sorry wrong link... he has 2 of them
http://www.samasamtennis.com/index.php/2011/05/the-most-important-stroke-in-tennis/
http://www.samasamtennis.com/index.php/2011/05/the-most-neglected-important-stroke-in-tennis/
serve and return.
the FH is over-discussed
- Thread starter luvforty
- Start date
sorry wrong link... he has 2 of them
http://www.samasamtennis.com/index.php/2011/05/the-most-important-stroke-in-tennis/
http://www.samasamtennis.com/index.php/2011/05/the-most-neglected-important-stroke-in-tennis/
serve and return.
5263
G.O.A.T.
What did I miss here. He seems to be saying the serve is king,
not the second serve rtn.
5263
G.O.A.T.
Lol, you are serious aren't you. :???:
Does not the above you posted prove my point? I rarely post or discuss the
common knowledge items like this about the second serve.
Is this really new and exciting info for you? How long have you been playing?
I expect you can go back at least 10 years and find stuff on that topic and
I'm pretty sure it was on tennisone 15 yrs ago. I surely would not want to
be rude to a new player in discovery, but since you were so arrogant with
your comments.... I'm being more direct.
you kidding? player doesn't serve into a vacuum... the serve and the return decides the outcome of the serves, and ultimately the match.
my 2nd serve return comment was targeted at the contest of the top pros, and the analysis is inline with the USATODAY article i quoted above, that article has a couple of charts showing the increasing correlation of the return.
bottom line is the serve and the return goes hand in hand, and the first exchange is the deciding factor.
ok, so you knew that 15 years ago, and now you have reached the conclusion of the mid ct ball after charting 50 matches... so instead of questioning your knowledge, I am questioning your logical thinking ability...
which one do you prefer?
5263
G.O.A.T.
I prefer you to sit and gloat that you have discovered a perspective that has
been around at least 15 yrs and likely more. It's probably in my Bill Tilden book.
Really, how long have you been playing?
I stated from the beginning that this was my theory/opinion, and
it was based on my charting results.
I don't expect that to compare to the Bible or something, but do enjoy
getting a chance to share the concept on here and also
to have you oppose it sooo strongly. Maybe if you stay around awhile, I'll
get to see what you say as the theory becomes more accepted.
Or maybe something will come along and show me I was wrong.
Your primitive use of stats didn't make a scratch.
DoctorBackhand
Rookie
Must every thread degenerate into this?
For once can thread live for 1 week without this happening?
For once can thread live for 1 week without this happening?
5263
G.O.A.T.
I agree and thought we were having a reasonable discussion till he made a
comment insulting the intelligence of the forum group, and how his use of
stats was beyond the posters.
Last edited:
Relinquis
Hall of Fame
Yes. Your brain is SOOOOO BIG we don't want to talk with you!
Many of us work with stats in our day jobs, or used to and now have our analysts handle that work. We care more about playing tennis than what shot a pro needs to focus on to advance in his career. The pro-stats you're talking about are not relevant to our tennis at the amateur level, when we're still trying to get the strokes and basic tactics down, i.e. check this video of me hitting XYZ shot, what am i doing wrong? Or, how do I beat a baseline grinder who keeps attacking my backhand? type threads.
People here aren't participating because they don't care about what the pros need to move from no.35 to no. 3 in the world. Maybe if you had a differently titled thread in the pro player discussion board you'd get all the fanbois and celebrity coaches chiming in. Try it. It might get you the data driven arguments about fat tails, outliers and heteroscedasticity that you are yearning for.
@ Dr.BH - i thought we had a legit discussion too, until he started the BS by saying that 15 years ago he already knew serve/return was the most important, (he didn't mention that in the 2 days worth of posting on the subject, but only started saying this after I showed overwhelming results on google), yet after charting 50 pro matches (with his admittedly possibly biased method), he has reached the conclusion of the mid ct ball.
so is it a) BS or is it b) he that impaired in thinking? .... i am giving him the benefit of the doubt on the b) and therefore had no option but pointing out the a).
then he keeps defending the a), so that left me with option b).
