Draw Fixing: An Official Study

First topic message reminder :

The link below shows a talk by an Estionian researcher Katarina Pijetlovic, giving an exposition on draw fixing at a Corruption in Sport Symposium in Koln.


Katarina's talk starts at around 13 minutes:

http://www.livestream.com/playthegame_dshs/video?clipId=pla_44809e94-aa04-46c7-9f1e-35b212ba9d46


She examines the pattern of draws at slam tournaments (French Open was not part of the study) between 2007-2011, drawing the conclusion that ITF organised draw fixing on behalf of Nike seeing that Djokovic fell in Federer's half of the draw statistically virtually impossible 12 out of 12 times.

Roland Garros was not taken into the study as it showed a healthy 50/50 pattern.



Interesting facts, e.g. I didn't know that seeds 3 and 4 are drawn by hand unlike all the other seeds/players that are computer drawn.
Draws are apparently public and televised, but not really accessible anywhere on Youtube.

To me, the most blatant example of draw fixing was the Isner Mahut match played in the first round last year ON COURT 18, just like at the record breaking match the year before!!!


Katarina did the research hoping it would interest sports journalists and encourage them to contact the players and ITF.


So far nothing came out of it.

Have a look with an open mind and share your thoughts.

actually, just to be pedantic Tenez, but 6H6T isn't even close to 50%, it's more like 22.5%

Mad for Chelsea wrote:actually, just to be pedantic Tenez, but 6H6T isn't even close to 50%, it's more like 22.5%

yes, good point...I meant 6 each.

You see you can understand when you want.

Tenez wrote:

Mad for Chelsea wrote:actually, just to be pedantic Tenez, but 6H6T isn't even close to 50%, it's more like 22.5%

yes, good point...I meant 6 each.

You see you can understand when you want.

eh? in a sequence of 12 coin tosses the likeliest outcome is having 6 heads and 6 tails (in total), but the probabiliy of this outcome is only about 22.5% (it's C(12,6)/2^12). In fact it's far more likely you'll get 7 of one and 5 of the other (probability roughly 38.6 %) and slightly more likely you'll get 8 of one and 4 of the other (about 24.1%).

Mad for Chelsea wrote:
eh? in a sequence of 12 coin tosses the likeliest outcome is having 6 heads and 6 tails (in total),

You are still thinking in sequences. That's where you get it wrong. the fact is the more you throw a coin the closer you get to have 50%T and 50%H.

noleisthebest wrote:

If you think there is nothing in the 12/12 "concidence" find another, even remotely close pattern for the top 2 players. any era.

I provided examples, which you conveniently ignore.

Tenez wrote:

Mad for Chelsea wrote:
eh? in a sequence of 12 coin tosses the likeliest outcome is having 6 heads and 6 tails (in total),

You are still thinking in sequences. That's where you get it wrong. the fact is the more you throw a coin the closer you get to have 50%T and 50%H.

yes that's true. However the probability of getting exactly n heads and n tails out of 2n coin tosses tends to 0 (it's C(2n,n)/2^(2n) which is asymptotic to some constant multiplied by n^(-1/2)). You're right though that the proportion of heads (and of tails) tends to 50% (law of large numbers). Basically, if you toss a coin a very large number of times you should expect the number of heads to be close to 50%, but not exactly 50%. Of course, 12 isn't really that large.

Summerblues has got the gist of this thread, and I enjoyed his summary. Thanks for reading through this, man, and I'm happy it pleased the enrd in you.

Barrystar has grasped the argument, but still seems confused about how the probabilities can be calculated when rankings change. It's the same of course, which is 1/2 to the power 12 since each Federer- Djoker meeting has a probability of 1/2.

Henman Bill and new balls sadly are yet to get the argument.

Lydian doesn't seem to want to get it, as does Hawk eye.

All this is just about calculating the odds. There is no debate here. This is science.

As I've mentioned before, the debate only comes when one tries to explain the fact thta such a low chance event has occured.

Mad for Chelsea, you're right, but that's a bit of a digression.

Laver fan, your examples are ridiculous. Do read them again and spot the flaw(s). if you do not, I can explain it but if you do, I do not want to waste my time. Do let me know.

laverfan wrote:

noleisthebest wrote:

If you think there is nothing in the 12/12 "concidence" find another, even remotely close pattern for the top 2 players. any era.

