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.

To be fair I think tennisanorak has made some good points here, but clearly the majority do not believe in this theory.

This would be my take on it:

Theory 1/ Coincidence and anomaly theory:
This is SF seedings through from 2008-2011 for AO, USO and Wimby.
1 vs. 4 and 2 vs. 3 a total of 7 times
1 vs. 3 and 2 vs. 4 a total of 5 times

Now there is 50% chance that it is 1vs 4 and 50% chance that it is 1vs 3.
So the equation to find the probability is:
(1/2 +1/2) to the power of 12.
So using nCr The probability of getting the 7 (1vs4) -5 (1vs3) sequence is- (^=to the power of)
0.5^7 x 0.5^5 x 12C7= 0.193359375
This is the joint second most likliest sequence, of course the most likely being that it is shared out 6 and 6.

So this theory is strong as the draw is actually done on seedings, so this sequence is entirely possible. The people arguing this theory will say that the Djokovic-Fed is a coincidence, and there is no reason to be suspicious.

Theory 2/ Conspiracy:
Now the people arguing this side of the story would point out that if it is possible that it was being rigged the 'SEEDS DON'T MATTER.' They would say that if the organisers wanted something they would get it done on the players they wanted, and with their seeds irrelevant as long as they are in the top 4.
This theory also has some evidence. The '12outof12' - of course relating to the Djokovic-Fed SF ratio is true. If the organisers were keen on having this semifinal then they could have organised it, irrelevant of the seedings. But then of course the people arguing on the other side will counter this and claim its a coincidence, and that if you look at the seed placement it will not be unusual.

Personally I'm not on either side here. The regular occurrence of the Djoko-Fed semi is eyebrow raising but looking at their seeds puts into perspective that this sequence is definitely possible. It depends on which way you look at it, but we can't prove there has been rigging, nor can we prove the opposite.

Benefits:
This is an interesting one. Some people have argued this, I feel, to prove their agenda against certain tennis players rather than be neutral.
One claim is that Nadal benefitted from not having Djokovic on his side of the draw between 2007-2010. Let's have a look at the facts. Nadal and Djokovic have been top 4 during this time so if they were in the same half they could have only met in the semis:
-Nadal has only reached 6 Grand Slam Hard Court Semifnals.
-Out of these 6 he played Djokovic in the finals anyway (2 of which they were seeded 1 and 2).
-In 2 of them he has gone on to lose, so there wasn't a possibility of him gaining anything from not facing Djoko in USO 2008+2009, the worst case for him was losing which he did to Del Pot and Murray anyway.
-The last one is the only time he could have possibly benefited- AO 2009. But here Nadal was tested in the SF against Verdasco and Djokovic had to retired due to heat so it was unlikely it would have made a big difference.
-Therefore we can say that even if the draws were rigged so that Nadal avoids Djokovic it hardly affected him. So the idea of this being a secret plot organised by Nadal is frankly ridiculous. Remember Djokovic was not as good before.
-Meanwhile if not anything Federer probably benefitted from it. Before his resurgence in late 2010, Djokovic lost 100% of the slam matches against Federer (apart from 2008 when Fed had mono). In fact Murray showed more signs he could have harmed Federer, but let's be honest he would have struggled against Fed in slams anyway.

Anyway overall I think Tenez and Tennisanorak have made some good points but they have been countered well. As I have shown it didn't really help Nadal that anyway so I'm not sure we can say he is behind this, if it was rigged.

hawkeye wrote:

Tenez wrote:

hawkeye wrote:Why don't they "fix" the draw for Murray at Wimbledon? "They" certainly don't want Nadal rather than Murray in the final there.

As far as we know Slam organisers know that Murray can beat Nadal in slams. Murray beating Federer as yet to happen.

Tenez

So your theory is that the draws are fixed to benefit both Nadal and Federer? But not at the FO as your "evidence doesn't cover this slam... And also not at Wimbledon... Because here they want Murray to win and not Nadal so they put Murray in Rafa's half...?

I hope these draw fixers concentrate because it would be very easy to put Federer and Djokovic in the same half for all the wrong reasons and then they would go and spoil everything...

The study DOES cover Wimbledon, why don't you listen to the link before commenting? The study covers Wimbledon, AO and USO from 2008 to 2011 that's 12 slams and in all of those Djokovic was in Federer's half.

barrystar wrote:

@Tenez - are you saying that at RG they'd put Djoko in Nadal's half to try and stop Nadal making the final?

