​	1.11​	51.9%​	50.0%​	9.3%​	62.2%​	55.0%​
9	Hubert Hurkacz [POL]	59​	1.11​	51.5%​	60.6%​	21.6%​	61.4%​	53.0%​
2	Carlos Alcaraz [ESP]	49​	1.09​	51.6%​	63.0%​	20.3%​	60.2%​	53.4%​
35	Aslan Karatsev [RUS]	54​	1.09​	51.4%​	40.0%​	10.4%​	59.0%​	53.4%​
24	Sebastian Korda [USA]	49​	1.08​	51.5%​	47.6%​	16.5%​	59.1%​	52.9%​
10	Taylor Fritz [USA]	56​	1.04​	50.8%​	57.1%​	23.3%​	56.7%​	51.4%​
12	Alex De Minaur [AUS]	50​	1.02​	50.4%​	47.4%​	15.1%​	50.8%​	50.4%​
8	Holger Rune [DEN]	20​	0.96​	49.2%​	33.3%​	16.4%​	41.8%​	48.0%​
	Average		1.08​	51.2%​	54.2%​	17.2%​	57.7%​	52.5%​

 

[RaulRamirez]

To frame it a bit differently. At quick glance at DR and TPW allows us to show how well overall a player is roughly doing and how much it has to do with his serve and return qualities. If you look at the table bigger servers have a slightly higher DR relative to the TPW. It also helps to track the development of players in a very quick and easy way.

First Djokovic had a far higher, incredible 55.4% TPW but was less dominant. This suggests a big serve+ improvement as the return game is much more sticky. Alcaraz with a DR of only 1.09 despite 51.6% TPW was greatly hampered by his weak serve. FAA is easy to spot as a very serve-dependent player, just like Hurkacz.

Sinner was earning more or less as many sets as expected, supporting my idea that a return to mean in 2024 should result in a higher win percentage...

		M​	DR​	TPW%​	TB W%​	TB/S​	S W%​	G W%​
29	Felix Auger Aliassime [CAN]	61​	1.13​	51.7%​	55.6%​	17.0%​	58.5%​	52.8%​
9	Hubert Hurkacz [POL]	59​	1.11​	51.5%​	60.6%​	21.6%​	61.4%​	53.0%​
8	Holger Rune [DEN]	20​	0.96​	49.2%​	33.3%​	16.4%​	41.8%​	48.0%​
								​

I had just seen that older thread you linked to -- thanks. A good discussion.

When I look at these numbers, I wonder how Rune could make it to #8. I'm also not sure why only 20 matches show for him. But was he clutch (not judging by TBW%), lucky or...? And I look at the two big servers mentioned - FAA and Hubi - and it's hard to see why one is ranked #29 and the other is #9.
I think that the more I look at this stat, to be frank, the less I like it -- and not only based on a couple outliers.

It breaks down points with an index that doesn't tell that much, and even if it was more telling within those parameters, winning the most points, per se, is not the goal of players. I also don't know how DR can drive strategy whereas other data certainly can.
I just read things like "How did Player X lose a match, or tourney, with a DR of 1.3?"
Well lots of ways. Even if he won more points (slightly more telling to me) or more games (slightly more), it doesn't mean that the winning player was lucky and he was unlucky.

It's a bit like break points. "Well, Player X" only converted...I don't know...3 of 22 opportunities."
Okay, but those who post these statements assume that getting break points was because of skill and not cashing them in was due to bad luck.

As a cross-sports thing, in (American) football, if a team outgains the other team in yardage (a good thing generally), they have a good chance of winning the game, but the goal is not to outgain the other team in yards; it's finding ways to win the game. The MLB team I root for leaves lots of runners on base, which is a little like accumulating break points and not cashing them in. Bad luck? No, not really.

Others may see all or some of this differently.

 

[Rovesciarete]

I had just seen that older thread you linked to -- thanks. A good discussion.

