Draw Fixing : A Real World Analysis - Part 1 (The 12 vs 12 Question)

by **laverfan** Tue 24 Apr 2012, 10:46 pm

First topic message reminder :

The fall out from Draw Fixing: An Official Study on 606v2 has created a deep divide. This followup article does not intend to cause any further . Please respect fellow posters' opinions and pursue a civilised debate.

Draws from 1974-2000 (post Open Era) for all four Slams were used. The data from 2001-2010 was not used to avoid conflicts with the ESPN study as well as look at an independent data set.

The Top 4 of each slam who were seeded 1 through 4 were considered.

The Top 4 are distributed where each half has seeds 1 and 2 at the opposite ends of the draw (64 or 128 players). Seeds 3 and 4 are placed to yield a 1-3, 2-4 or 1-4, 2-3 combination in each half.

1. There was some contention in the earlier article about the distribution of seeds #3 and #4 across the draws 'randomly'.

a. Australian Open (1974-2000) has 12 occurrences of 1-3, 2-4 (44%) combination, and 15 occurrences of 1-4, 2-3 (56%).

(AO had no 1986 Slam, but had two in 1977 (January and December). AO also had 'BYEs' for the Top 4 from 1983-1987).

b. French Open (1974-2000) has 15 occurrences of 1-3, 2-4 (56%) combination, and 12 occurrences of 1-4, 2-3 (44%).

c. Wimbledon (1974-2000) has 14 occurrences of 1-3, 2-4 (52%) combination, and 13 occurrences of 1-4, 2-3 (48%).

d. US Open (1974-2000) has 11 occurrences of 1-3, 2-4 (41%) combination, and 16 occurrences of 1-4, 2-3(59%).

2. Federer vs Djokovic (12vs12) raised a followup question whether similar scenario has happened before. The answer is yes, it has.

a. Connors vs McEnroe (9 projected meetings)

1979 - USO
1980 - FO, W, USO
1981 - FO
1983 - USO
1984 - FO, USO
1985 - W

b. Connors vs Vilas (7 projected meetings)

1975 - W
1976 - USO
1978 - USO
1979 - FO
1982 - FO, USO
1983 - W

c. Edberg v Lendl (9 projected meetings)

1986 - USO
1987 - FO, W
1989 - AO, USO
1990 - W, USO
1991 - AO, FO

d. Edberg v Stich (6 projected meetings)

1991 - USO
1992 - FO, W
1994 - AO, FO
1995 - AO

e. Edberg v Wilander (6 projected meetings)

1987 - USO
1988 - AO, FO, W, USO
1989 - W

The case of Connors is unique because Connors v Vilas is spread over a 8 year span, as an example. If USO wanted Connors all the time, which is understandable, FO and W also wanted a piece of the 'possibly rigging' pie.

Primary sources of data are ATP, ITF and Wikipedia. Spot validation was done using the AO draws from the official website and ITF. Wikipedia has some issues with specific AO draws, which were checked with the official site. I have not yet looked Sampras closely who was very dominant during his tenure.

Please point out any errors. E&OE (as I always indicate). This will be posted on another website that I visit (for full disclosure).

by lydian Fri 27 Apr 2012, 4:09 pm

Hey good work reckoner...you're obviously an emoticon king Smile

- oh sorry it should be king

by reckoner Fri 27 Apr 2012, 4:27 pm

FINALLY I am king of 606! king

by **JuliusHMarx** Fri 27 Apr 2012, 4:30 pm

reckoner wrote:FINALLY I am king of 606!

I am so tempted to delete that post just because I can Smile

(Just kidding)

by lydian Fri 27 Apr 2012, 4:37 pm

Like the order of the emotes there...!

Could do with a beer too - oh to heck with it, an after work jar beckons!

by reckoner Fri 27 Apr 2012, 4:45 pm

JuliusHMarx wrote:
reckoner wrote:FINALLY I am king of 606!

I am so tempted to delete that post just because I can
(Just kidding)

You can't delete the king on a Friday!!! It'd be... unconstitutional or something.

by reckoner Fri 27 Apr 2012, 4:46 pm

lydian wrote:Like the order of the emotes there...!