I am not saying that. I am just saying that the French simply ran a normal draw cause they had no intention or motives "to sort it out". They had their share of Fedals and got lots of success through it. Their draw shows a 50/50 of having Djoko on either side in that period. A normal occurence. In that respect not considering the FO makes perfect sense. As mentioned it adds to the case for rigging if anything.

Tenez wrote:

I am not saying that. I am just saying that the French simply ran a normal draw cause they had no intention or motives "to sort it out". They had their share of Fedals and got lots of success through it. Their draw shows a 50/50 of having Djoko on either side in that period. A normal occurence. In that respect not considering the FO makes perfect sense. As mentioned it adds to the case for rigging if anything.

Interesting. The French are an anomaly then according to you, the only ones who don't cheat.
Are you sure you are neutral here Tenez?

Amritia3ee* wrote:

Tenez wrote:

I am not saying that. I am just saying that the French simply ran a normal draw cause they had no intention or motives "to sort it out". They had their share of Fedals and got lots of success through it. Their draw shows a 50/50 of having Djoko on either side in that period. A normal occurence. In that respect not considering the FO makes perfect sense. As mentioned it adds to the case for rigging if anything.

Interesting. The French are an anomaly then according to you, the only ones who don't cheat.
Are you sure you are neutral here Tenez?

what's nationality got to do with the discussion, stay on the topic please.

noleisthebest

Of course I realise that this "Official study" covers Wimbledon! But at Wimbledon as Tenez pointed out the draw is fixed so that poor Rafa loses in the semi to Murray. How is that draw fix helpful to poor Rafa?

I'm sure both Federer and Djokovic are delighted with this fix though. They would much rather play Murray than that dastardly Nadal yet again...

I'm sorry NITB.
But I still don't we can miss out the French Open from the data as they are 'more trustworthy', because they don't fit in to the conspiracy theory, no?

legendkillarV2 wrote:

Tenez wrote:

legendkillarV2 wrote:There is a fine line difference to proof and circumstantial statistics.

I said solid proof. Yes there is a chance it;s down to chance. By calling a 1 in 4k chance a "gaping hole" is laughable.

Missing other Slam events in the year is gaping and yes laughable.

Not it's not. You coudl take just the USO and AO it woudl still be a weird occurence.

As I said earlier, the chance for an Isner/Mahut in WImby 2011 after their epic 2010 is 1 in 96. Really dodgy if you ask me....but possible. Having the AO and USO sorting Fed/Djoko for 4 years running is 1/254 extremely unlikely. 1 in 4096 is ...well if you don;t want to see it fine....but gaping hole. No.

Tenez do you agree with 'Theory 2' at the top of Page 5.
I must admit you have made some good points here, but I'm still going to support LK et co. as i believe they have a slightly stronger case.

hawkeye wrote:noleisthebest

Of course I realise that this "Official study" covers Wimbledon! But at Wimbledon as Tenez pointed out the draw is fixed so that poor Rafa loses in the semi to Murray. How is that draw fix helpful to poor Rafa?

I'm sure both Federer and Djokovic are delighted with this fix though. They would much rather play Murray than that dastardly Nadal yet again...

For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams.
Fed is obviously a miles better player on hard courts as well as grass, so naturally, it was Nadal who needed a bit of a push.

Nadal has said that he found Djokovic most difficult to play (and recent 7 losses prove that he knew what he was talking about), Federer does not find Djokovic that difficult to beat, so naturally Djokovic was going to end up in Federer's half.
Nadal has little problem beating Murray, compared to beating Djokovic.
I can't be any simpler and more obvious than this.

Tenez did you also my stats point in 'Theory 1'.
The chance of the seed permutation happening was 0.19, which was the second most likliest possibility. What do you say about that?
I'm not saying this in an aggressive way, i'm just genuinely interested as I'm not sure you've countered that point yet.

Amritia3ee* wrote:Tenez did you also my stats point in 'Theory 1'.
The chance of the seed permutation happening was 0.19, which was the second most likliest possibility. What do you say about that?
I'm not saying this in an aggressive way, i'm just genuinely interested as I'm not sure you've countered that point yet.

he'll ignore it, as it doesn't suit his theory, much as these students ignored anything that didn't suit theirs...

Amritia3ee* wrote:Tenez do you agree with 'Theory 2' at the top of Page 5.
I must admit you have made some good points here, but I'm still going to support LK et co. as i believe they have a slightly stronger case.

A stronger case would be a case that shows 1/4096 is a normal oddity when considering top 4 players draws over 4 years. I have not read or seen anything like this thus far.