When I look at these numbers, I wonder how Rune could make it to #8. I'm also not sure why only 20 matches show for him. But was he clutch (not judging by TBW%), lucky or...? And I look at the two big servers mentioned - FAA and Hubi - and it's hard to see why one is ranked #29 and the other is #9.
I think that the more I look at this stat, to be frank, the less I like it -- and not only based on a couple outliers.

My bad, I simply forgot that the first table was from 2023 and the second from 2021. It allows us to compare the development of some players.

It breaks down points with an index that doesn't tell that much, and even if it was more telling within those parameters, winning the most points, per se, is not the goal of players. I also don't know how DR can drive strategy whereas other data certainly can.
I just read things like "How did Player X lose a match, or tourney, with a DR of 1.3?"
Well lots of ways. Even if he won more points (slightly more telling to me) or more games (slightly more), it doesn't mean that the winning player was lucky and he was unlucky.

It's a bit like break points. "Well, Player X" only converted...I don't know...3 of 22 opportunities."
Okay, but those who post these statements assume that getting break points was because of skill and not cashing them in was due to bad luck.

Luck is always present but IIRC elite players outperform normal ATP players in big points. However maybe much less so than many fans seem to think. Sinner played this season eleven matches on clay which is a bit small as a sample. Nobody will say that he had a great clay season but look at the table.

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
2	Carlos Alcaraz [ESP]	28​	1.31​	4131​	55.0%​	6​	3-3​	50.0%​	8.6%​	70​	57-13​	81.4%​	648​	397-251​	61.3%​	1:26​	34.5​	35.1​
4	Jannik Sinner [ITA]	11​	1.29​	1815​	54.7%​	6​	3-3​	50.0%​	19.4%​	31​	21-10​	67.7%​	281​	167-114​	59.4%​	1:32​	32.7​	33.5​
1	Novak Djokovic [SRB]	15​	1.29​	2727​	54.5%​	10​	8-2​	80.0%​	23.3%​	43​	33-10​	76.7%​	406​	237-169​	58.4%​	1:11​	24.8​	23.5​
10	Taylor Fritz [USA]	17​	1.26​	2673​	53.6%​	7​	3-4​	42.9%​	15.6%​	45​	28-17​	62.2%​	427​	237-190​	55.5%​	1:27​	32.9​	33.2​
6	Stefanos Tsitsipas [GRE]	22​	1.25​	3129​	53.5%​	6​	4-2​	66.7%​	11.3%​	53​	38-15​	71.7%​	509​	288-221​	56.6%​	1:15​	31.2​	31.7​

He has won more total points than Nole and had the same DR. In this case he played a lot of good matches against weaker opposition than Nole faced. Alcaraz was simply incredible this season despite not winning in Rome or RG. Both Jannik and Charlie underperformed a lot in the tiebreaks and Sinner even more in sets won. Fritz had some great runs but didn't get big wins so we don't remember it as much.

DC helps but isn't enough alone.

 

[TripleATeam]

I see DR (Dominance Ratio) cited fairly often in threads, and I wanted to start a post about it, and to ask what value this stat has, if any, for you.
Then, I saw a very good, analytical thread on DR that was started 6 years ago, slightly before my time here.
(I'm not a statistician, and only recall about .013% of my one college statistics class from decades ago, so I can't offer a deep numerical drive.)

My questions are more general, and perhaps, more common-sense driven, although there may be some holes in my own understanding:

Do you find that DR has great (or any) value as a predictor - as a predictive analytic?
If it does have value, does it have value to you in analyzing actual results -- whether for a match, a tourney or for a longer duration?
Are there other stats that you prefer as predictive analytics, and do you apply them to actual results?

To generalize - and drawing from mostly team sports like baseball - I see the value of some stats as predictors of future success (which could help scouts and GMs, etc.) but I find them to not be helpful when it comes to analyzing actual results, including comparing/ranking players for various awards. In tennis, my sense is that the predictors aren't as well-developed [compared to baseball] let alone of much use in analyzing actual results.

Its main use is as a reflective stat, not a predictive one. DR shows part of a "quality of game", but that is deceptive when we get into easy win territory.