I provided examples, which you conveniently ignore.

which examples, when were Federer and Nole not in the same half of the draw in Wimbledon, USO and AO 2008-2011?

Tennisanorak wrote:Mad for Chelsea, you're right, but that's a bit of a digression.

I know, I was merely disputing Tenez's statement that there was a 50% chance Novak and Federer would end up in opposite sides half the time.

noleisthebest wrote:Barry is defending the honour of Wimbledon here.

Not really I'm being entirely practical and hard-headed .

I don't have any view on whether the guys who run Wimbledon are any more honourable than anyone else - I rather doubt it because they are all the same sort of people.

The AELTCC are in the fortunate position of not needing to care about money half as much as, say, the USTA. The evidence is there for all to see that the USTA do more to the USO scheduling to maximise TV money than at Wimbledon. If Wimbledon belonged to the LTA I bet they'd be more nakedly commercial about it than the AELTCC.

The fact that the AELTCC react differently to the USTA in the face of different commercial pressures doesn't make them more honourable, you know, don't judge a guy till you've walked a mile in his shoes and all that....

Tennisanorak wrote:Summerblues has got the gist of this thread, and I enjoyed his summary. Thanks for reading through this, man, and I'm happy it pleased the enrd in you.

Barrystar has grasped the argument, but still seems confused about how the probabilities can be calculated when rankings change. It's the same of course, which is 1/2 to the power 12 since each Federer- Djoker meeting has a probability of 1/2.

Henman Bill and new balls sadly are yet to get the argument.

Lydian doesn't seem to want to get it, as does Hawk eye.

All this is just about calculating the odds. There is no debate here. This is science.

As I've mentioned before, the debate only comes when one tries to explain the fact thta such a low chance event has occured.

And it's clear you know what you are talking about.

Tennisanorak wrote:Lydian doesn't seem to want to get it...

When discussing Nadal and Sampras with Lydian this is a constant! I have argued for what 5 years with Lydian but those 2 players carry far too much emotion to get a proper admission of facts from Lydian.

He would even go as far as saying Pete and Nadal don;t know what they are talking about if they contradict him. No kidding.

barrystar wrote:

Not really I'm being entirely practical and hard-headed .

I don't have any view on whether the guys who run Wimbledon are any more honourable than anyone else - I rather doubt it because they are all the same sort of people.

And because it's down to human nature, we know that when there are big stars and lots of money involved, little twicks sre usually done. Slowing down conds is one (BTW Wimbledon has also always refused to admit it officially) and rigging draw can be another one.

Tennisanorak wrote:Laver fan, your examples are ridiculous. Do read them again and spot the flaw(s). if you do not, I can explain it but if you do, I do not want to waste my time. Do let me know.

So give me an explanation.

There must be a private rule where the number 1 ranked player gets to choose his semi final opponent. Djokovic Federer streak is simply not possible, you don't need proof to make these assumptions!

Can't wait for the French Open draw to take place for all the conspiracy theorists like Tenez and tennis Anorak to mull over.

Oh wait a minute! Of course that's the exception (or at least one of them) to their rather flexible rigging theory. Goodness only knows what they'll make of the Olympics draw!

Tennisanorak wrote:
Barrystar has grasped the argument, but still seems confused about how the probabilities can be calculated when rankings change. It's the same of course, which is 1/2 to the power 12 since each Federer- Djoker meeting has a probability of 1/2.

OK - then it's still the case that whichever outcome of the 12 draws you have is still 1/4096, with one overall pattern of meetings no more or less likely than the other. Couple that with the fact that draws are based on numbers and not names and I don't find it too bizarre.

Tenez says it's not about sequences but I think you do, which chimes with me because the actual seeding is what comes before the draw and it is the seeding changes which creates the sequence that the numbers draw has to follow each time to achieve the desired result.

This debate is slowly descending into poster accusations, rather than the OP.

"Lydian doesn't seem to want to get it, as does Hawk eye."

