Net Approaches in Pro Tennis: A New Analysis

[johnyandell]

Let's get your thought son Jeremy Rosen's first article, "Net Approaches in Pro Tennis: A New Analysis"

[doctorhl]

The article demonstrates how hard it is to define success in sport performance because of confounding variables. Confounding variables complicate “ cookie cutter” templates, whether used for physical or mental performance

[jeremyrosen]

The article demonstrates how hard it is to define success in sport performance because of confounding variables. Confounding variables complicate “ cookie cutter” templates, whether used for physical or mental performance

Yup! Templates have an important purpose for understanding the game, but it's essential to put them in context.

[jimlosaltos]

Perhaps Rosen would consider Craig's misleading analysis similar to Simpson's Paradox. Where showing a high result by one measure can distort another. For example: Federer loses a lot of 5-set matches, a higher percentage than one would expect for his overall success. But, that can be explained because he rarely loses big matches before a deciding set. So, it's actually a sign of strength.

https://en.wikipedia.org/wiki/Simpson%27s_paradox

Wiki "Simpson's paradox, which goes by several names, is a phenomenon in probability and statistics, in which a trend appears in several different groups of data but disappears or reverses when these groups are combined. This result is often encountered in social-science and medical-science statistics^[1][2][3] and is particularly problematic when frequency data is unduly given causal interpretations" Batting averages[edit]

A common example of Simpson's paradox involves the batting averages of players in professional baseball. It is possible for one player to have a higher batting average than another player each year for a number of years, but to have a lower batting average across all of those years. This phenomenon can occur when there are large differences in the number of at bats between the years. Mathematician Ken Ross^[18] demonstrated this using the batting average of two baseball players, Derek Jeter and David Justice, during the years 1995 and 1996:^[19]

Derek Jeter	12/48	.250	183/582	.314	195/630	.310
David Justice	104/411	.253	45/140	.321	149/551	.270

In both 1995 and 1996, Justice had a higher batting average (in bold type) than Jeter did. However, when the two baseball seasons are combined, Jeter shows a higher batting average than Justice. According to Ross, this phenomenon would be observed about once per year among the possible pairs of players.

[jeremyrosen]

Perhaps Rosen would consider Craig's misleading analysis similar to Simpson's Paradox. Where showing a high result by one measure can distort another. For example: Federer loses a lot of 5-set matches, a higher percentage than one would expect for his overall success. But, that can be explained because he rarely loses big matches before a deciding set. So, it's actually a sign of strength.

https://en.wikipedia.org/wiki/Simpson%27s_paradox

Wiki "Simpson's paradox, which goes by several names, is a phenomenon in probability and statistics, in which a trend appears in several different groups of data but disappears or reverses when these groups are combined. This result is often encountered in social-science and medical-science statistics^[1][2][3] and is particularly problematic when frequency data is unduly given causal interpretations" Batting averages[edit]


A common example of Simpson's paradox involves the batting averages of players in professional baseball. It is possible for one player to have a higher batting average than another player each year for a number of years, but to have a lower batting average across all of those years. This phenomenon can occur when there are large differences in the number of at bats between the years. Mathematician Ken Ross^[18] demonstrated this using the batting average of two baseball players, Derek Jeter and David Justice, during the years 1995 and 1996:^[19]

Derek Jeter	12/48	.250	183/582	.314	195/630	.310
David Justice	104/411	.253	45/140	.321	149/551	.270

In both 1995 and 1996, Justice had a higher batting average (in bold type) than Jeter did. However, when the two baseball seasons are combined, Jeter shows a higher batting average than Justice. According to Ross, this phenomenon would be observed about once per year among the possible pairs of players.

Yes, Craig's analysis can at times be an example of Simpson's paradox. Specifically when there's a situation in which you're less likely to win when approaching than when staying back both on confounders and not on confounders, yet you're more likely to win when approaching than staying back overall. In the specific match in my article, that wasn't the case, but odds are there are matches out there where that is the case.

[johnyandell]

I learn something every day on Tennisplayer

[jimlosaltos]

Yes, Craig's analysis can at times be an example of Simpson's paradox. Specifically when there's a situation in which you're less likely to win when approaching than when staying back both on confounders and not on confounders, yet you're more likely to win when approaching than staying back overall. In the specific match in my article, that wasn't the case, but odds are there are matches out there where that is the case.

Ah, found the articles about Fed and Simpson's Paradox, but it was about losing matches where the winner won less than 50% of total points. At that time in 2014 Fed had the worst such record, winning 4 vs 24 loses when winning more points.