Amritia3ee* wrote:Tenez did you also my stats point in 'Theory 1'.
The chance of the seed permutation happening was 0.19, which was the second most likliest possibility. What do you say about that?
I'm not saying this in an aggressive way, i'm just genuinely interested as I'm not sure you've countered that point yet.

If you are serious about rigging the draw to avoid Federer/Murray semis or Djoko/Nadals, you are not going to look at 1/3 1/4 2/3 2/4. You want Djoko and Nadal on opposite sides. As simple as that.

I have checked quickly to see whether we had recording of those draws on youtube or else...and strangely we have not. Weird isn't it?

Tenez wrote:

Amritia3ee* wrote:Tenez do you agree with 'Theory 2' at the top of Page 5.
I must admit you have made some good points here, but I'm still going to support LK et co. as i believe they have a slightly stronger case.

A stronger case would be a case that shows 1/4096 is a normal oddity when considering top 4 players draws over 4 years. I have not read or seen anything like this thus far.

But isn't it the case that any sequence of those 12 draws would have had a 1/4096 probability?

amritia, you wont get the answers you're looking for, there is only one story to be had for the "theorists" and if a point is made that counters the theory its discarded and 1/4000 thrown back at you...even though the seed placements were never aligned 12/12 (1 in 4096)...they keep focusing on names, not seeds (names move around rankings...seeds never move per se)...this is the basic premise that would have the analysis laughed out of any stats journal submission.

noleisthebest wrote:

hawkeye wrote:noleisthebest

Of course I realise that this "Official study" covers Wimbledon! But at Wimbledon as Tenez pointed out the draw is fixed so that poor Rafa loses in the semi to Murray. How is that draw fix helpful to poor Rafa?

I'm sure both Federer and Djokovic are delighted with this fix though. They would much rather play Murray than that dastardly Nadal yet again...

For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams.
Fed is obviously a miles better player on hard courts as well as grass, so naturally, it was Nadal who needed a bit of a push.

Nadal has said that he found Djokovic most difficult to play (and recent 7 losses prove that he knew what he was talking about), Federer does not find Djokovic that difficult to beat, so naturally Djokovic was going to end up in Federer's half.
Nadal has little problem beating Murray, compared to beating Djokovic.
I can't be any simpler and more obvious than this.

I think you may have got a little confused. Not surprising as this is such a complex subject. Don't worry this has nothing to do with maths though. It's about exactly what the fixers are trying to fix. At Wimbledon as Tenez points out they don't want to fix a Nadal Federer final. This is what Tenez had to say earlier in this thread

"As far as we know Slam organisers know that Murray can beat Nadal in slams. Murray beating Federer as yet to happen."

That's why they put Murray in Nadals half so Murray can beat Nadal. This just applies at Wimbledon.

Now that I think about it I'm not sure about who the fixers would like to win the other semi. What do you think? Federer because the fixers always want him in the final? Or Djokovic because Murray has a better chance of beating him in the final? It would be a waste of time fixing a Murray win over Nadal in the semi's if he is to lose in the final...

I did say it was a complex subject... But there is no point doing all those complex calculations involving statistics and probability if your not sure exactly what it is your trying to fix.

noleisthebest wrote:For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams

For the umpteenth time the seeds placement distribution ratios dont fit your 'theory'...its all conspirational conjecture and supposition!

lydian wrote:

noleisthebest wrote:For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams

For the umpteenth time the seeds placement distribution ratios dont fit your 'theory'...its all conspirational conjecture and supposition!

of course they do.

barrystar wrote:

Tenez wrote:

Amritia3ee* wrote:Tenez do you agree with 'Theory 2' at the top of Page 5.
I must admit you have made some good points here, but I'm still going to support LK et co. as i believe they have a slightly stronger case.

A stronger case would be a case that shows 1/4096 is a normal oddity when considering top 4 players draws over 4 years. I have not read or seen anything like this thus far.

But isn't it the case that any sequence of those 12 draws would have had a 1/4096 probability?

Any sequence yes but i am sure you see the difference beteen a sequence a set of results. Having Djoko and Federer on the same side has a 50% chance in theory. The more you throw the coin (the more draw you have, the closer you should get to those 50% value. Here we hace 0.02% which is extremely small compared to 50%.

lydian wrote:

noleisthebest wrote:For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams

For the umpteenth time the seeds placement distribution ratios dont fit your 'theory'...its all conspirational conjecture and supposition!