"your examples are ridiculous"

"far too much emotion to get a proper admission of facts"

Absolutely wonderful.

summerblues wrote:Fourth (a side note),

Some of the posters here suggest that the draws were rigged to maximize the probability of the most lucrative - i.e., Fedal - final, and therefore Djokovic was being put to Federer's draw because Federer was more likely to make the final anyway.

Some of the same posters are suggesting that this year's MC draw is also rigged by giving Nadal an easy draw. Surely if the same argument as above applied, the draw would be rigged to give Nadal the more difficult draw? Clearly, by far the most lucrative final would be Djoker-Nadal. Also, Nadal has proven over the years that he will be in the final irrespective of the opposition. As good as Djoker is, it is him who must be viewed as the more likely one to slip and so the one in need of an easier draw.

Not sure I agree with the above but I thought your first three contributions were superb. Accurate and without bias.

newballs wrote:Can't wait for the French Open draw to take place for all the conspiracy theorists like Tenez and tennis Anorak to mull over.

Oh wait a minute! Of course that's the exception (or at least one of them) to their rather flexible rigging theory. Goodness only knows what they'll make of the Olympics draw!

YOu must be the kind of guy that still believes that Bush and co really thought that Saddam had WMDs, aren't you? Only us fools believed it was a conspiration.

Tenez wrote:

SB says it "can" make sense, even if it can also be down to some strange events as they can happen. It's a much more scientific appraoch than yours and HB.

Totally wrong. Me and SB made most of the same points and hold basically the same views. What significant difference is there between my views or approach and SB's?

Tenez wrote:

newballs wrote:Can't wait for the French Open draw to take place for all the conspiracy theorists like Tenez and tennis Anorak to mull over.

Oh wait a minute! Of course that's the exception (or at least one of them) to their rather flexible rigging theory. Goodness only knows what they'll make of the Olympics draw!

YOu must be the kind of guy that still believes that Bush and co really thought that Saddam had WMDs, aren't you? Only us fools believed it was a conspiration.

I wonder if you carry a gas mask while walking down the street or ever drink water.

Take it easy Tenez. Draw fixing does not kill anyone, but Saddam did. Just because Dubya wanted Oil and lucrative Military deals does not make Dubya as evil as Saddam (wonder about Cheney though), does it? He even supplied Saddam to fight Iran.

Tenez wrote:

newballs wrote:Can't wait for the French Open draw to take place for all the conspiracy theorists like Tenez and tennis Anorak to mull over.

Oh wait a minute! Of course that's the exception (or at least one of them) to their rather flexible rigging theory. Goodness only knows what they'll make of the Olympics draw!

YOu must be the kind of guy that still believes that Bush and co really thought that Saddam had WMDs, aren't you? Only us fools believed it was a conspiration.

Slightly different scenario methinks.

I don't believe that they knew Saddam did not have WMD's. I do believe that they only had the evidence to suspect that he might have WMD's but recklessly said that they knew he did because that was the 'legal' justification that they needed for a war that they wanted to have anyway whilst hoping against hope that: (a) they'd be able to find some WMD's afterwards to prove their suspicions; or, (b) if they did not they'd be forgiven when everyone realised what a good idea it was after all. Defining WMD so vaguely was part of wanting to give themselves wriggle room. I think it was more reprehensible on the part of Blair because he pinned his colours far more aggressively and exclusively to the WMD mast than Bush did.

Even then, you'll see that it was part of the natural human tendency to avoid a direct lie - i.e. saying he had them when they knew for certain he did not - and instead to use slightly greyer (or hopeful) sophistry.

Even so - utterly reprehensible on the part of politicians starting a war.

laverfan wrote:

Tenez wrote:

newballs wrote:Can't wait for the French Open draw to take place for all the conspiracy theorists like Tenez and tennis Anorak to mull over.

Oh wait a minute! Of course that's the exception (or at least one of them) to their rather flexible rigging theory. Goodness only knows what they'll make of the Olympics draw!

YOu must be the kind of guy that still believes that Bush and co really thought that Saddam had WMDs, aren't you? Only us fools believed it was a conspiration.

I wonder if you carry a gas mask while walking down the street or ever drink water.

Take it easy Tenez. Draw fixing does not kill anyone, but Saddam did. Just because Dubya wanted Oil and lucrative Military deals does not make Dubya as evil as Saddam (wonder about Cheney though), does it? He even supplied Saddam to fight Iran.