Nothing indicates a player that wins 6-0, 6-1 over their opponent is the favorite against a player that went 6-3, 6-2 in the prior round. It may tend to be correct more often than not, but the main reason is that any gap leads to a victory in tennis. There's little reason to push yourself to get the 6-0 if you've already almost secured the 6-2 victory.

When we look at a match we can more objectively say if a stronger player lost, or how close the match was by using DR. But as a predictor its value is lost since they are two completely separate comparisons with so many factors influencing the match.

A long running average of DR may provide a decent suggestion of a winner, as say Fritz might be in the QF with a DR of 0.9 average and Djokovic could meet him with 2.3. Djokovic likely winner.

But when it's close, say a final where Alcaraz comes in with 1.7 and Djokovic with 1.6, there's little reason to put any stock into the fact that Alcaraz is slightly ahead in DR.

 

[Rovesciarete]

I think we have to distinguish between DR's analytical value for a match and its predictive quality for the long run. The dominance ratios of Djokovic against Sinner in the last week are largely the result of two excellent serve performances, one of them bordering on the realm of absurdity with all the line-hitting.

DJOKOVIC Recent Results All results | Top​

Mouse over column headers (on all tables) for stat definitions.

Tournament​	Surface​	Rd​	Rk​	vRk​	​	Score​	DR​	A%​	DF%​	1stIn​	1st%​	2nd%​	BPSvd​	Time​
Tour Finals	Hard	F	1​	4​	(1)Djokovic d. (4)Jannik Sinner [ITA]	6-3 6-3	2.64​	28.3%​	0.0%​	69.6%​	90.6%​	64.3%​	2/2​	1:43​
Tour Finals	Hard	SF	1​	2​	(1)Djokovic d. (2)Carlos Alcaraz [ESP]	6-3 6-2	1.57​	3.6%​	1.8%​	67.3%​	81.1%​	55.6%​	4/4​	1:28​
Tour Finals	Hard	RR	1​	4​	(4)Jannik Sinner [ITA] d. (1)Djokovic	7-5 6-7(5) 7-6(2)	1.13​	20.0%​	1.0%​	63.0%​	81.0%​	59.5%​	1/3​	3:09​
Tour Finals	Hard	RR	1​	9​	(1)Djokovic d. (9)Hubert Hurkacz [POL]	7-6(1) 4-6 6-1	1.43​	7.4%​	1.2%​	63.0%​	74.5%​	73.3%​	0/1​	2:05​
Tour Finals	Hard	RR	1​	8​	(1)Djokovic d. (8)Holger Rune [DEN]	7-6(4) 6-7(1) 6-3	1.07​	9.3%​	3.7%​	68.5%​	74.3%​	41.2%​	3/6​	3:04​

Sinners suffered a rough serve patch in the final. Interestingly both were quite aggressive with their first serves as both are at the very top of second serve win percentage.


I do think that Sinner will do better in slams than in masters in years to come as the longer format increases the predictive value of TPW and DR...

SINNER Recent Results All results | Top​

Mouse over column headers (on all tables) for stat definitions.

Tournament​	Surface​	Rd​	Rk​	vRk​	​	Score​	DR​	A%​	DF%​	1stIn​	1st%​	2nd%​	BPSvd​	Time​
Tour Finals	Hard	F	4​	1​	(1)Novak Djokovic [SRB] d. (4)Sinner	6-3 6-3	0.38​	10.8%​	1.4%​	66.2%​	57.1%​	48.0%​	5/8​	1:43​
Tour Finals	Hard	SF	4​	3​	(4)Sinner d. (3)Daniil Medvedev [RUS]	6-3 6-7(4) 6-1	1.15​	9.6%​	1.9%​	57.7%​	83.3%​	47.7%​	2/2​	2:29​
Tour Finals	Hard	RR	4​	1​	(4)Sinner d. (1)Novak Djokovic [SRB]	7-5 6-7(5) 7-6(2)	0.89​	12.7%​	1.7%​	60.2%​	78.9%​	55.3%​	2/3​	3:09​
Tour Finals	Hard	RR	4​	6​	(4)Sinner d. Stefanos Tsitsipas [GRE]	6-4 6-4	1.76​	17.6%​	3.9%​	70.6%​	88.9%​	60.0%​	0/0​	1:25​
Tour Finals	Hard	RR	4​	8​	(4)Sinner d. (8)Holger Rune [DEN]	6-2 5-7 6-4	1.51​	13.1%​	0.0%​	58.3%​	83.7%​	60.0%​	4/5​	2:3​