Probably the most ridiculous post. Even Amri has some understanding of the issue. It reminds me one of your ridiculous assumption once on old 606: A 55g ball weight is irrelevant to the trajectory of the ball when hit with such power by modern players.

noleisthebest wrote:of course they do.

They do when you want them to.
But as hawkeye observes, besides the stats abberations, the premise for who TDs would want to face who is even open for conjecture. You see your theory is based on a rock solid NADAL MUST AVOID DJOKOVIC at all costs premise. However...until 2011 there was no reason to see Djokovic as a threat to Nadal...Nadal's H2H vs Nole was better than Federer's H2H vs Nole - but broadly similar actually. They might as well as fixed it either way! Either way your "theory" doesnt stack up...

Tenez wrote:Probably the most ridiculous post.

lol
Ridiculous because you have no counter to it...because it doesnt suit your conspiracy theory

lydian wrote:

Tenez wrote:Probably the most ridiculous post.

lol
Ridiculous because you have no counter to it...because it doesnt suit your conspiracy theory

counter what? you keep losing the arguments one after the other!

List one argument you won against me!

Tenez/NITB, do you think the fixing will continue or do you think it's stopped now?

lydian wrote:

noleisthebest wrote:For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams

For the umpteenth time the seeds placement distribution ratios dont fit your 'theory'...its all conspirational conjecture and supposition!

I think you are being a little unfair here. What they are saying is that the "seeds placement distribution" which should be random marched precisely in time with something that was not random, the actual seeding of Federer and Djoko.

There are two ways of looking at that:

* it's a coincidence - which is my preference because I don't go for conspiracies in the absence of overwhelming evidence

* it tends to show that the "seeds placement distribution" was not random but influenced to match precisely the seedings of individual players to produce the desired result of preventing Nadal from having to face Djoko before the final and thus increasing the chance of a Fedal final.

I don't think that's a completely absurd suggestion, even though I strongly disagree with it.

Another reason why I think the conspiracy does not work is that it tends to view a 4 year period as if it conformed uniformly to a pattern which really only existed for parts of those 4 years.

Djoko has always been a fairly tough opponent for Nadal, even according to Nadal himself, but until the beginning of 2011 Nadal had a 16:7 H2H vs. Djoko, consisting of 9-0 on clay, 2-0 on grass, and 5-7 on hard. Djoko had never beaten Nadal in a best-of-five match and of his 7 hard wins 4 were on the slick surfaces of Paris, TMF, and Cincinnatti. Prior to 2011 Nadal's record vs. Murray was 9:4, 2-0 on grass, 2-0 on clay, and 5-4 on hard. Murray had twice beaten Nadal at an HC slam and pushed him very hard at the Australian Open in 2007. It is not the case that someone trying to rig a draw would have concluded at the start of every slam in 2007-2011 apart from RG that Nadal's best chance would involve Djoko being on the other side of the draw. At Wimbledon it would have made no difference such was his superiority on grass over either of them, and on hard there was next to nothing between either of them. Of course, I recognise that Murray has not always been No. 4 in this period, but nor has Djoko always been the most dangerous opponent for Nadal outside Fedal either.

Yeah course I do. Oh dear Tenez...is that it then? It boils down to a "my dad is bigger than your dad" type argument? ...purlease. As for specifics, there have been many but I'm not going down that road for obvious reasons - this forum doesnt need it. We both know you are completely driven by the need to discredit Nadal. Fair enough, thats your raison d'etre...just dont expect others to fall in line behind you. Viva la difference, viva la debate...

Henman Bill wrote:Tenez/NITB, do you think the fixing will continue or do you think it's stopped now?

Why would you expect us to know? We have just found a weird pattern and more concerning is the fact that top seeds at the USO get lower ranked players in their first round. DO you deny that 1 in 250k chance as well? This shows draws are rigged. There is no reason to believe they are going to stop. What is more difficult to predict is who is going to benefit and whether it's worse taking the risk as it seems it has been for the last 4 years.

lydian wrote:Yeah course I do. Oh dear Tenez...is that it then? It boils down to a "my dad is bigger than your dad" type argument? ...purlease. As for specifics, there have been many but I'm not going down that road for obvious reasons - this forum doesnt need it. We both know you are completely driven by the need to discredit Nadal. Fair enough, thats your raison d'etre...just dont expect others to fall in line behind you. Viva la difference, viva la debate...

Talking cliche as usual...but no substance as usual.

barrystar...I dont think the conspiracy theory is unfair per se, I just dont see it supported by proper evidence. The seeds placement is what it is. You either want to make that leap of faith and believe its all rigged or you dont.

Djoko has always been a tough opponent for Federer too dont forget.