@relinquis - no, the pro stats don't apply to recs... I already said it repeatedly.... you and most people don't care... that's cool.
my point was not on the pro stats alone.... if any player/coach has some methodology that improves their winning, I am all ears.... I know people using spreadsheets, but at this day and age, we can do much better than spreadsheets.
so is it a) BS or is it b) he that impaired in thinking? .... i am giving him the benefit of the doubt on the b) and therefore had no option but pointing out the a).
then he keeps defending the a), so that left me with option b).
@relinquis - no, the pro stats don't apply to recs... I already said it repeatedly.... you and most people don't care... that's cool.
my point was not on the pro stats alone.... if any player/coach has some methodology that improves their winning, I am all ears.... I know people using spreadsheets, but at this day and age, we can do much better than spreadsheets.
arche3
Banned
I think the statistics are interesting actually. The pros all have very polished and efficient games. So it makes sense to look at them to see what the actual statistics are of how matches are won and lost. Assuming a player has a very well rounded game these insights can actually help for the non pro advanced player or developing juniors.
Relinquis
Hall of Fame
luvforty, have you checked out the Babolot play & connect racquet? maybe with technological advances we'll be able to collect these stats at the amateur level.
http://tt.tennis-warehouse.com/showthread.php?t=441461
http://tt.tennis-warehouse.com/showthread.php?t=441461
5263
G.O.A.T.
I agree, the stats are interesting. That is why I enjoyed the early part of the
discussion, but I just don't agree with the common interpretation that was offered
related to them.
5263
G.O.A.T.
Above is an excellent example of how the important subtle items are misinterpreted.
I didn't say I knew they were the most important. I said I knew it was
a common theory that the second serve rtn was important.
A very big difference, but if you can't get that, explains why you can't get the
stats.
Also I never said the serve was not important...Only that the second sever rtn
skills are overrated, and not as important as dealing with the mid ct ball.
I guess you are a new player since you won't tell your time in the game and it's
fine that you just realized this common notion about the game (which may be
correct). I just thought it was worth pointing out when you started insulting
everyone for not being smart enough to get this.
just checked.. thx...
i think - technologies always get better and cheaper.. smart rackets, smart shoes (miCoach), how about smart balls, or amateur version of shot-tracker.
now after the match, arche3 can go analyzing all this data with his son - serve%, 2nd serve risk level (see julian's thread), return points won/lost, average court position (hugging baseline or pushed back), net points... you name it.
how about equipment fitting, racket/string combo, measurement of ball speed, spin rate with different possibilities.
how about scouting reports, opponents tendencies.
unlimited possibilities.
isn't that much sweeter than just give a pad on the back and say 'well, maybe we need some more serve practice'.
some of the stuff is already available, but most are still in infancy or too expensive for the recs.
what are the industry trends, what are the high performance people using to squeeze every ounce of winning out of a player?
survival of the fittest.
arche3
Banned
Like " moneyball" for an individual player in tennis. The advanced stats tracking ultimately comes down to whom is interpreting the data though.
5263
G.O.A.T.
Important to realize.
well... mis-interpretation happens all the time on the internet.. that's a shame.
so instead of mid ct ball vs. serve/return.... it was mid ct ball vs. 2nd serve return.... wow, huge difference there... so in your mind serve and return are so far removed from each other.... the server hits a ball into the vacuum, and the receiver just has this random ball to hit back.
read all the posts again... you said the MOST IMPORTANT difference maker is the mid ct ball, and I am saying it's false.... doesn't matter it's the serve/return combo that defeats your theory, or it's specifically the 2nd serve return..... I am referring to the 2nd serve return because the serve/return stats happen to pointing to it, while 10 years ago the serving stats maybe more relevant.
I have played the game for 30 years, and have known the importance of serve/return for maybe 20 years.... but that does not matter, so stop putting yourself on the seniority high ground.
so did you or did you not know, 15 years ago, that the serve/return was the most important?