 

[dking68]

It won't happen in a slam match or in any other big match. He won in straights where he needed to, Wimbledon SF & ATP Finals F.

His DR over Sinner in his 3 set win at Wimbledon this year was 1.03. Essentially Sinner played at the same level as Nole but lost because he couldn’t convert the break piints

 

[Rovesciarete]

His DR over Sinner in his 3 set win at Wimbledon this year was 1.03. Essentially Sinner played at the same level as Nole but lost because he couldn’t convert the break piints

He was closer in a sense than the year before. Tennis can not be broken down into a single all-mighty stat but DR or TPW are good predictors for long-time success. Still look at the guys winning the highest percentage of points at the slams in the last two years. After a 2021 just below a brutally strong Medvedev Djokovic dominates the tour stat-wise for two consecutive years.

Alcaraz has a slight lead in dominance ration while Sinner won more points but also faced lower ranked players.

2023

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	28​	1.42​	5707​	56.1%​	19​	17-2​	89.5%​	20.4%​	93​	83-10​	89.2%​	873​	547-326​	62.7%​	2:00​	36.2​	35.4​
2	Carlos Alcaraz [ESP]	19​	1.34​	3791​	54.8%​	7​	5-2​	71.4%​	10.9%​	64​	52-12​	81.3%​	583​	350-233​	60.0%​	1:50​	32.6​	33.0​
4	Jannik Sinner [ITA]	16​	1.34​	3399​	54.6%​	5​	2-3​	40.0%​	8.5%​	59​	42-17​	71.2%​	534​	320-214​	59.9%​	2:18​	37.5​	39.0​
3	Daniil Medvedev [RUS]	17​	1.20​	3547​	53.5%​	10​	5-5​	50.0%​	17.2%​	58​	40-18​	69.0%​	543​	315-228​	58.0%​	2:21​	41.3​	40.5​

2022

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	12​	1.45​	2395​	56.0%​	3​	2-1​	66.7%​	7.0%​	43​	34-9​	79.1%​	382​	235-147​	61.5%​	1:52​	31.2​	33.6​
3	Daniil Medvedev [RUS]	15​	1.27​	3290​	53.9%​	9​	6-3​	66.7%​	17.0%​	53​	39-14​	73.6%​	508​	292-216​	57.5%​	2:08​	36.1​	34.9​
4	Jannik Sinner [ITA]	19​	1.22​	4247​	53.5%​	7​	6-1​	85.7%​	9.9%​	71​	50-21​	70.4%​	665​	378-287​	56.8%​	2:20​	37.5​	37.7​
2	Carlos Alcaraz [ESP]	19​	1.24​	4539​	53.2%​	13​	5-8​	38.5%​	18.1%​	72​	52-20​	72.2%​	701​	408-293​	58.2%​	2:23​	37.8​	35.9​

Lot of guys on the forum will say that Sinner hasn't the legs to compete when the game goes long - and they were right. However they fail to make the next mental step. He won a similar percentage of points as Alcaraz and below only the great Nole while suffering from endurance and at least a change of serve. What happens if he naturally continues to improve both the body and the serve?

 

[nolefam_2024]

He was closer in a sense than the year before. Tennis can not be broken down into a single all-mighty stat but DR or TPW are good predictors for long-time success. Still look at the guys winning the highest percentage of points at the slams in the last two years. After a 2021 just below a brutally strong Medvedev Djokovic dominates the tour stat-wise for two consecutive years.