Tenez wrote:Talking cliche as usual...but no substance as usual.

"W" - whatever...zzzzzzz

barrystar wrote:

lydian wrote:

noleisthebest wrote:For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams

For the umpteenth time the seeds placement distribution ratios dont fit your 'theory'...its all conspirational conjecture and supposition!

I think you are being a little unfair here. What they are saying is that the "seeds placement distribution" which should be random marched precisely in time with something that was not random, the actual seeding of Federer and Djoko.

There are two ways of looking at that:

* it's a coincidence - which is my preference because I don't go for conspiracies in the absence of overwhelming evidence

* it tends to show that the "seeds placement distribution" was not random but influenced to match precisely the seedings of individual players to produce the desired result of preventing Nadal from having to face Djoko before the final and thus increasing the chance of a Fedal final.

I don't think that's a completely absurd suggestion, even though I strongly disagree with it.

Another reason why I think the conspiracy does not work is that it tends to view a 4 year period as if it conformed uniformly to a pattern which really only existed for parts of those 4 years.

Djoko has always been a fairly tough opponent for Nadal, even according to Nadal himself, but until the beginning of 2011 Nadal had a 16:7 H2H vs. Djoko, consisting of 9-0 on clay, 2-0 on grass, and 5-7 on hard. Djoko had never beaten Nadal in a best-of-five match and of his 7 hard wins 4 were on the slick surfaces of Paris, TMF, and Cincinnatti. Prior to 2011 Nadal's record vs. Murray was 9:4, 2-0 on grass, 2-0 on clay, and 5-4 on hard. Murray had twice beaten Nadal at an HC slam and pushed him very hard at the Australian Open in 2007. It is not the case that someone trying to rig a draw would have concluded at the start of every slam in 2007-2011 apart from RG that Nadal's best chance would involve Djoko being on the other side of the draw. At Wimbledon it would have made no difference such was his superiority on grass over either of them, and on hard there was next to nothing between either of them. Of course, I recognise that Murray has not always been No. 4 in this period, but nor has Djoko always been the most dangerous opponent for Nadal outside Fedal either.

I answered some of this in previous posts you have not read it seems but there was no reason at all to rig 2007 as Djoko was really a non existent entity then. However he was already showing his potential (almost beating Nadal at SW19 before being injured) and was young so everybody knew he was going to get stronger and Djoko started 2008 with a bang!

And if you don;t believe they are rigged how do you explain the low ranking players for the top seeds (one in 250k chance!!!).

I am aurprised by your reasoning here which is usually pretty good. As I said many times in my earlier posts I would not believe in 1/4000 chance but there would still be some doubts in my mind. The fact they are clearly rigging the draws for the first rounds gives me almost 0 doubt for the top 4 seeds.

Tenez wrote:And if you don;t believe they are rigged how do you explain the low ranking players for the top seeds (one in 250k chance!!!).

The ESPN study is purely based on USO. The current 'conspiracy' is based on 12 slams.

Picking arbitrary data to 'prove' your theory.

The South Korean, who cloned in the lab comes to mind.

Very sound post from Barrystar, not least because it's similar to much of what I said.

The US Open early rounds draw I haven't looked at yet so my comments are only on the other discussion. It could be more interesting. I do recall Roger Federer drawing dangerous players floating low in the rankings like Nalbandian, Henman and Hewitt quite a few times in the first 3 rounds of slams so top players don't get it all their own way all the time. I mean Henman in R2 at Wimbledon in what, 2006, was about as hard a draw as you could get for R2 at Wimbledon. Mind you, he still schooled him.

Mind you, what difference does it make. When Federer drew Henman in 2006 he probably thought "Crap, now I'll have to get to 90% for R2 instead of cruise controlling through to R4 in third gear".

By the way, he had Gasquet in R1. Gasquet was extremely well known as an up and coming young player. Hardly the kind of player you would want to give someone if you wanted to fix someone an easy path. He went on to reach the Canada final later that year (winning the first set 6-2 against Federer before losing) and US Open R4 and finished the year ranked 18.

I feel vindicated, excellent research Nitb. By the way I posted a study finding the exact same thing done by ESPN statisticians and posted it last year. Now this is the second study we have seen by mathemeticians finding that the draws at ATP tournaments are not random. Once it is established that the draws aren't random they are therefore manufactured. I am glad I came back online today to see this. I remember the howling from all corners dismissing my theories as idle consipiracy theories. Now we have two studies that confirm the facts.