Again, you really don't know what you are talking about I am afraid.

barrystar wrote:I don't believe that they knew Saddam did not have WMD's.

Ouahh! Must be the English "innocent until proven guilty" thing.

I, like billions others I guess, knew without having their "Intelligence". It's called reading the opponent's game. Very important in tennis too. It really stunk from day one! I will always remember the smile Putin had on his face when Tony presented him with the "dossier".

I am sure you are aware they had decided to go to Irak and finish the job before 9/11, don;t you? It's now well documented.

Tennisanorak wrote:Summerblues has got the gist of this thread, and I enjoyed his summary. .........Henman Bill and new balls sadly are yet to get the argument.

What an odd comment. If you read back to the first couple of pages you'll see that now I think I already made most of the same points as Summerblues and hold viritually identical views.

And I got the argument throughout...the reasons given for discarding FO from the sample...the justifications given for discarding 2007...the reasons why Federer and Djokovic would need to be paired together...etc etc...all very easy to understand. Just because I don't agree or make opposing comments doesn't mean I don't "get" the opposing views. What is it that I supposedly don't understand. Name even one thing. Good luck with that. I found everything on the thread easy to understand.

Also, Nadal has proven over the years that he will be in the final irrespective of the opposition. As good as Djoker is, it is him who must be viewed as the more likely one to slip and so the one in need of an easier draw.

Djokovic has rarely struggled on clay, he struggled more on hardcourt than clay in 2011 despite his lack of a Roland Garros.

So far we've had coin toss comparisons, odds of a boy versus a girl, apples falling on Newton's head and even WMDs.

At this rate we'll get to that much cherished "Theory of Everything" although most of us well read types know that to be 42 anyway.

Mad for Chelsea wrote:in which case I misunterstood and I apologise, but my point remains. Ignoring RG and including Wimbledon makes no sense whatsoever (with regards to Djokovic), as you can make a decent case for Murray being a superior grass court player to Djokovic upto 2011.

Exactly.

So NITB says the idea of possible draw fixing was to make Fedal finals, and the idea is that they started fixing the draws in 2008. So in 2008, 2009 and 2010 they started potentially fixing Wimbledon to to have Federer-Djokovic finals. But Djokovic reached R2 in 2008 and QF in 2009. Murray reached a QF and SF in those years. Roddick reached a final in 2009. Would making Djokovic avoid Nadal in the semi really have made any difference? He failed to even make the semi in the first two years of the alleged fixing. It wouldn't have made a difference. It would have made more sense to have Nadal avoid Roddick in 2006-2009 when he was still a danger.

What about the slams Nadal didn't even play and wasn't even in the draw, like Wimbledon 2009 (are there any others)? He wasn't even in the draw and yet that slam is used as one of the ones where the draw was supposedly fixed to create Fedal finals to get to the over 4000-1 odds!

But then the argument is..they weren't all fixed, just some. And yet still the ones that don't fit the fixing argument are included in the 4000-1 stats. While the FO can be discarded from those stats apparently.

Surely we have to either add the FO in to the stats or remove Wimbledon, reducing the numbers to way below the 4000-1 level (or 1 in 4096 level or whatever it was). Unless anyone can really explain why they would have fixed a Federer-Djokovic semi final in 2009 to create a Federer-Nadal final, in a tournament Nadal did not even enter?

Henman Bill wrote:

Mad for Chelsea wrote:in which case I misunterstood and I apologise, but my point remains. Ignoring RG and including Wimbledon makes no sense whatsoever (with regards to Djokovic), as you can make a decent case for Murray being a superior grass court player to Djokovic upto 2011.

Exactly.

So NITB says the idea of possible draw fixing was to make Fedal finals, and the idea is that they started fixing the draws in 2008. So in 2008, 2009 and 2010 they started potentially fixing Wimbledon to to have Federer-Djokovic finals. But Djokovic reached R2 in 2008 and QF in 2009. Murray reached a QF and SF in those years. Roddick reached a final in 2009. Would making Djokovic avoid Nadal in the semi really have made any difference? He failed to even make the semi in the first two years of the alleged fixing. It wouldn't have made a difference. It would have made more sense to have Nadal avoid Roddick in 2006-2009 when he was still a danger.