Alcaraz has a slight lead in dominance ration while Sinner won more points but also faced lower ranked players.

2023

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	28​	1.42​	5707​	56.1%​	19​	17-2​	89.5%​	20.4%​	93​	83-10​	89.2%​	873​	547-326​	62.7%​	2:00​	36.2​	35.4​
2	Carlos Alcaraz [ESP]	19​	1.34​	3791​	54.8%​	7​	5-2​	71.4%​	10.9%​	64​	52-12​	81.3%​	583​	350-233​	60.0%​	1:50​	32.6​	33.0​
4	Jannik Sinner [ITA]	16​	1.34​	3399​	54.6%​	5​	2-3​	40.0%​	8.5%​	59​	42-17​	71.2%​	534​	320-214​	59.9%​	2:18​	37.5​	39.0​
3	Daniil Medvedev [RUS]	17​	1.20​	3547​	53.5%​	10​	5-5​	50.0%​	17.2%​	58​	40-18​	69.0%​	543​	315-228​	58.0%​	2:21​	41.3​	40.5​

2022

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	12​	1.45​	2395​	56.0%​	3​	2-1​	66.7%​	7.0%​	43​	34-9​	79.1%​	382​	235-147​	61.5%​	1:52​	31.2​	33.6​
3	Daniil Medvedev [RUS]	15​	1.27​	3290​	53.9%​	9​	6-3​	66.7%​	17.0%​	53​	39-14​	73.6%​	508​	292-216​	57.5%​	2:08​	36.1​	34.9​
4	Jannik Sinner [ITA]	19​	1.22​	4247​	53.5%​	7​	6-1​	85.7%​	9.9%​	71​	50-21​	70.4%​	665​	378-287​	56.8%​	2:20​	37.5​	37.7​
2	Carlos Alcaraz [ESP]	19​	1.24​	4539​	53.2%​	13​	5-8​	38.5%​	18.1%​	72​	52-20​	72.2%​	701​	408-293​	58.2%​	2:23​	37.8​	35.9​

Lot of guys on the forum will say that Sinner hasn't the legs to compete when the game goes long - and they were right. However they fail to make the next mental step. He won a similar percentage of points as Alcaraz and below only the great Nole while suffering from endurance and at least a change of serve. What happens if he naturally continues to improve both the body and the serve?

Sinner doesn't have the legs yet. But he might next year. Most likely. I think he is close to winning a slam but we will only know when it happens.

 

[Rovesciarete]

Sinner doesn't have the legs yet. But he might next year. Most likely. I think he is close to winning a slam but we will only know when it happens.

We will see, I agree. My simplistic take is quite important to anchor the expectations. Still I would put quite a bit of value that he has improved a lot in specific matchups, something DR struggles to show. Maybe DR versus the top 20 is better at that and in this case Sinner struggled mightily in 2022, far less so in 2023.

 

[Rovesciarete]

So this is sorted after DR against the top 20:

2023

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	30​	1.25​	5137​	53.7%​	19​	15-4​	78.9%​	22.4%​	85​	65-20​	76.5%​	808​	458-350​	56.7%​	2:01​	42.6​	42.3​
2	Carlos Alcaraz [ESP]	34​	1.16​	5599​	52.2%​	15​	10-5​	66.7%​	16.9%​	89​	62-27​	69.7%​	864​	467-397​	54.1%​	1:56​	44.3​	42.2​
3	Daniil Medvedev [RUS]	39​	1.10​	6038​	51.5%​	18​	11-7​	61.1%​	18.9%​	95​	61-34​	64.2%​	944​	506-438​	53.6%​	1:60​	49.1​	46.4​
4	Jannik Sinner [ITA]	29​	1.10​	4697​	51.4%​	14​	8-6​	57.1%​	18.2%​	77​	47-30​	61.0%​	718​	378-340​	52.6%​	2:08​	48.4​	47.:​