The draws have been fixed for years to put Novak into Roger's half in the grandslams. This was done so that slams could protect a fedal final. But now it is interesting to see how they are going to start tampering with it now that Novak is not an easy touch in the semi by any stretch of the imagination. What is interesting is that now Novak continually is getting Andy Murray this year.

I love how Tenez jumps on board to support this study but I produced a study coming to a similar conclusion and at the time tenez was part of those downplaying my comments.

Barry star- ah, you put it beautifully. You are a star! : )

That is exactly it for the layman. How can a random pairing of seeds follow a non-random seeding of Federer and Djokovic 12 times in a row?

From there on, you can believe it's an incredible coincidence or that there is something going on.

In fact, given any low probability event, these are the only two ways to reason it out

(a) That a miraculous event happened

(b) That the probabilities were wrongly estimated.

You people seem to prefer option (a)

Many will go with (b) here, i.e. that the probabilities were wrongly calculated. In other words, it really wasn't random with a probability of 1/2.

Just out of curiosity, if this had happened 50 times in a row rather than 12 times, would people still have called it a coincidence? So what is the threshold when people stop calling it a coincidence and question things? I'm really curious to know this.

socal1976 wrote:I feel vindicated, excellent research Nitb. By the way I posted a study finding the exact same thing done by ESPN statisticians and posted it last year. Now this is the second study we have seen by mathemeticians finding that the draws at ATP tournaments are not random. Once it is established that the draws aren't random they are therefore manufactured. I am glad I came back online today to see this. I remember the howling from all corners dismissing my theories as idle consipiracy theories. Now we have two studies that confirm the facts.

The draws have been fixed for years to put Novak into Roger's half in the grandslams. This was done so that slams could protect a fedal final. But now it is interesting to see how they are going to start tampering with it now that Novak is not an easy touch in the semi by any stretch of the imagination. What is interesting is that now Novak continually is getting Andy Murray this year.

I love how Tenez jumps on board to support this study but I produced a study coming to a similar conclusion and at the time tenez was part of those downplaying my comments.

I don't remember the ESPN thread, I probably was "in exile". Will look it up.

As for this whole player- seeds debate, here's an illustration.

Let’s say that for four consecutive grand slams. Federer was seeded 1,2, 3 and 4 and Nalbandian 33, 48, 59 and 72.

Let’s say the draw put 1 Vs 33, 2 Vs 48, 3 Vs 59 and 4 Vs 72.

Seedings-wise, that’s okay, but player-wise, that would be Federer Vs Nalbandian in 4 continuous slams. The probability would be less than one in a million of this happening, around 1 in 100 million in fact! (Each of those has a likelihood of 1/96 since a seeded player can meet any of 96 unseeded players. So we're looking at 1/ 96 to the power of 4).

I want to know what your reaction would be if this actually happens. Would you say

(a) Nothing wrong- the seedings were perfectly random and the draws are random.

(b) Very fishy- the draws kept following the seedings of Federer and Nalbandian.

This might seem an extreme illustration, but it is identical to the argument we are making for Federer and Djokovic.

People, please read this carefully before replying. I do not have an agenda. I just am curious to see how people think about this issue.

laverfan wrote:

Tenez wrote:And if you don;t believe they are rigged how do you explain the low ranking players for the top seeds (one in 250k chance!!!).

The ESPN study is purely based on USO. The current 'conspiracy' is based on 12 slams.

Picking arbitrary data to 'prove' your theory.

The South Korean, who cloned in the lab comes to mind.

Wrong Laverfan they also found that the draw in the women's side of RG was very fishy as well. And the Espn study if we are referring to the same one I put up last year only looked at the first two rounds of the draw it did not address other areas in the draw that may or may not have been tampered. There is no surprise that tournament directors have tried their best to favor the interests of themselves and by extension fedal has benefitted.

It is now looking like all those people who called me a conspiracy theorist now are on the back foot. This is the second such piece of research we have found now that sheds serious doubt on the randomness of the draw. I am sure the tournament organizers who really are not regulated or auditted in this process are not beyond mixing and matching the draws here and there within reason to help them maximize potential ratings and matchups.

Nitb, I posted it a while back under something like "PHd statistician finds draws are manufactured". You should look it up because two different groups of statisticians looked at the draws of the Grandslam over the last few years, they focused it only on the first two rounds. They found that the USO mens and womens draw and the RG women's draw had a very high likelihood of not being random and therefore tampered with. The study did not directly address the years that Novak was continually put into fed's half over and over again. The study found that the top seeds got easier first two round draws then they otherwise would have had if the draws had been randomly assigned.