What about the slams Nadal didn't even play and wasn't even in the draw, like Wimbledon 2009 (are there any others)? He wasn't even in the draw and yet that slam is used as one of the ones where the draw was supposedly fixed to create Fedal finals to get to the over 4000-1 odds!

But then the argument is..they weren't all fixed, just some. And yet still the ones that don't fit the fixing argument are included in the 4000-1 stats. While the FO can be discarded from those stats apparently.

Surely we have to either add the FO in to the stats or remove Wimbledon, reducing the numbers to way below the 4000-1 level (or 1 in 4096 level or whatever it was). Unless anyone can really explain why they would have fixed a Federer-Djokovic semi final in 2009 to create a Federer-Nadal final, in a tournament Nadal did not even enter?

People also believed the Earth was flat once upon a time...

Mad for Chelsea wrote:in which case I misunterstood and I apologise, but my point remains. Ignoring RG and including Wimbledon makes no sense whatsoever (with regards to Djokovic), as you can make a decent case for Murray being a superior grass court player to Djokovic upto 2011.

NO that's not a good explanation at all.

Wimbledon 2007 had Djoko really threatening Nadal before he had to retire. It showed clearly how DJoko could handle Rafa's game. Djoko is the one reaching the semis in the last 3 slams of 2007, in fact even reaching the final at the USO and winning the AO08. (Nadal was nowhere to be seen in those HC slams so Djoko was the real threat outside clay). So clearly when came Wimbledon 2008 the one people were looking at was Djoko not Murray (only a Murray fan woudl think that way). Djoko had a very weird patch in that 2008-2011 period but until that weird patch happened he was the real danger as proven by his results in 2011 and even 2010, not Murray. Sure Murray has won Nadal twice but quite unexpected. Nadal being very tired in that summer USO and then the knee. The fact is 2011 proves that Djoko was always a much feared opponent than Murray and as mentioned Nadal said it long before 2011.

FO is also very different than WImbledon cause at the FO they have done everything to speed up the courts and certainly not to help Nadal.

Nadal is a much more popular player than Djoko everywhere but at the French.

But if you don;t want to see it...no-one can help.

Henman Bill, all I meant was the argument about the probability being one in 4k. I see this being contested again and again by saying that the draw is based on seedings. It doesn't matter. The probability is still 1 in 4000 of 12 Fed- Djoker matches. There is absolutely no debate about that.

What that means is open to debate. Whether excluding French Open is right is up for debate. But not the fact that the probability is 1 in 4000 itself. It's an elementary probability calculation.

But as long as people don't accept the calculation and find nothing abnormal, it is tough to proceed. That is the starting point of the argument.

Nobody has replied to my Federer- Berdych illustration which brings out the point I was trying to make.

Laverfan, just give me some time. I will tell you why I think your Federer- Soderling and Federer- Roddick data points are not anything like a disproof.

Henman Bill wrote:What about the slams Nadal didn't even play and wasn't even in the draw, like Wimbledon 2009 (are there any others)? He wasn't even in the draw and yet that slam is used as one of the ones where the draw was supposedly fixed to create Fedal finals to get to the over 4000-1 odds!

But then the argument is..they weren't all fixed, just some. And yet still the ones that don't fit the fixing argument are included in the 4000-1 stats. While the FO can be discarded from those stats apparently.

Surely we have to either add the FO in to the stats or remove Wimbledon, reducing the numbers to way below the 4000-1 level (or 1 in 4096 level or whatever it was). Unless anyone can really explain why they would have fixed a Federer-Djokovic semi final in 2009 to create a Federer-Nadal final, in a tournament Nadal did not even enter?

Djoko and Fed were on the same side of the Wimbledon draw in 2009, and Nadal was drawn out of the hat that tournament on the opposite side, but his place in the draw was taken by JM de P so it does count.

This is an excellent thread (well done again NITB) as it enables reading of the posters mind, and in particular the logos that is in us (in most cases ). It's also interesting to see how emotions can affect our logic (those who do not want to see rigging and those who want to see it all cost).