2022

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	24​	1.28​	3511​	53.7%​	14​	10-4​	71.4%​	23.0%​	61​	44-17​	72.1%​	557​	307-250​	55.1%​	1:52​	44.2​	46.1​
2	Carlos Alcaraz [ESP]	27​	1.11​	4730​	51.5%​	17​	9-8​	52.9%​	22.1%​	77​	47-30​	61.0%​	736​	392-344​	53.3%​	2:14​	47.0​	45.9​
10	Taylor Fritz [USA]	17​	1.09​	3037​	50.7%​	15​	6-9​	40.0%​	30.0%​	50​	26-24​	52.0%​	479​	249-230​	52.0%​	2:11​	44.4​	43.9​
11	Casper Ruud [NOR]	23​	1.08​	3615​	50.9%​	11​	7-4​	63.6%​	18.6%​	59​	33-26​	55.9%​	585​	303-282​	51.8%​	1:58​	45.8​	44.9​
29	Felix Auger Aliassime [CAN]	27​	1.06​	3865​	50.7%​	16​	7-9​	43.8%​	23.9%​	67​	35-32​	52.2%​	606​	306-300​	50.5%​	1:55​	46.3​	48.2​
	10+ players	​	​	​	​	​	​	​	​	​	​	​	​	​	​	​	​	​
4	Jannik Sinner [ITA]	14​	0.91​	2636​	48.8%​	7​	2-5​	28.6%​	16.3%​	43​	16-27​	37.2%​	407​	193-214​	47.4%​	2:29​	48.4​	47.3​

This explains to some extent why Nole and Charlie did what Janni could not. They were stronger and knew how to play against the best while he did not.

Taylor Fritz is always ranked highly thanks to his good form and excellent serve. Might have to look into it more.

 

[nolefam_2024]

So this is sorted after DR against the top 20:

2023

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	30​	1.25​	5137​	53.7%​	19​	15-4​	78.9%​	22.4%​	85​	65-20​	76.5%​	808​	458-350​	56.7%​	2:01​	42.6​	42.3​
2	Carlos Alcaraz [ESP]	34​	1.16​	5599​	52.2%​	15​	10-5​	66.7%​	16.9%​	89​	62-27​	69.7%​	864​	467-397​	54.1%​	1:56​	44.3​	42.2​
3	Daniil Medvedev [RUS]	39​	1.10​	6038​	51.5%​	18​	11-7​	61.1%​	18.9%​	95​	61-34​	64.2%​	944​	506-438​	53.6%​	1:60​	49.1​	46.4​
4	Jannik Sinner [ITA]	29​	1.10​	4697​	51.4%​	14​	8-6​	57.1%​	18.2%​	77​	47-30​	61.0%​	718​	378-340​	52.6%​	2:08​	48.4​	47.:​

2022

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	24​	1.28​	3511​	53.7%​	14​	10-4​	71.4%​	23.0%​	61​	44-17​	72.1%​	557​	307-250​	55.1%​	1:52​	44.2​	46.1​
2	Carlos Alcaraz [ESP]	27​	1.11​	4730​	51.5%​	17​	9-8​	52.9%​	22.1%​	77​	47-30​	61.0%​	736​	392-344​	53.3%​	2:14​	47.0​	45.9​
10	Taylor Fritz [USA]	17​	1.09​	3037​	50.7%​	15​	6-9​	40.0%​	30.0%​	50​	26-24​	52.0%​	479​	249-230​	52.0%​	2:11​	44.4​	43.9​
11	Casper Ruud [NOR]	23​	1.08​	3615​	50.9%​	11​	7-4​	63.6%​	18.6%​	59​	33-26​	55.9%​	585​	303-282​	51.8%​	1:58​	45.8​	44.9​
29	Felix Auger Aliassime [CAN]	27​	1.06​	3865​	50.7%​	16​	7-9​	43.8%​	23.9%​	67​	35-32​	52.2%​	606​	306-300​	50.5%​	1:55​	46.3​	48.2​
	10+ players	​	​	​	​	​	​	​	​	​	​	​	​	​	​	​	​	​
4	Jannik Sinner [ITA]	14​	0.91​	2636​	48.8%​	7​	2-5​	28.6%​	16.3%​	43​	16-27​	37.2%​	407​	193-214​	47.4%​	2:29​	48.4​	47.3​

This explains to some extent why Nole and Charlie did what Janni could not. They were stronger and knew how to play against the best while he did not.