This is what I have been saying, you have financial interests at stake, no oversight of the process, and of course these guys are going to try to play with the process to maximize their returns.

Or here's another example.

Say at the beginning of Australian open, Federer is seeded 1 and Berdych seeded 7.

Probability of Berdych being in Federer's quarter is 1/4 (any of seeds 7-8 can be drawn to meet Federer).

When the draws come out, you find that Berdych is drawn to play Federer.

No big deal.

Federer remains #1 at French Open. Berdych slips to #8.

The draws come out. #1 is drawn to play #8. Federer- Berdych again. No big dea stilll, though it's a 1/4 probability event twice in a row.

Wimbledon- Federer #1 Berdych # 6. Again drawn to play each other.

USO Federer #1 Berdych # 5 same story.

Australia Open Fed #1 Berdych # 6 They play again

French Open Fed #1 Berdych # 7 they play yet again.

Now statistically the #1 seed has drawn Numbers 7,8,6,5, 6 and 7 in consecutive slams. No pattern there.

Yet player-wise, the seedings keep following the same two players.

Tell me- would you find it bizarre or say that it is fine if you look at it seedings-wise?

The probability of this hypothetical event is 1/4 to the power of 6, which is exactly the same as 1/4096 in the case of Federer- Djokovic playing 12 times.

Only, in this case, it involves Federer and Djokovic who are much more prominent and much more discussed than Federer and Berdych.

And to think some people think this is a conspiracy theory when it is a genuine expression of incredulity and surprise!

TA - if the random "seeds placement distribution" marched exactly hand in-hand with the seedings of two particular top players for 50 slams so that they were on the same (or opposite) sides of the draw every time I'd think that the onus of explaining that the draw had been random would sit far more firmly on the organisers than it does with just 12 slams.

On the current evidence I'm unimpressed that RG has been excluded, I can't see any reason for keeping Nadal away from Djokovic at Wimbledon, there is a feasible explanation that changing permutations in the seedings of the top 3 players have been random for 12 out of 16 slams, and the alternative of officials lying about what they have been doing for a fairly limited gain prospect is far less feasible to me. Remember, there's no room for ifs and buts if the draws were fixed, several people across 3 of the most prestigious tournaments in the game have been acting dishonestly and directly contrary to what they say that they do (a huge downside if they are nailed) against a relatively small upside.

I really need overwhelming evidence to believe that people with so much to lose would behave in such a short-sighted way for such little gain.

socal1976 wrote:Nitb, I posted it a while back under something like "PHd statistician finds draws are manufactured". You should look it up because two different groups of statisticians looked at the draws of the Grandslam over the last few years, they focused it only on the first two rounds. They found that the USO mens and womens draw and the RG women's draw had a very high likelihood of not being random and therefore tampered with. The study did not directly address the years that Novak was continually put into fed's half over and over again. The study found that the top seeds got easier first two round draws then they otherwise would have had if the draws had been randomly assigned.


This is what I have been saying, you have financial interests at stake, no oversight of the process, and of course these guys are going to try to play with the process to maximize their returns.

OK.
This study has a completely different focus and a more thorough one. It attempts to show how business has been ruining tennis, firstly by creating a financially lucrative rivalry, then propping it up to such an extent that 2 year reanking was at stake, draws fixed, courts slowed down and one player (happens to be my favourite one ) had to had it extra hard to overcome it all.
A few other bits and bobs, but these are the main ones. You have to agree, pretty serious stuff with far-reaching consequences.

The question I am almost scared to ask is: if and how much players knew about it and had their say in it all. I suppose that's the bit we never get to find out unless they chose to unburden their consciences like Agassi did with his drug abuse.

Well, Barrystar, it is a question of where you draw the line. To me, a probability of 1 in 4096 is low enough to start questioning things, especially when coupled with another study by ESPN that shows that the first round draws in US Open were random only with a probability of, what, 3 in a million?

If the first round computer generated draw is not random (and the one in 3 million probability is surely low enough to conclude that), why can't the handpicked seeded draws not be random as well?

This isn't proof, but it has reached the point where you've got to at least wonder whether something is going on. From that point to a proof is of course a long way!

barrystar wrote:TA - if the random "seeds placement distribution" marched exactly hand in-hand with the seedings of two particular top players for 50 slams so that they were on the same (or opposite) sides of the draw every time I'd think that the onus of explaining that the draw had been random would sit far more firmly on the organisers than it does with just 12 slams.