I felt Socal wanted to see it at all cost for instance when he first mentioned it but this study (though I have not personally verified the first draw stats but can't see why I shoudl not believe it) is simply proving it beyond reasonable doubt.

The sceptical should argue against the validity of the first draw rigging. If they trust it to be a valid stat, they clearly have no other choice to accept that draws are rigged.

Yeah, the first study was even more surprising. Guess why it didn't attract this much of heated argument. That's right- because no particular player was involved.

Unfortunately, here we mention the top 4 players by name and immediately people think it's player bashing. It suddenly becomes about defending your player and logic goes out of the window.

If the study had said top 4 players players A,B, C D and not specified their names or the era, almost everyone would have agreed that 12/12 is surprising!

Tennisanorak wrote:Yeah, the first study was even more surprising. Guess why it didn't attract this much of heated argument. That's right- because no particular player was involved.

Unfortunately, here we mention the top 4 players by name and immediately people think it's player bashing. It suddenly becomes about defending your player and logic goes out of the window.

If the study had said top 4 players players A,B, C D and not specified their names or the era, almost everyone would have agreed that 12/12 is surprising!

Yes but it;s also interesting to see how people can be "scared" of being part of a conspiration theory. They would rather believe in amazing absurd draw stats than being seen as a conspirationist. That's why I said it's a good thread to read the minds.

I like the Mahut/Isner case even though there was only 1 in 96 chance to have it. Sure can be down to luck but I was already then very suspicious. Happened at Wimbledon not at the French or USO, right the following year, etc... 1 in 4096 simply is too good. 1/4k + 1/250k shoudl be a no brainer.

barrystar wrote:

Henman Bill wrote:What about the slams Nadal didn't even play and wasn't even in the draw, like Wimbledon 2009 (are there any others)? He wasn't even in the draw and yet that slam is used as one of the ones where the draw was supposedly fixed to create Fedal finals to get to the over 4000-1 odds!

But then the argument is..they weren't all fixed, just some. And yet still the ones that don't fit the fixing argument are included in the 4000-1 stats. While the FO can be discarded from those stats apparently.

Surely we have to either add the FO in to the stats or remove Wimbledon, reducing the numbers to way below the 4000-1 level (or 1 in 4096 level or whatever it was). Unless anyone can really explain why they would have fixed a Federer-Djokovic semi final in 2009 to create a Federer-Nadal final, in a tournament Nadal did not even enter?

Djoko and Fed were on the same side of the Wimbledon draw in 2009, and Nadal was drawn out of the hat that tournament on the opposite side, but his place in the draw was taken by JM de P so it does count.

Assuming your memory is correct here then fair enough, my point about Nadal 2009 can be ignored. Sorry for that. It was just a side point to my main point through, which is that fixing Nadal to avoid Djokovic at Wimbledon would not have give Nadal a significant enough advantage to make it worthwhile, although we can see Tenez making an argument the other way so we have arguments on both sides here.

Henman Bill wrote: although we can see Tenez making an argument the other way so we have arguments on both sides here.

Yes and beginning of 2008 Nadal had 3 FOs no slam outside. That is a bit more than "an other side" isn't it?

I am sure you realise it's much more likely that Nadal winning his first 3 slams at the FO were just down to chance (1 in 32) than getting 12/12.

SO that's my argument for the sceptics. I say that Nadal is no better on clay than anywhere else. He was just lucky that his first 3 slams were at the FO and only conspiracionists woudl say he was better on clay over 5 sets cause they have only 32/33 chance to be right.

Now, Laver fan. let's get back to the first example you quoted of something you thought was an improbable pattern. I quote you below:

"Federer vs Soderling

2010 USO Q Robin Soderling (SWE) 5 W 6-4, 6-4, 7-5
2010 FO Q Robin Soderling (SWE) 7 L 6-3, 3-6, 5-7, 4-6
2009 USO Q Robin Soderling (SWE) 12 W 6-0, 6-3, 6-7(6), 7-6(6)
2009 W R16 Robin Soderling (SWE) 12 W 6-4, 7-6(5), 7-6(5)
2009 FO W Robin Soderling (SWE) 25 W 6-1, 7-6(1), 6-4
2008 W R64 Robin Soderling (SWE) 41 W 6-3, 6-4, 7-6(3)"

First of all, we remove 2009 FO out of the picture because the two were on opposite sides of the draw. This is the most startling error you have made. We are looking it at how often they were drawn to play each other, so why did you include a tournament where the two met as late as possible, i.e. the final? It only weakens your hypothesis. We are left with the following 5 data points.