Could do with a beer too - oh to heck with it, an after work jar beckons!

Great minds think alike I am OFF TO THE PUB in 7 minutes!

by **laverfan** Fri 27 Apr 2012, 4:51 pm

We have

1. Broken Record -

2. Yikes - Beware (Stolen from another forum) - Draw Fixing : A Real World Analysis - Part 1 (The 12 vs 12 Question) - Page 3 Yikes

by **JuliusHMarx** Fri 27 Apr 2012, 4:53 pm

Broken Record - great album - http://www.lloydcole.com/music/brokenrecord/main.html

by lydian Fri 27 Apr 2012, 4:59 pm

I'm jealous reckoner - enjoy!

Cool LF...I must be blind! Headscratch

Not heard of them for a while JHM...Perfect Skin and all that music

by **JuliusHMarx** Fri 27 Apr 2012, 5:00 pm

Solo artist now (since about 1990!). Saw him live last month at a small venue in Birmingham.

by lydian Fri 27 Apr 2012, 5:02 pm

Good to see he's still around...not many from the 80s still actively touring these days.

by summerblues Sat 28 Apr 2012, 12:48 am

laverfan wrote:I removed the 4 opposite-half scenarios to make the 12vs12 and the 9 meetings comparable. If you assume 14/16 (87.5%). Then it is 9/13 (69.2%) in the Connors v McEnroe case.

Statistically 9 (CvM) or 12 (DvF) are not any different in the same half, are they?

Statistically, neither 9/9 nor 12/12 are interesting, because in both cases we are cherry-picking exactly those draws that we like. I would say 12/12 is a touch better because at least they were able to just throw out one slam rather than have to throw out across slams but that is a relatively minor detail. There is, however, a world of difference between 9/13 and 14/16. The fair probabilities would be as follows:

(a) probability of being drawn in the same half at least 9 times out of 13: about 1 in 8.
(b) probability of being drawn in the same half at least 14 times out of 16: about 1 in 478.

So your example is nowhere near similar to Federer-Djokovic. If anything, you confirmed that indeed it is very rare to find a pattern like theirs - you searched through around 30 years of data and came nowhere near that. Note that there are two main reasons why the second one is much less likely. First, the percentage is higher (87.5% vs 69.2%). Second, the sample size is a bit bigger. The sample size also matters. For example, if we had two players drawn in the same half 4 times out of 4 cases, i.e., 100% times, the probability would not be extremely small - only 1 in 16.

laverfan wrote:
summerblues wrote:...If they had been drawn 14 times out of 28 possible or indeed 30 times out of 60 possible, it would be entirely uninteresting.
You are willing to accept a 70-30 in the CvM case, but now want a 50-50 split in the DvF case. Is that fair?

Statements like these make me stop and wonder whether you are really interested in exploring the data or whether you just want to dismiss anything that goes against your view. As I said on the other thread, I find this puzzling because I generally view you as a pretty open-minded poster.

If you read my statement - either stand alone as quoted here, or perhaps even more obviously in the context it was written, it should be clear that I am not talking about asking for a 50-50 split.

Finally, I would like to say that I do not claim that the draws were rigged. In fact, I would not necessarily say that I lean in that direction. All I am hoping for is to agree that the probability of the 14/16 pattern observed in Federer-Djokovic is quite low - about 1 in 478 in fact. There really is not much room for debate there.

Once we agree on that, we can then ask ourselves whether we think that this is just one of those rare events that are bound to occasionally happen in perfectly fair draws or whether something other is happening here. The arguments there will be of the sort mentioned here already - such as the incentives to rig the draws versus risks involved, our perceived level of corruption in tennis circles (if any), etc.

by **laverfan** Sat 28 Apr 2012, 3:51 pm

summerblues wrote:So your example is nowhere near similar to Federer-Djokovic. If anything, you confirmed that indeed it is very rare to find a pattern like theirs - you searched through around 30 years of data and came nowhere near that.

Did you look at the Graf v Sabatini (75%) and Evert v Mandalikova (76.9%) example?

summerblues wrote:Note that there are two main reasons why the second one is much less likely. First, the percentage is higher (87.5% vs 69.2%).