I still don't see your way out of either option a) or b)
sure - but don't compare yourself to the moneyball wizkid yet.... that kid provided a concrete interpretation that is different from the old farts.
but so far, instead of providing anything concrete, all you have is your mid ct ball theory (with no details), and a claim that the direct cause-effect relationship between serve/return to winning/ranking needs to consider other factors of how the points play out.
how? where is the detail? what is the methodology to take other factors into consideration?
there is revolutionary analysis, like the moneyball kid, or there is analysis paralysis - where the straightforward facts are being ignored, and attention is misdirected to minutia.
5263
G.O.A.T.
I didn't compare myself to them at all, but seems you are trying to
I doubt you even got the premise of moneyball.
Last edited:
I didn't compare myself to them at all, but seems you are trying to
I doubt you even got the premise of moneyball.
the premise to me is 2 fold
1) thru data mining, and open minded thinking, you can find diamond in the rough, that is only if the method is scientific..... therefore the mid ct ball is not a diamond, but rather a fools gold
2) the establishment is the establishment - it was foolish for him not to take the Boston offer..... so, using a 50-match chart to topple the established cause effect relationship between serve/return and win/ranking, is, well.......
5263
G.O.A.T.
5263
G.O.A.T.
1) thru data mining, and open minded thinking, you can find diamond in the rough, that is only if the method is scientific..... therefore the mid ct ball is not a diamond, but rather a fools gold
..
Lets just take a look at how this misses the mark, based on the movie anyway.
A diamond in the rough is something exceptional that just needs to be polished
up to show how exceptional it is....like a young draft choice that needs to spend
time in the minors. Baseball already found Diamonds in the rough, along with
the star diamonds, already polished.
The big key here is not finding a Rough Diamond. It's about recognizing
the impact of the economic disparity between the enormous costs of polished &
rough diamonds vs the Bargain basement prices of the broken and flawed rejects
that can still get the job done.
Thru the data mining they could find bargain rejects that could still serve the
purpose, but fit in the limited payroll they worked with.
So no, they didn't find diamonds in the rough that just need polishing, they
found cheap rejects that polish wouldn't help, but could still fit their needs.
And No, not only thru Scientific methods, but good common sense and some
perspective. Statistics do serve well to provide data, evidence, and support
for the conclusions, but....
Who out there didn't look at $150mil contracts and say they could get 5-7 rock
solid players for that and make a better team than putting that in one player
who can get hurt or go cold, sticking you with a big salary you can't move??
Good common sense would keep most of us from getting trapped in a bid war
to get the Prized Diamond. The team owner avoided that trap.
And as to the Establishment is the Establishment...that changed too.
Many of the teams started using some of these same methods to improve their
rosters after that season.
Yes, it's a huge challenge to turn that big ship, but this movie is about some
unlikely folks doing just that. You could say MoneyBall is not only about finding
undervalued assets thru data mining, but also about how people have some
great ideas out there and having the guts to see it through.
How sometimes they can see it through for Big Change
and how throwing money at things is often not the best approach.
Last edited:
5263
G.O.A.T.
Not aware I need a methodology to get the concept of a movie.
I looked back over the thread and think we may have gotten sidetracked.
Seems you were first saying the Bh was most more important to improve.
Then return of second serve,
then just return of serve in general, and
then Serve and any return of serve.
You are only as good as your worst shot, so whatever is worst.
Also, Are you saying the pick from above is the most important or the difference
maker? or both?
Just want to get it straight, what your point is, thanks.
Last edited:
the thread of the title is 'FH is over-discussed'.... I was just saying too much attention was put on the FH.