"Federer vs Soderling

2010 USO Q Robin Soderling (SWE) 5 W 6-4, 6-4, 7-5
2010 FO Q Robin Soderling (SWE) 7 L 6-3, 3-6, 5-7, 4-6
2009 USO Q Robin Soderling (SWE) 12 W 6-0, 6-3, 6-7(6), 7-6(6)
2009 W R16 Robin Soderling (SWE) 12 W 6-4, 7-6(5), 7-6(5)
2008 W R64 Robin Soderling (SWE) 41 W 6-3, 6-4, 7-6(3)"

2010 USO- 1/4th chance
2010 FO- 1/4th chance
2009 USO- 1/4th chance
2009 Wimbledon- 1/4th chance

Multiplying all these gives a probability of 1/256 which is much less than 1/4096.

This is further weakened by the fact that there are so many gaps in your data. Basically, you've just picked up a few points across years.

Do you really think this is the same as 4 continuous years in 3 different grandslams?

In spite of trying to spot a pattern among all possible players pairs and leaving out so many data points arbitrarily, the best you could come up with was a 1/256 chance, not even near a 1/4096 chance.

This is like my saying that Federer and Djokovic met 8 times between 2003 and 2009 (Probability of that is 1/256). There's nothing remarkable about that if they were not in a row.

Please do not mention leaving out the FO as an arbitrary selection. That is no way anywhere near as random as just picking some 5 data points across 3 years and ignoring others.

Tennisanorak wrote:Henman Bill, all I meant was the argument about the probability being one in 4k. I see this being contested again and again by saying that the draw is based on seedings. It doesn't matter. The probability is still 1 in 4000 of 12 Fed- Djoker matches. There is absolutely no debate about that.

What that means is open to debate. Whether excluding French Open is right is up for debate. But not the fact that the probability is 1 in 4000 itself. It's an elementary probability calculation.

But as long as people don't accept the calculation and find nothing abnormal, it is tough to proceed. That is the starting point of the argument.

On the draw/seedings issue I don't think that was me, so maybe you made a mistake there, if you were attributing that to me as part of your criticism that I don't get it.

In a sense you're right, it it an elementary probability calculation. But what if I throw H-T-H-H-H-T-T-H-H-T-H-H and then after the fact say that the odds of that exact sequence occuring are 4096-1. But wouldn't that be true of any sequence. Does that really make it a surprise result? What if I say that the odds of the 1,3,4,5,8,9,11 and 12th all being heads were 256-1, ignoring the times it were tails. Again, that is the correct odds but it is very selective.

The odds of anyone winning the lottery is negligibly small, but someone must. The odds on any of us having been born are astronomically low, and yet here we are. There is no question that many many things happen and therefore it is highly probable, even inevitable, that things of low probability will occur. The question here is whether the odds are long enough to defeat these things (given the number of years, events and sports it is quite inevitable that hundreds and thousand to one events will occur out of the many hundreds of thousands of things and premutations that there are) or whether or not the events are too selective, or not. Not sure if we agree on that?

Indeed TA. Federer and Soderling are top 10. Take any stat between 2 top 10s and you have 10 * 10 more chance to find weird stats. So 250th/100th is 1/2.5 nothing special about finding weird stats between 2 top 10.

HH I bet on a similar vein that you could take just about any group of selected tennis draws (after all nobody can deny that it was the researcher did) and finds some amazing statistical thread that links them .

Also didn't someone, somewhere once state "there are lies, dam lies and then there are statistics."?

Tennisanorak wrote:Now, Laver fan. let's get back to the first example you quoted of something you thought was an improbable pattern. I quote you below:
....
First of all, we remove 2009 FO out of the picture because the two were on opposite sides of the draw. This is the most startling error you have made. We are looking it at how often they were drawn to play each other, so why did you include a tournament where the two met as late as possible, i.e. the final?