See previous statement.

summerblues wrote:Finally, I would like to say that I do not claim that the draws were rigged.... All I am hoping for is to agree that the probability of the 14/16 pattern observed in Federer-Djokovic is quite low - about 1 in 478 in fact. There really is not much room for debate there.

The challenge here is that just because the probability is low, does not mean such an event will not occur. Correct? There are many events of low or even lower probability, which do occur, for example the Krakatoa Volcano explosion. Positing that the occurrence of such an event is tantamount to 'rigging' is what I have strong objections too.

summerblues wrote:Once we agree on that, we can then ask ourselves whether we think that this is just one of those rare events that are bound to occasionally happen in perfectly fair draws or whether something other is happening here. The arguments there will be of the sort mentioned here already - such as the incentives to rig the draws versus risks involved, our perceived level of corruption in tennis circles (if any), etc.

The area that you allude to is a highly divisive and subjective area of interpretation. My role is to provide data and analysis.

Thanks for responding.

by Guest Sat 28 Apr 2012, 4:08 pm

laverfan wrote: The challenge here is that just because the probability is low, does not mean such an event will not occur. Correct? There are many events of low or even lower probability, which do occur, for example the Krakatoa Volcano explosion. Positing that the occurrence of such an event is tantamount to 'rigging' is what I have strong objections too.

The old Krakatoa paradox. If I had a pound for every time that old Krakatoan chestnut was raised I would be a very rich person - well at least in a country in which the cost of living was very low.

ps: those Krakatoan balls would be red hot rather than warm.

by summerblues Sat 28 Apr 2012, 4:52 pm

laverfan wrote:Did you look at the Graf v Sabatini (75%) and Evert v Mandalikova (76.9%) example?

I only looked at McEnroe-Connors because that is the only one where I independently verified your stats (it does take fair amount of effort - which I quite admire and can appreciate, since I am also statistically inclined). However, I have no reason to think that the data in your other cases would not hold up. Assuming it does, then we get:

(a) 1-in-14 probability for Graf-Sabatini (9/12)
(b) 1-in-22 probability for Evert-Mandlikova (10/13)

So better than McEnroe-Connors but still very far off Federer-Djokovic - not in the same league really.

laverfan wrote:
The challenge here is that just because the probability is low, does not mean such an event will not occur. Correct?

Agreed, ultimately that is the question, but I would first like to make sure that we all agree that the probability is really relatively low and that there is no misunderstanding there. 1-in-478 is not insanely low but it is quite low and that is what the probability is - no getting around that.

Also, I think that is more or less as far as stats and math can get us. Beyond that the debate has to enter the subjective, and thus more contentious, territory.

I do not think that looking at more data sets adds that much to the debate. To the extent it does, I think it works quite differently from what you are suggesting. Suppose you looked through the data and found that 1-in-478 probability cases were happening left and right all over the place. The conclusion from that would not be "look, this happens quite often, so it is not such a strange thing". The conclusion would be "1-in-478 cases are happening way more frequently then they should, this does indeed suggest something strange is happening in tennis". So, if anything, you should be happy that you cannot find anything similar to Federer-Djokovic. At least that indicates that the pattern we observed with them is very rare - as it indeed should be if the draws were all fair. However, as we do not really have enough data points (we would need more like 3,000 years instead of 30 years worth of data), hard to draw any strong conclusions from that.

BTW, I hope you do not mind arguing about this. I do like arguments but do not want to make anyone feel uncomfortable, so if you are tired of it we may call it a day.

by lydian Sat 28 Apr 2012, 8:50 pm

Good stats work guys, I dont profess to be knowledgeable on stats/probabilities (its not my area of training) but I see that the original 1 in 4096 chance of 12/12 has been stated often and now reduced down to 1 in 478 given inclusion of FO. On the face of it it appears to still be a small probability of the run seen happening...and I can see the basis of the question being asked about draws.