Taylor Fritz is always ranked highly thanks to his good form and excellent serve. Might have to look into it more.

And also who the REAL number 1 was last year.

 

[Rovesciarete]

And also who the REAL number 1 was last year.

There is no doubt about that as much as I admire Alcaraz. This also goes for 2021, here the DR against the top 20. I'm starting to like that specific selection as it is broad enough to give us a decent sample size but selects higher quality opponents. The consistency of Nole is outstanding, winning between 53.5 and 53.7% of total points and having ad DR between 1.26 and 1.28 over three years!

Daniil was doing very well that year, in part because he had to face less Novak and Alcaraz was not yet there. Once again we see Janni suffering much more against the top players than others. FAA had two strong years against the competition, similar to Stefanos.

2021

Rk	Player	M​	DR​	Points​	TPW%​	TBs​	TB W-L​	TB W%​	TB/S​	Sets​	Set W-L​	S W%​	Gms​	Game W-L​	G W%​	Time/Mt​	Min/Set​	Sec/Pt​
1	Novak Djokovic [SRB]	24​	1.26​	4545​	53.5%​	13​	7-6​	53.8%​	16.7%​	78​	53-25​	67.9%​	709​	407-302​	57.4%​	2:27​	45.1​	46.4​
3	Daniil Medvedev [RUS]	26​	1.15​	3945​	52.2%​	11​	5-6​	45.5%​	15.9%​	69​	44-25​	63.8%​	628​	345-283​	54.9%​	1:47​	40.3​	42.3​
14	Grigor Dimitrov [BUL]	11​	1.11​	1536​	51.5%​	4​	2-2​	50.0%​	14.8%​	27​	15-12​	55.6%​	248​	128-120​	51.6%​	1:46​	43.1​	45.4​
7	Alexander Zverev [GER]	27​	1.10​	4521​	51.1%​	15​	9-6​	60.0%​	20.0%​	75​	45-30​	60.0%​	708​	365-343​	51.6%​	2:07​	45.6​	45.4​
6	Stefanos Tsitsipas [GRE]	20​	1.05​	3583​	51.3%​	8​	5-3​	62.5%​	14.0%​	57​	33-24​	57.9%​	572​	303-269​	53.0%​	2:18​	48.6​	46.4​
29	Felix Auger Aliassime [CAN]	15​	1.05​	2424​	50.2%​	2​	2-0​	100.0%​	4.8%​	42​	22-20​	52.4%​	412​	204-208​	49.5%​	2:04​	44.1​	45.9​
5	Andrey Rublev [RUS]	15​	1.01​	2130​	50.1%​	4​	2-2​	50.0%​	11.4%​	35​	17-18​	48.6%​	331​	165-166​	49.8%​	1:43​	44.1​	43.4​
35	Aslan Karatsev [RUS]	14​	1.01​	2300​	50.2%​	2​	1-1​	50.0%​	5.0%​	40​	21-19​	52.5%​	368​	186-182​	50.5%​	2:06​	44.0​	45.9​
4	Jannik Sinner [ITA]	18​	0.98​	3056​	49.9%​	8​	6-2​	75.0%​	16.7%​	48​	24-24​	50.0%​	470​	232-238​	49.4%​	2:08​	47.8​

 

[Rovesciarete]

I often try to come in from different angles to understand a subject. A fine post by Jeff Sackmann about Andy Roddick helps to think about dominance ratio. Look at those numbers, lots of 1.25 plus, similar to the current form of Alcaraz or Sinner. When did he have the highest success?