Here is the original paper by RYAN RODENBERG is an assistant professor of sports law at Florida State University. and Jeff Sackman et al, of Tennis Abstract:

https://www.ingentaconnect.com/content/uwic/ujpa/2013/00000013/00000003/art00019

And the more reader-friendly Atlantic article:

https://www.theatlantic.com/entertainment/archive/2014/01/what-every-pro-tennis-player-does-better-than-roger-federer/283007/

Excerpt from ATM:

At the other end of the Simpson’s Paradox spectrum was, of course, Roger Federer. In completed matches, he was 4-24 in contests where the winner prevailed on less than 50 percent of the total points. Federer’s winning percentage in these matches (14.29 percent) was the worst among all 72 players in the sample who participated in at least 20 matches of this type during their careers. This result surprised us, as it differed wildly from other players who had similarly won multiple Grand Slam singles titles. Andre Agassi, Rafael Nadal, Pete Sampras, Sergi Bruguera, Marat Safin, Lleyton Hewitt, Yevgeny Kafelnikov, and Gustavo Kuerten were all .500 or better in Simpson’s Paradox matches. Jim Courier was the only player worse than 50-50 in such matches, with a non-alarming 11-15 record.

[arturohernandez]

Yup! Templates have an important purpose for understanding the game, but it's essential to put them in context.

Thank you for this article!!!!

Craig wrote that the net yields a 70% advantage today and yesterday. The problem is that people hardly come in these days. I wondered if this was not due to people being passed more often.

70% of 10 times a match does not equal 70% of 60 times a match.

If players are coming in less, then they may be simply reacting to what is happening on the court.

Regressions are nice but even better would be an experiment. Ask players to practice certain styles and then calculate win percentages.

What happens if SABR was used a lot more?
What happens if someone actually serves and volleys?

These would be very interesting experiments that might or might not confirm the regressions.

Tennis is so homogeneous these days that there are no natural experiments left.

[jeremyrosen]

Ah, found the articles about Fed and Simpson's Paradox, but it was about losing matches where the winner won less than 50% of total points. At that time in 2014 Fed had the worst such record, winning 4 vs 24 loses when winning more points.

Here is the original paper by RYAN RODENBERG is an assistant professor of sports law at Florida State University. and Jeff Sackman et al, of Tennis Abstract:

https://www.ingentaconnect.com/content/uwic/ujpa/2013/00000013/00000003/art00019

And the more reader-friendly Atlantic article:

https://www.theatlantic.com/entertainment/archive/2014/01/what-every-pro-tennis-player-does-better-than-roger-federer/283007/

Excerpt from ATM:

At the other end of the Simpson’s Paradox spectrum was, of course, Roger Federer. In completed matches, he was 4-24 in contests where the winner prevailed on less than 50 percent of the total points. Federer’s winning percentage in these matches (14.29 percent) was the worst among all 72 players in the sample who participated in at least 20 matches of this type during their careers. This result surprised us, as it differed wildly from other players who had similarly won multiple Grand Slam singles titles. Andre Agassi, Rafael Nadal, Pete Sampras, Sergi Bruguera, Marat Safin, Lleyton Hewitt, Yevgeny Kafelnikov, and Gustavo Kuerten were all .500 or better in Simpson’s Paradox matches. Jim Courier was the only player worse than 50-50 in such matches, with a non-alarming 11-15 record.

Thanks for the links! Section 6 of the paper is particularly interesting; I wonder if anyone's looked for examples of match fixing where instead of losing a match on purpose, a player only loses a set. Also, Jeff Sackmann is the creator of the Match Charting Project, so if you're looking to analyze charted matches for academic use, that's a great place to go.

[jeremyrosen]

Thank you for this article!!!!

Craig wrote that the net yields a 70% advantage today and yesterday. The problem is that people hardly come in these days. I wondered if this was not due to people being passed more often.

70% of 10 times a match does not equal 70% of 60 times a match.

If players are coming in less, then they may be simply reacting to what is happening on the court.

Regressions are nice but even better would be an experiment. Ask players to practice certain styles and then calculate win percentages.

What happens if SABR was used a lot more?
What happens if someone actually serves and volleys?

These would be very interesting experiments that might or might not confirm the regressions.

Tennis is so homogeneous these days that there are no natural experiments left.

That's a very good point that 70% of 10 times is not the same thing as 70% of 60 times. Also, it would be great to get players to participate in those sort of experiments, but the experiments would be hard to control because players will probably still be more likely to come to net if they're confident it's going to work than if they aren't confident. We'd need a player who's willing to test out strategies even if he or she ends up losing as a result.

[drsous]

Thanks for the article. Some interesting ways to look at the stats. For me personally I am comfortable with not worrying about Confounders or maybe it’s just simpler for me to understand. I personally feel if someone sets the point up well or selects the correct time to come in, that is a part of the Net game which is inseparable. A great net player will use the previous serve or approach to creat the floater, they will also use their ground strokes or tactics to set up a ball that is easier to approach on. No one will purposely give you a short ball or a floater. You earn it with your game and so I am happy to count all approaches. But I think the article was very interesting.

[jeremyrosen]

Thanks for the article. Some interesting ways to look at the stats. For me personally I am comfortable with not worrying about Confounders or maybe it’s just simpler for me to understand. I personally feel if someone sets the point up well or selects the correct time to come in, that is a part of the Net game which is inseparable. A great net player will use the previous serve or approach to creat the floater, they will also use their ground strokes or tactics to set up a ball that is easier to approach on. No one will purposely give you a short ball or a floater. You earn it with your game and so I am happy to count all approaches. But I think the article was very interesting.