But my question is this...is the probability further, and fundamentally, affected by the way tennis draws work?, i.e. we know that 1 v 2 and 3 v 4 go automatically into opposite halves meaning that 1 and 2 can only meet in the final and 3 and 4 have 1/2 chance of meeting in semi or final. However, because Federer and Djokovic have rarely (if ever?) been seeded #1 & 2, or #3 & 4, i.e. they wouldnt automatically be placed into opposite halves, does this mean they had greater chance of being drawn against each other in the semis (than final) - or is it still 50/50? My feeling was that because they are always in the top 4 but usually seeds 2 or 3...this has given them a greater chance than normal of meeting in the semis than would ordinarily be the case if a completely level playing field applied. Be interested to hear thoughts on that...and again I'm no stats expert so apologies if this is a naive question! Shocked

by Guest Sat 28 Apr 2012, 9:55 pm

If I flip a coin five times and it comes up heads each time - then the next time I flip a coin it is still a one in two chance it will come up tails. They are called "independent events" - and whatever has occurred in the past doesn't influence what occurs in the future.

In reality, even a coin flip can be non-random (I can flip a coin so that it spins exactly the same number of times to "load" the result), and the coin itself might be double-headed Smile

by hawkeye Sat 28 Apr 2012, 10:06 pm

So everyones still discussing this.... Zzzzzz....

Anyway I have a solution before every slam (other tournaments dont count of course) or now that I think of it before every slam that isn't the FO. Nore Staat (being the expert at coin tossing) should toss a coin. If it comes up tails Djokovic gets Federer. If it comes up heads Nadal gets Federer. If Nore Staat gets it wrong then they simple toss the coin again until they get it right...

by Guest Sun 29 Apr 2012, 12:31 am

hawkeye wrote:... Anyway I have a solution before every slam (other tournaments dont count of course) or now that I think of it before every slam that isn't the FO. Nore Staat (being the expert at coin tossing) should toss a coin. If it comes up tails Djokovic gets Federer. If it comes up heads Nadal gets Federer. If Nore Staat gets it wrong then they simple toss the coin again until they get it right...

With my sleight of hand and "lucky" coin(s) I would be happy to oblige.

by **laverfan** Sun 29 Apr 2012, 12:41 am

summerblues wrote:BTW, I hope you do not mind arguing about this. I do like arguments but do not want to make anyone feel uncomfortable, so if you are tired of it we may call it a day.

I am not uncomfortable at all. The more mathematical stuff, the better. Hug

lydian wrote:Be interested to hear thoughts on that...and again I'm no stats expert so apologies if this is a naive question! :shocked:

I will respectfully defer this to SummerBlues. He is much better than I. Wink

I consider myself a bit biased (like Nore Staat's double-headed coin).

by summerblues Sun 29 Apr 2012, 6:59 am

laverfan wrote:I am not uncomfortable at all. The more mathematical stuff, the better.

Glad to hear, I will then try to go on a bit further and look at Lydian's question:

No, it does not matter, the probabilities are still the same, as long as one of them is seed #1 or #2 and the other one #3 or #4 (which I understand was the case in all those slams). Seed #1 is placed in the top half and seed #2 in the bottom half. There are only two ways seeds #3 and #4 can be drawn - either seed #3 is in the top half and seed #4 in the bottom half or vice versa. In other words we have two possible semifinal draws - each with the 50% chance of happening:

(a) #1 vs #3 and #2 vs #4, or
(b) #1 vs #4 and #2 vs #3

Say Federer is seeded #2 and Djokovic #4. Then they will be in the same half in case (a) but not in case (b) - exactly 50% chance. The same goes for any other seeding option (again, as long as one of them is either #1 or #2, and the other either #3 or #4).

Now, it does not matter if from one slam to the next their seedings change. The same argument as above will always apply to each individual slam - so the probability will be 50% each time. So, say the probability of getting 12 consecutive semifinal meetings will be 1-in-4096 even though to the "naked eye" this may look different from having 12 consecutive draws where #1 and #3 seeds are drawn together, which has the same probability.

One thing worth noting (we touched upon it with HB on the old thread) is that certain implicit "cherry-picking" may be happening even in perfectly good looking studies. Let's consider again the 12/12 example (a bit simpler to work with than the 14/16 one):

(a) Say, Federer-Djoko are in the same half 12/12 times. Well, this has a chance of 1-in-4096 as mentioned many times before.
(b) But what if Federer-Djoko ended up in the opposite half 12/12 times? Once again, we would have said that this had 1-in-4096 probability
(c) What if instead #1 and #3 seeds were drawn in the same half 12/12? Again, 1-in-4096.
(d) Finally what if #1 and #4 seeds in the same half 12/12? Of course, 1-in-4096 again.