I did mention that recs usually have very weak bh and even worse volley so perhaps the low-fruits are in those areas but not the fh.... but I have no stats to go with.... nobody cares to collect data for the recs.
but data for the tour is available, therefore I tried to use the ATP data to illustrate the common notion that the serve is the most important shot, and I made another point that it's really the "first exchange", i.e. the serve and the return, because after that, both players have the same role - a "rallier".
fortunately ATP has both serving and returning data available, so it was convenient to see the correlation between the return stats and the player ranking.... and coincidentally dom had another thread about the top 4, so I put the scope on the top4.
to summarize the point - at the elite level, the "first exchange" is the most important shot(s), this does not conflict with the old saying of "serve is the most important", because the serve is not hit into a vacuum, its effectiveness depends on what the returner does..... and a few links from google - an usatoday article, an article from a tennis pro, expressed the same view so I quoted them.
so - in a tennis point, there is the "first exchange", where A is the server and B is the returner, after the first exchange, both A and B are ralliers..... so the question is, a) does the better server/returner win, or b) does the better rallier win? and the answer is -
if player has similar performance in their serving / returning stats in relation to their ATP ranking, e.g. rafa and ferrer, (whom I call the 'symmetrical players) then both answers a) and b) are possible. the explanation can be that either the 'first exchange' is not making the difference, or these 2 guys happen to do serving/returning equally well.
but if the player is asymmetrical, fed, murray, joke, and majority of the top guys, then the explanation can only be a), since we assume their 'rallier' performance being roughly the same during serving vs. receiving games, so the difference maker has to the 'first exchange' when they are in the 'server'/'returner' role.
however, i did acknowledge many times that these stats do not apply to recs, and your mid court ball theory may very well be valid for the recs.... if one 3.5 guy dinks a serve to another 3.5 guy who dinks back, then sure, whoever punishes the mid ct ball wins.
the thread started to de-emphasize on the FH, then took on quite a life of its own lol.
I did mention that recs usually have very weak bh and even worse volley so perhaps the low-fruits are in those areas but not the fh.... but I have no stats to go with.... nobody cares to collect data for the recs.
but data for the tour is available, therefore I tried to use the ATP data to illustrate the common notion that the serve is the most important shot, and I made another point that it's really the "first exchange", i.e. the serve and the return, because after that, both players have the same role - a "rallier".
fortunately ATP has both serving and returning data available, so it was convenient to see the correlation between the return stats and the player ranking.... and coincidentally dom had another thread about the top 4, so I put the scope on the top4.
to summarize the point - at the elite level, the "first exchange" is the most important shot(s), this does not conflict with the old saying of "serve is the most important", because the serve is not hit into a vacuum, its effectiveness depends on what the returner does..... and a few links from google - an usatoday article, an article from a tennis pro, expressed the same view so I quoted them.
so - in a tennis point, there is the "first exchange", where A is the server and B is the returner, after the first exchange, both A and B are ralliers..... so the question is, a) does the better server/returner win, or b) does the better rallier win? and the answer is -
if player has similar performance in their serving / returning stats in relation to their ATP ranking, e.g. rafa and ferrer, (whom I call the 'symmetrical players) then both answers a) and b) are possible. the explanation can be that either the 'first exchange' is not making the difference, or these 2 guys happen to do serving/returning equally well.
but if the player is asymmetrical, fed, murray, joke, and majority of the top guys, then the explanation can only be a), since we assume their 'rallier' performance being roughly the same during serving vs. receiving games, so the difference maker has to the 'first exchange' when they are in the 'server'/'returner' role.
however, i did acknowledge many times that these stats do not apply to recs, and your mid court ball theory may very well be valid for the recs.... if one 3.5 guy dinks a serve to another 3.5 guy who dinks back, then sure, whoever punishes the mid ct ball wins.
the thread started to de-emphasize on the FH, then took on quite a life of its own lol.
Last edited:
5263
G.O.A.T.
I missed this excellent post earlier. Well stated point here. And just to note,
while both of them are excellent servers in their way, neither are legendary
like Goran, Pete, or Roddick. The Fh is their main and best weapon and each
have chinks in the armor of their Bhs.
That said, other than the seat of the pants stat of 15% factor on Fhs, I do agree that
most every player would benefit greatly from getting more balance in their
games by improving, serves, Rtn, and Bhs along with some Fh work.
5263
G.O.A.T.
Looking back over the thread and this post, Imo you made quite a few good pts,
even though we don't agree on what the 2ond serve stats reflect.