You're welcome. For what it's worth, when I'm playing a match, I'm not thinking about statistics. I come in when I think it'll help me win the point and stay back when I think it won't. Where the stats are useful is after the match when I can watch the tape and rigorously check whether my thought process on when to come in was on target. In other words, was I not quick enough to come in or too quick to come in, maybe in general or maybe in specific situations? And if I'm using stats to make that determination, that's where I have to consider potential confounders to make sure the conclusions I draw from the stats are as accurate as possible.

[arturohernandez]

That's a very good point that 70% of 10 times is not the same thing as 70% of 60 times. Also, it would be great to get players to participate in those sort of experiments, but the experiments would be hard to control because players will probably still be more likely to come to net if they're confident it's going to work than if they aren't confident. We'd need a player who's willing to test out strategies even if he or she ends up losing as a result.

My daughter decided she wanted to switch from a two handed to a one handed topspin backhand at the age of 11. My son who is six years older did the same around the age of 12. I hit with a one handed backhand.

At some point, the two hander felt unnatural to her. Coaches complained for years to me about her one hander. My son even said it was a mistake. There is a reason that very few women hit one handers these days. She has yet to play another girl who hits with a one handed backhand. ZERO. The thinking is that girls need that other hand in order to hit a good backhand.

We have no skin in the game. A college scholarship would be great but is not our goal. We have the luxury of her playing with one hand. I have watched her backhand let her down at times. Players would just key in on the shot. But she kept working at it and it has gotten better. At 15 she is able to stay in rallies and vary the spin to that troubles other players. While girls with two handers will show some improvement in their backhands, my daughter's one hander is likely to keep improving even more.

I think natural experiments are important for tennis. I am waiting for some player to decide to serve and volley just for the heck of it. I have asked my daughter to do it sometimes when she gets bored of serving and playing points. Maybe I will ask her to do it more. When was the last time we saw a woman with a one handed backhand serve and volley?

Mauresmo? Navratilova?

[don_budge]

My daughter decided she wanted to switch from a two handed to a one handed topspin backhand at the age of 11. My son who is six years older did the same around the age of 12. I hit with a one handed backhand.

At some point, the two hander felt unnatural to her. Coaches complained for years to me about her one hander. My son even said it was a mistake. There is a reason that very few women hit one handers these days. She has yet to play another girl who hits with a one handed backhand. ZERO. The thinking is that girls need that other hand in order to hit a good backhand.

We have no skin in the game. A college scholarship would be great but is not our goal. We have the luxury of her playing with one hand. I have watched her backhand let her down at times. Players would just key in on the shot. But she kept working at it and it has gotten better. At 15 she is able to stay in rallies and vary the spin to that troubles other players. While girls with two handers will show some improvement in their backhands, my daughter's one hander is likely to keep improving even more.

I think natural experiments are important for tennis. I am waiting for some player to decide to serve and volley just for the heck of it. I have asked my daughter to do it sometimes when she gets bored of serving and playing points. Maybe I will ask her to do it more. When was the last time we saw a woman with a one handed backhand serve and volley?

Mauresmo? Navratilova?

You must be extremely proud of your daughter...because I sure as hell am. Work on the physical conditioning. If you can get her to be stronger, faster, quicker it will be a deal breaker. There is no reason she couldn't serve and volley provided the serve is good enough. Surely it is a longer road to hoe. Surely it is the more difficult. But to stick with it is the challenge. It takes courage to look down the road four or five years. The other girls are more or less following the herd and what if anything differentiates one from the other.

Crack the whip. Hire a personal trainer. She can do it I tell you. Good on you arturohernandez teaching your offspring something more important than tennis. It's about life...life choices.

[jeremyrosen]

My daughter decided she wanted to switch from a two handed to a one handed topspin backhand at the age of 11. My son who is six years older did the same around the age of 12. I hit with a one handed backhand.

At some point, the two hander felt unnatural to her. Coaches complained for years to me about her one hander. My son even said it was a mistake. There is a reason that very few women hit one handers these days. She has yet to play another girl who hits with a one handed backhand. ZERO. The thinking is that girls need that other hand in order to hit a good backhand.

We have no skin in the game. A college scholarship would be great but is not our goal. We have the luxury of her playing with one hand. I have watched her backhand let her down at times. Players would just key in on the shot. But she kept working at it and it has gotten better. At 15 she is able to stay in rallies and vary the spin to that troubles other players. While girls with two handers will show some improvement in their backhands, my daughter's one hander is likely to keep improving even more.

I think natural experiments are important for tennis. I am waiting for some player to decide to serve and volley just for the heck of it. I have asked my daughter to do it sometimes when she gets bored of serving and playing points. Maybe I will ask her to do it more. When was the last time we saw a woman with a one handed backhand serve and volley?

Mauresmo? Navratilova?

Best of luck to her, I hope it works out! It would be nice to see more diversity of playing styles.