If any of the above happened, we would likely note (correctly) that the event had a 1-in-4096 chance. However, there are 4 such events, so the chance of one of them happening is 4-in-4096, i.e., 1-in-1024. So a researcher could check for any of these patterns and if they found one they could have reported a 1-in-4096 event, without telling us that they really were checking more than just that one pattern, and thus gave themselves a better than 1-in-4096 chance to succeed.

by socal1976 Sun 29 Apr 2012, 9:45 am

The draws were rigged to protect the fedal finals and probably rigged in otherwise, ie the coincidental Isner v. Mahut rematch. At this point now that two different studies have looked at the draws and have determined non randomness I feel that the onus the burden of proof shifts to those who are trying to argue that yes the 1-4000 or 1-400 longshot did come in without a the result being constructed.

by **JuliusHMarx** Sun 29 Apr 2012, 12:34 pm

socal1976 wrote:The draws were rigged to protect the fedal finals and probably rigged in otherwise, ie the coincidental Isner v. Mahut rematch. At this point now that two different studies have looked at the draws and have determined non randomness I feel that the onus the burden of proof shifts to those who are trying to argue that yes the 1-4000 or 1-400 longshot did come in without a the result being constructed.

2008 AO Federer #1 - Djokovic #3 (Djokovic in Federer's half) 1-3, 2-4

So the AO 2008 organisers looked at the Nadal/Djoko H2H, saw that it was 6-2 in Rafa's favour, figured Djoko would beat him (even though Rafa was yet to reach a AO semi at that point), and rigged it in some unspecified manner so that he was drawn to play Davydenko in the semi instead (if it played out according to the seeding).

Even though no studies have ever produced evidence of draw-rigging at the AO. And no-one can offer any explanation of how the draw would be rigged, when the balls are drawn out of a hat by a player, with an audience (apart from 'warm balls', which would require the player to be in on it).

The USO 2008 then decided that the best way for Fed to get to the final was to place him in the same half as Djoko, who had beaten him in the AO semi-final, (thus keeping Rafa away from Djoko, even though Rafa had just beaten Djoko in the Olympics).

Yeah, it all makes sense.

by Guest Sun 29 Apr 2012, 1:10 pm

JuliusHMarx wrote:... Yeah, it all makes sense. ...

It makes more sense than anti-aircraft missiles on top of London residential / council flats:
http://news.uk.msn.com/uk/missiles-may-protect-olympics
http://www.bbc.co.uk/news/uk-17884897

Personally I am offering the British Military access to my balcony for stationing one of their Gatling type guns, as you can never be too sure nowadays.

by **JuliusHMarx** Sun 29 Apr 2012, 1:17 pm

I'm offering my goldfish tank for a Trident nuclear sub.

by lydian Sun 29 Apr 2012, 7:52 pm

That's some tank you have JHM Wink

Thanks for the stats reasoning summerblues, appreciated

The stats show the run is unusual but the leap to actual rigging is a subjective one. And the real-world case, i.e. how it could be done without being found out across all majors besides French - and funny how they all started doing this in 2008 so there had to be collusion across the majors too, has to be proved.

Speaking of which, I agree with the "draw rigging anti-rationale". I have never bought into this "Nadal must avoid Djokovic at all costs" plot. Its easy to now retrospectively say that after last year but there would be no reason to say that before mid-2011.
So with clay taken out of the equation (as it doesnt fit the 12/12 dataset) we are left with grass and HC slams to examine for rigging. In those, Djokovic was as big a threat to either player given as JHM says Nole had just beaten Federer on HC at AO (and earlier too). So why pit him just against Federer? On grass Nadal had nothing to fear from Djokovic pre June 2011...nor HC either actually. The rig-plot through real world eyes just doesnt add up for me - it seems the conspiracy theorists have jumped on this 'data' and assumed the draw to be rigged, then sought to prove their case retrospectively. Yes the 12/12 run for names is unusual but the real world reasons for draw rigging dont support keeping Nadal away from Djokovic. Plus as JHM says...how could they a) practically do it, and b) assume to get away with it?