Fh is likely way over scrutinized in general and players would benefit greatly from
developing better overall skills to go with their weapon of choice.
I do expect they try to an extent, but it's just not working out for them.
(referring to Inspector's post) with the risk of sounding arrogant, here is another example of different level of analyzing.
anecdotal arguments like the one from Inspector is exactly why internet debates go in circles to infinity.... because another party can provide equally if not more convincing arguments for the contrary.. e.g. Agassi saying fed is better than pete because of Fed's bh, and fed saying rafa so tough because of rafa's backhand.
stats is needed to reveal the true story... see my previous post, why the FH is not the difference maker...first of all, a definition... we are talking about THE MOST IMPORTANT, not ONE OF THE MOST IMPORTANT.
so fed's FH is one of the most important, no doubt, but what is THE MOST IMPORTANT, (not just for Fed, we are talking about the entire tour, which ATP collects stats for).
and if you follow my paragraph about symmetrical/asymmetrical player above, hopefully you can make sense out of it.
anecdotal arguments like the one from Inspector is exactly why internet debates go in circles to infinity.... because another party can provide equally if not more convincing arguments for the contrary.. e.g. Agassi saying fed is better than pete because of Fed's bh, and fed saying rafa so tough because of rafa's backhand.
stats is needed to reveal the true story... see my previous post, why the FH is not the difference maker...first of all, a definition... we are talking about THE MOST IMPORTANT, not ONE OF THE MOST IMPORTANT.
so fed's FH is one of the most important, no doubt, but what is THE MOST IMPORTANT, (not just for Fed, we are talking about the entire tour, which ATP collects stats for).
and if you follow my paragraph about symmetrical/asymmetrical player above, hopefully you can make sense out of it.
moneyball is awesome.
I think for the top pros, the stats are already widely used, because so much is at steak, and so much data is available..... but we recs don't get to know what's going on.
I think coaches who live on teaching fees and those who live on % of player earnings have VERY different business models....
the former has to promote his methodology, to make as many people aware as possible.
the latter has to operate under the radar with trade secretes unrevealed..... too much $$$ at steak, any competitive edge is gold.
I think for the top pros, the stats are already widely used, because so much is at steak, and so much data is available..... but we recs don't get to know what's going on.
I think coaches who live on teaching fees and those who live on % of player earnings have VERY different business models....
the former has to promote his methodology, to make as many people aware as possible.
the latter has to operate under the radar with trade secretes unrevealed..... too much $$$ at steak, any competitive edge is gold.
arche3
Banned
I'm actually not a numbers person at all...but I loved the story of moneyball and how it changed everything they thought.
Would it be weird if I showed up to rec tennis matches with some statistics college kid on a tablet inputing every ball i hit ?
WildVolley
Legend
Until now, the difficulty in doing anything more than cursory statistical analysis of tennis has been gathering good data to work with. With the shot-tracker type technology that records every ball hit, the speed, and the trajectory, it should be possible to do some fairly sophisticated statistical analysis.
Well before moneyball, I had an economics professor who was trying to get a college to help him set-up a "sportametrics" department that would apply statistical analysis to sports. I guess he was tired of writing economics papers.
So the argument so far in this thread has been about having a good serve and a good return game. I think we can all agree on that.
However, what does a good return game look like? Does the Agassi high-risk strategy still work today?
Well before moneyball, I had an economics professor who was trying to get a college to help him set-up a "sportametrics" department that would apply statistical analysis to sports. I guess he was tired of writing economics papers.
So the argument so far in this thread has been about having a good serve and a good return game. I think we can all agree on that.
However, what does a good return game look like? Does the Agassi high-risk strategy still work today?
5263
G.O.A.T.
Interesting question, but tennis is so different in how it's played and scored,
I'm not sure how much the stats help.
In tennis, stats are quite lacking to begin with, just glossing the surface of
what is happening, and really mostly just tell how things are currently going
in general.
I do know they buy the hawkeye data though and am in contact with a coach
using it.