On the current evidence I'm unimpressed that RG has been excluded, I can't see any reason for keeping Nadal away from Djokovic at Wimbledon, there is a feasible explanation that changing permutations in the seedings of the top 3 players have been random for 12 out of 16 slams, and the alternative of officials lying about what they have been doing for a fairly limited gain prospect is far less feasible to me. Remember, there's no room for ifs and buts if the draws were fixed, several people across 3 of the most prestigious tournaments in the game have been acting dishonestly and directly contrary to what they say that they do (a huge downside if they are nailed) against a relatively small upside.

I really need overwhelming evidence to believe that people with so much to lose would behave in such a short-sighted way for such little gain.

who do you expect to lose and how much to think it's "so much"? Corruption on high levels is not a new thing.

TA... both your 'constructions' are data chosen to fit your player-affinity theory. This is the same case as Federer-Djokovic, despite rankings movement, and in the Federer-Djokovic case, just within Top 4 for almost 5 years.

Your distribution is reversing the 1/4096 event and showing that it will occur.

Laver fan- unfortunately, that just seems to weaken your argument. Yeah, those were constructions. But in the other case, I had a theory and the data fits it to a T. The theory has the same probability as the construction!

I again don't agree with hand picked. I am not picking the Masters series, am I? I am choosing the two players first and picking grand slam data for four years, from 2008 to 2011!

Your only valid point is the French Open draw. But I've already given you the reason why I think it's a valid exception. The relative strengths of the players were perceived to be different on clay. Nadal could defeat anyone on clay, so who he played in the semis didn't matter. Same for Federer as well. This is borne out by history too. The only player to beat either Federer or Nadal at RG for the past 7 years has been Soderling. That is how dominant the two have been.

Clearly, there is no need to give either a favourable draw at the French Open.

So there you go. I choose my two players, both very prominent (three if you include Murray), I form a hypothesis and I find that data for the past 4 years supports my hypothesis. The probability of this happening is 1/4096 if the draws really are random.

Where is this selective data which was chosen? if you think selective choosing of data is so easy, please give me an alternative hypothesis involving players of similar calibre ( so please don't pick stuff like "Monaco and Troicki played in the second round in 3 consecutive slams") and choose suitable data to justify your hypothesis. The hypothesis must have a probability of 1/4096 of happening.

If you can do that, I'll accept that my argument was flawed.

I await your reply.

Excellent post by Tennisanorak and Nitb, it is funny how people want to just pretend that there is no way that those with a vested financial interest and zero accountability would never stoop so low as to tamper with the draw.

At this point there is a strong rational basis for further investigation and changing of the draw process. Handing it to 3rd party with real oversight and transparency. Something that frankly we have literally none of the tennis tour.

noleisthebest wrote:

hawkeye wrote:noleisthebest

Of course I realise that this "Official study" covers Wimbledon! But at Wimbledon as Tenez pointed out the draw is fixed so that poor Rafa loses in the semi to Murray. How is that draw fix helpful to poor Rafa?

I'm sure both Federer and Djokovic are delighted with this fix though. They would much rather play Murray than that dastardly Nadal yet again...

For the upteenth time, the draw "arrangement" was to ensure (as much as possible) that Federer and Nadal meet in finals at slams.
Fed is obviously a miles better player on hard courts as well as grass, so naturally, it was Nadal who needed a bit of a push.

Nadal has said that he found Djokovic most difficult to play (and recent 7 losses prove that he knew what he was talking about), Federer does not find Djokovic that difficult to beat, so naturally Djokovic was going to end up in Federer's half.
Nadal has little problem beating Murray, compared to beating Djokovic.
I can't be any simpler and more obvious than this.

Sorry, no offence meant, but I've highlighted the flaw in this argument. The highlighted sentence is correct for 2011, but the data under study is for 2008-2011. And since supposedly the fixing occurred throughout that period, 2011 trends are hardly relevant to the supposed decision to fix, since the alleged fixing would have been established in 2008 and 2009.

In 2008-2010 Djokovic NEVER lost to Nadal in a slam. But Murray beat him in 2008 at the US Open, and also added another victory (due to injury) at Australian Open 2010.

As for Nadal "has said that he found Djokovic most difficult to play (and recent 7 losses prove that he knew what he was talking about" I think we need to clarify that to begin with Nadal was racking up win after win against Djokovic. Djokovic has become a better player now. Most players have worse records against him now.

Since we are keen to avoid clay, I bet if you take Murray and Djokovic's ranking points for 2008, 2009 and 2010 excluding clay, that they would be about the same. Apart from clay, they were on a par in those years.