Oh...and additionally, Murray beat Nadal in 2 HC slams at USO & AO...so using the rig-reasoning if anything why not keep Nadal away from Murray as much as keeping him away from Djokovic?

There are too many issues with the dataset and real-world holes for this plot to make any semblance of realistic sense. In my opinion!

by socal1976 Sun 29 Apr 2012, 8:47 pm

Yes ignoring the fact that when this draw rigging was taking place murray also had a superior h2h v. fed so in a way fed was getting a little extra protection with the seeding. That is why the motivation existed to not give Novak to Nadal and to not give Fed, Murray. In 08, 09, and 10 if you asked fed if would want to play novak or Murray he would say Novak. If you asked Nadal in 08, 09, 10 who he would prefer to play he would probably say murray, coincidentally that happened over and over again.

The fact remains with all this talk about balls being picked out of tubes and so forth 3 sets of statisticians have analyzed the draws and have found that with a very high likelihood these draws are NOt Random. That means on some level they are manufactured. A draw that in anyway is non-random and played with is rigged. It is like virginity, you either are virgin or it isn't. The draw either is random or it isn't, and therefore on some level manufactured.

Here is another trend I am observing that I am certain the draw rigging committees will continue to benefit from a fortunate coincidence. Whenever a tournament has Novak, and only 2 of the remaining big 4. The world number 1 will "coincidentally" get the lower ranked big 4 member in his half and will have to beat two big 4 guys where federer or Nadal will get a clear half of their field. Dubai only murray, fed, and novak show up; guess what murray is in Novak's half. Monte Carlo only Nadal, Novak, and Murray show up; guess what Murray is in Novak's half. So far this year 2 for 2 I am sure that will be another one of those happy coincidences for the tournament directors. More discrimination against Novak and more protection for fedal.

by hawkeye Sun 29 Apr 2012, 9:10 pm

This is all Federers fault! Sneakily dropping down to a ranking of three. Seriously that goes against all known rules of nature. IMO it is a tactic. He can now cause pandemonium, panic and sleepless nights amongst the players. Where will he land and whose rightful place in the final will he steal?

Hopefully it is only a matter of time before Federer regains a fixed place in the draw...

by **JuliusHMarx** Sun 29 Apr 2012, 9:13 pm

3 set of statisticians? I thought there were 2 studies - one by ESPN statisticians and one by a law academic and even the law academic admitted it didn't prove anything.
As for asking Fed/Nadal who they would prefer to play, that's just pure speculation, which you've interpreted in your own way, to fit your existing views - there's no evidence that e.g. Fed would have preferred to play Djoko instead of Murray.
There is no evidence at all that e.g. the AO or Wimbledon have ever rigged any draws - none at all!
And you wonder why the term 'conspiracy theorist' get used - maybe because it's entirely a theory you have about a conspiracy.

by Guest Sun 29 Apr 2012, 10:34 pm

This thread is turning into a rig-marole (sorry couldn't resist Smile

)

by **JuliusHMarx** Sun 29 Apr 2012, 10:54 pm

Nore Staat wrote:This thread is turning into a rig-marole (sorry couldn't resist )

I know, but I can't resist my daily fix of it.

by Guest Sun 29 Apr 2012, 11:44 pm

My view is that there has probably been some "loading" of the "dice" in a few tournaments, but probably nothing "major".

Didn't some cricketer lose the toss 14 times in a row? I can't seem to find anything on it over the interweb.

by socal1976 Mon 30 Apr 2012, 2:40 am

Yes julius 3 sets of statisticians. The Espn study was so shocking in its findings that ESPN sent their stats guys analysis out to a thrid party who looked at the numbers and the first guys stat methodology and stated that the first statisticians findings were accurate. That is where I get 3 sets of statisticians from julius.