Like if back in the heyday of S&V, stats likely showed coming to net was the way to win since the best players were doing it. Now S&V wouldn't fare as
well in stats. Doesn't really mean it is actually the key and may be just a fad
or trend that is self perpetuating. People come up with theories to support
their belief and stats show the current trend.
Main use now is for trends, like he will go cross ct in xyz situation.
Last edited:
them pros all have the play-books against each other, tendencies under all situations..where a guy serves to defend BP, shot patterns... it's the scout's job to figure all that out and make the player aware.
arche3 you will be so busy during them junior tournaments scouting on your son's next round opp while keeping an eye on his own match lol.... you have work cut out for ya.
arche3 you will be so busy during them junior tournaments scouting on your son's next round opp while keeping an eye on his own match lol.... you have work cut out for ya.
that's my thinking as well about the pros.... them pro coaches must have been crunching all the shot tracker data to figure out shot patterns, not for his own player, but for the opps.... tremendous advantage without having to 'feel out the other guy' during the first few games.
http://www.atpworldtour.com/Rankings/MatchFacts.aspx
as for if Agassi will be successful... i don't know... look at the link above, rafa and joker tops the 'return games won' while they rank very high on returning both the first and 2nd serve.... murray has good success in return games won, he is on top returning 2nd serve, but only decent against 1st serve.... and Fed is dismal in the return department for his standards due to the 1hbh.
so I guess the story of the top4 says, you gotta be able to go after both the 1st and 2nd serves.... so perhaps Agassi will be successful.
http://www.atpworldtour.com/Rankings/Top-Matchfacts.aspx?y=1991&s=0#
the ATP stats go all the way back to 1991 when Edberg was #1 !
and if you highlight the stats area and copy paste into a spreadsheet, it lay out nicely into different columns
so anybody curious enough can slice dice the data easily just by using some basic Excel correlation formulas, and show which column is the most relevant for success.
the ATP stats go all the way back to 1991 when Edberg was #1 !
and if you highlight the stats area and copy paste into a spreadsheet, it lay out nicely into different columns
so anybody curious enough can slice dice the data easily just by using some basic Excel correlation formulas, and show which column is the most relevant for success.
1st Srv 1st Srv w 2nd srv w srv gm w bp won bp sv pts rtn 1 pts rtn 2 rtn gm wn
0.046853021 0.268923042 0.562500096 0.44739878 0.302596267 0.2950009 0.181784048 0.222073506 0.282499512
sorry, had formatting error....
for 1991, the top 3 are -
2nd Srv Won
Srv Game Won
BP Won.
0.046853021 0.268923042 0.562500096 0.44739878 0.302596267 0.2950009 0.181784048 0.222073506 0.282499512
sorry, had formatting error....
for 1991, the top 3 are -
2nd Srv Won
Srv Game Won
BP Won.
Last edited:
arche3
Banned
Lol. I'm one of the coaches for my sons football team. Last season (6th grade junior football league) I started tracking our play calling on 3rd down conversions. It basically told us to always pass. We were much more likely to get a 1st down passing than running. No matter how many yards were needed. We started passing only on 3rd downs on second half of season and we were much more consistent in converting.
Even though the other teams scouted us and knew we always passed on 3rd it didn't really matter. We ran a spread offence on 3rd and the qb always has the option to run. But the first option is to look for one of the receivers.
Sorry way off topic. I'm talking about football ....
for 2012
1st Srv 1st Srv w 2nd srv w srv gm w bp won bp sv pts rtn 1 pts rtn 2 rtn gm wn
0.06429448 0.522678196 0.614081536 0.650474016 0.248547356 0.536628334 0.349750106 0.288385001 0.38757496
the top 3 are
Srv Gm won
2nd Srv won
BP Save
the top 3 specific to a certain stroke are
2nd Srv won
1st Srv won
pts rtn 1st Srv.
the stats above are for the top 100, so beyond the top 4, the picture is a bit different....
what does this all mean?