We remove FO, because they are on the opposite side. OK. Which is what 12vs12 did anyway.

Tennisanorak wrote:It only weakens your hypothesis.

No, it does not, but let us continue...

Tennisanorak wrote:
2010 USO- 1/4th chance
2010 FO- 1/4th chance
2009 USO- 1/4th chance
2009 Wimbledon- 1/4th chance

Multiplying all these gives a probability of 1/256 which is much less than 1/4096.

OK.

Tennisanorak wrote:This is further weakened by the fact that there are so many gaps in your data. Basically, you've just picked up a few points across years.

Exactly what was done by discarding the FO in the 12vs12.

Tennisanorak wrote:Do you really think this is the same as 4 continuous years in 3 different grandslams?

To prove you are hypothesis, you choose 'discreteness' vs 'continuous' data with a subjective interpretation. Why discard the FO 2007-2011? Is FO not part of slams?

Tennisanorak wrote:In spite of trying to spot a pattern among all possible players pairs and leaving out so many data points arbitrarily, the best you could come up with was a 1/256 chance, not even near a 1/4096 chance.

This is like my saying that Federer and Djokovic met 8 times between 2003 and 2009 (Probability of that is 1/256). There's nothing remarkable about that if they were not in a row.

Please do not mention leaving out the FO as an arbitrary selection. That is no way anywhere near as random as just picking some 5 data points across 3 years and ignoring others.

The pattern is based on subjectivity. Look at the Del Potro example, USO 2009, has him in the other half. You can apply the same logic. The window is 2007-2011 based on the OP. SPuranik made the same argument of it being 4 years, not 5 years.

Henman Bill wrote:

Tennisanorak wrote:Henman Bill, all I meant was the argument about the probability being one in 4k. I see this being contested again and again by saying that the draw is based on seedings. It doesn't matter. The probability is still 1 in 4000 of 12 Fed- Djoker matches. There is absolutely no debate about that.

What that means is open to debate. Whether excluding French Open is right is up for debate. But not the fact that the probability is 1 in 4000 itself. It's an elementary probability calculation.

But as long as people don't accept the calculation and find nothing abnormal, it is tough to proceed. That is the starting point of the argument.

On the draw/seedings issue I don't think that was me, so maybe you made a mistake there, if you were attributing that to me as part of your criticism that I don't get it.

In a sense you're right, it it an elementary probability calculation. But what if I throw H-T-H-H-H-T-T-H-H-T-H-H and then after the fact say that the odds of that exact sequence occuring are 4096-1. But wouldn't that be true of any sequence. Does that really make it a surprise result? What if I say that the odds of the 1,3,4,5,8,9,11 and 12th all being heads were 256-1, ignoring the times it were tails. Again, that is the correct odds but it is very selective.

The odds of anyone winning the lottery is negligibly small, but someone must. The odds on any of us having been born are astronomically low, and yet here we are. There is no question that many many things happen and therefore it is highly probable, even inevitable, that things of low probability will occur. The question here is whether the odds are long enough to defeat these things (given the number of years, events and sports it is quite inevitable that hundreds and thousand to one events will occur out of the many hundreds of thousands of things and premutations that there are) or whether or not the events are too selective, or not. Not sure if we agree on that?

That's where you exposed your logical flaws I am afraid.


Yes, the odds times the number of throws makes things happen. You can throw a coin 4096 times and you might get one, 2 or even 3 times only tales in a row....or simply none.

billions of people mating a few 1000 times in their lives releasing billions of spermatozoids each time they mate will nott make us human a rarity. In fact if you noticed, we are even more than yesterday despite all the precaution we take nowadays not to multiply too much.

Now you have a player like Nadal who wants to control everything including his water bottles...and the controlled draws become a really oddly timed occurence...in the history of tennis.

Tenez wrote:
Again, you really don't know what you are talking about I am afraid.

Where is the logic in that statement?

newballs wrote:Also didn't someone, somewhere once state "there are lies, dam lies and then there are statistics."?

Didn't this guy get killed by a storm's thunderbolt?

laverfan wrote:

Tenez wrote:
Again, you really don't know what you are talking about I am afraid.

Where is the logic in that statement?

None. Just trust me.