RODDICK Tour-Level Seasons Top​

Mouse over column headers (on all tables) for stat definitions. Click on years for results from that season.

Year​	Win%​	Set%​	Game%​	TB%​	Hld%​	Brk%​	A%​	DF%​	1stIn​	1st%​	2nd%​	SPW​	RPW​	TPW​	DR​	Best​
2012	59.0%​	56.6%​	51.0%​	54.5%​	84.2%​	17.6%​	11.7%​	2.4%​	66.2%​	74.8%​	53.1%​	67.5%​	34.4%​	50.9%​	1.06​	W (2x)
2011	68.0%​	63.1%​	54.1%​	57.1%​	86.6%​	21.5%​	16.7%​	2.8%​	65.2%​	76.8%​	55.7%​	69.4%​	36.3%​	52.4%​	1.19​	W (Memphis)
2010	72.7%​	67.8%​	54.9%​	57.5%​	91.1%​	17.6%​	16.0%​	1.9%​	68.5%​	79.2%​	56.9%​	72.2%​	34.9%​	52.9%​	1.25​	W (2x)
2009	76.6%​	73.1%​	56.2%​	73.3%​	90.8%​	19.3%​	15.7%​	1.8%​	70.1%​	79.1%​	57.2%​	72.5%​	35.5%​	53.4%​	1.29​	W (Memphis)
2008	73.5%​	68.0%​	55.6%​	57.8%​	91.2%​	19.2%​	19.0%​	2.8%​	67.6%​	79.6%​	55.6%​	71.8%​	35.6%​	53.1%​	1.27​	W (3x)
2007	77.1%​	73.1%​	55.9%​	74.5%​	90.9%​	17.8%​	16.0%​	2.1%​	65.0%​	80.4%​	56.7%​	72.1%​	34.0%​	52.8%​	1.22​	W (2x)
2006	70.0%​	67.9%​	56.2%​	50.0%​	90.5%​	22.4%​	15.0%​	1.9%​	67.0%​	78.8%​	54.9%​	70.9%​	37.2%​	53.6%​	1.28​	W (Cincinnati Masters)
2005	79.7%​	72.4%​	56.8%​	54.7%​	92.5%​	20.8%​	17.0%​	2.5%​	65.8%​	81.5%​	55.8%​	72.7%​	36.0%​	53.7%​	1.32​	W (5x)
2004	80.4%​	76.4%​	57.1%​	72.5%​	90.9%​	22.5%​	17.1%​	3.1%​	63.0%​	80.5%​	55.9%​	71.4%​	37.5%​	53.7%​	1.31​	W (4x)
2003	79.1%​	72.5%​	56.5%​	62.5%​	91.5%​	20.9%​	14.5%​	2.8%​	61.8%​	80.1%​	57.8%​	71.5%​	36.4%​	53.6%​	1.28​	W (6x)
2002	71.8%​	67.2%​	55.0%​	55.9%​	87.6%​	23.0%​	13.1%​	3.7%​	61.1%​	78.6%​	54.1%​	69.1%​	36.9%​	52.7%​	1.19​	W (2x)
2001	71.4%​	70.4%​	55.4%​	65.6%​	90.4%​	19.7%​	13.2%​	4.3%​	59.1%​	79.1%​	56.5%​	69.9%​	36.3%​	52.7%​	1.21​	W (3x)
2000	44.4%​	42.9%​	48.5%​	50.0%​	83.0%​	14.0%​	12.8%​	3.8%​	55.9%​	77.1%​	51.3%​	65.7%​	31.2%​	48.5%​	0.91​	QF (Washington)
Career	74.0%​	69.6%​	55.6%​	62.1%​	90.1%​	20.3%​	15.5%​	2.7%​	64.8%​	79.3%​	55.9%​	71.1%​	36.0%​	53.0%​	1.24 ​

Surprise! Right before he achieved his first DR of 1.30. Why?


In short he was a bit lucky, then Federer happened and he focused more and certain favourable venues...

Fritz seems to be rather similar to Andy in some regard, very high DR for his point haul.