1st Srv 1st Srv w 2nd srv w srv gm w bp won bp sv pts rtn 1 pts rtn 2 rtn gm wn
0.06429448 0.522678196 0.614081536 0.650474016 0.248547356 0.536628334 0.349750106 0.288385001 0.38757496
the top 3 are
Srv Gm won
2nd Srv won
BP Save
the top 3 specific to a certain stroke are
2nd Srv won
1st Srv won
pts rtn 1st Srv.
the stats above are for the top 100, so beyond the top 4, the picture is a bit different....
what does this all mean?
5263
G.O.A.T.
but didn't we know that?
isn't it more important to know how they did that? other than serve 135 out wide ace.
martini1
Hall of Fame
I still believe without a serve and a good bh there is just no way. People started learning the fh as their first shot, and naturally it is the shot that got the most hours logged, and the first one to become a weapon.
Somebody can have a 5.0 fh but with a 3.0 bh they will win very few points as long as the other guy can simply redirect the ball to the bh side.
Somebody can have a 5.0 fh but with a 3.0 bh they will win very few points as long as the other guy can simply redirect the ball to the bh side.
Yes - here it is..
first of all the raw data in a cleaner format - year 2012
Stat / correlation
1st Srv% / 0.06429448
1st Srv pts won / 0.522678196
2nd srv pts won / 0.614081536
srv gm won / 0.650474016
bp won / 0.248547356
bp sv / 0.536628334
pts rtn 1st srv / 0.349750106
pts rtn 2nd srv / 0.288385001
rtn gm won / 0.38757496
so the goal is to figure out what is really happening from this set of data.
first of all, we can say that the 1st Srv% is almost irrelevant, which makes sense.... a short guy serving 70% and a tall guy serving 70% can bring vastly different results.
2ndly, we want to use only the stats related to a specific shot, so we use these 4 stats -
1st Srv pts won / 0.522678196
2nd srv pts won / 0.614081536
pts rtn 1st srv / 0.349750106
pts rtn 2nd srv / 0.288385001
the rest of them, games/breaks, bp won/sv, we don't know what exactly happened (point started with a 1st srv or 2nd srv), so we are gonna leave them alone.
so on surface, the above 4 stats says, winning is more correlated to the srv pts, less to the rtn pts.
question is, do higher ranked players, perform better in srv pts, because they serve better, or because they rally better (fh, bh, volleys, drop shots etc).
assuming their rally shot performance being roughly the same in their srv pts and rtn pts, then -
if the ranking has roughly the same correlation to srv pts and rtn pts (symmetry), then we can say that ranking is more dependent on the rally performance.
if the ranking has asymmetrical correlation to srv pts and rtn pts, then we can say that ranking is more dependent on the srv or the rtn, which ever with the higher correlation (stats above pointing to the srv).
therefore considering the asymmetrical nature of the above 4 stats, the conclusion is that the srv is the most correlated to winning.... and the 2nd serve slightly more important than the 1st.
hopefully this analysis makes sense.
by the way, the 1991 data tells a similar story, the corresponding 4 stats are
0.268923042
0.562500096
0.181784048
0.222073506
and interestingly in 1991, the 2nd srv is much more important than the other 3.
so the saying is valid after all -
YOU ARE ONLY AS GOOD AS YOUR 2ND SERVE !!
Last edited:
5263
G.O.A.T.
If that is what you were saying all along, we really got sidetracked.
I thought you felt the key was rtn of second serve.
Are you sure?
"Stat / correlation
1st Srv% / 0.06429448"
OR
"Stat / correlation
1st Srv% / 0.6429448"
Please see a difference-ONE LESS ZERO in the second case?
Is it
"Stat / correlation
1st Srv% / 0.06429448"
OR
"Stat / correlation
1st Srv% / 0.6429448"
Please see a difference-ONE LESS ZERO in the second case?
initially only had the top 4 to look at...... the analysis today was on the top 100.
no... 0.06429448 is correct.... amazing huh.... 1st serve% is irrelevant....
well, in the sense that player A serving higher 1st srv% does not mean he is better than player B serving at lower 1st srv%.
Similar threads