Watch out for the new scam. Whenever it is 3 of the 4 top guys and not all 4 they will give Novak the extra semi against murray. Did it to him in Dubai, would have done it to him at Monte Carlo as well but Berdy won in the quarter. How much extra protection does fedal need from the serbian terminator?

by **JuliusHMarx** Mon 30 Apr 2012, 8:29 am

socal1976 wrote:Yes julius 3 sets of statisticians. The Espn study was so shocking in its findings that ESPN sent their stats guys analysis out to a thrid party who looked at the numbers and the first guys stat methodology and stated that the first statisticians findings were accurate. That is where I get 3 sets of statisticians from julius.

Watch out for the new scam. Whenever it is 3 of the 4 top guys and not all 4 they will give Novak the extra semi against murray. Did it to him in Dubai, would have done it to him at Monte Carlo as well but Berdy won in the quarter. How much extra protection does fedal need from the serbian terminator?

OK - 1 study (ESPN), with 2 sets of statisticians and 1 study (the recent one) with 1 law academic and no statisticians. That would be 2 sets of statisticians in total.

As for the anti-Djoko conspiracy theory - that's exactly what it is. Who did Djoko get in the World Tour Final groups? Not Fedal. Who did he get in the AO semis? Not Fed - they let Rafa take care of danger-man Fed, before handing Djoko his bunny in the final. Paris 2011 (without Rafa) - Murray and Fed in same half.

Now you're basing a new conspiracy theory on the huge sample of 2 whole tournaments since then. Those 1 in 4 odds are staggering! Hey, it's not paranoia if they're really out to get you! Believe what you will about that, but at least accept that it's a theory about a conspiracy, not anything based on evidence.

by hawkeye Mon 30 Apr 2012, 9:41 am

socal1976 wrote:Whenever it is 3 of the 4 top guys and not all 4 they will give Novak the extra semi against murray.

I don't know. If it's Nadal that can do the draw fixing he might choose to take Murray in exchange for Ferrer...

by socal1976 Tue 01 May 2012, 4:33 am

All I said Julius was that I am keeping an eye on this new trend in the draw matchups. Giving Novak the extra semi against a big 4 competitor. The facts remain that 3 different statisticians found a high likelihood that the draws are not random. That means that the draws are on some level manufactured. The evidence here goes way beyond guys on the internet talking about coin tosses.

by **JuliusHMarx** Tue 01 May 2012, 8:46 am

socal1976 wrote:The facts remain that 3 different statisticians found a high likelihood that the draws are not random.

No.
One set of statisticians found...etc. Another set peer-reviewed and confirmed.
As for the second study - Katarina Pijetlovic is not a statistician, any more than you or I are. She's a law lecturer, who claims no expertise in statistics.

So those 'facts' that remain, aren't facts at all.

If I were to ask e.g. how many qualified statisticans have found any evidence of draw-rigging at the AO, the answer would be zero.

If there are 7000 people on this site, the odds of any one given person in the world being on this forum are 1 in 1000000. The odds of any 2 given people being on this forum are 1 in 1000000000000. That's the odds that socal and I should both be on this site. Yet it's happened. It's rigged, I tell you!

by lags72 Tue 01 May 2012, 9:30 am

Ah Julius, I'm glad to see you neatly debunking the notion that theories somehow magically become 'facts' ...... just because people say they are.

Not that I'm about to enter the eye of the storm here. It all looks way too complex for me. But it does give a welcome opportunity to trot out my favourite quote about statistics, which is ............ that statistics are often misused in the way a drunk uses lamp-posts : for support rather than illumination

by lydian Tue 01 May 2012, 9:55 am

good comments JHM and lags72...love that quote lags!

At the end of the day I dont see the tennis world and media in uproar about it...surely if the data had any credence there would be some serious firefighting going on with the major TDs by now.

by **laverfan** Tue 01 May 2012, 2:08 pm

lags72 wrote:Ah Julius, I'm glad to see you neatly debunking the notion that theories somehow magically become 'facts' ...... just because people say they are.

Not that I'm about to enter the eye of the storm here. It all looks way too complex for me. But it does give a welcome opportunity to trot out my favourite quote about statistics, which is ............ that statistics are often misused in the way a drunk uses lamp-posts : for support rather than illumination

Statistics is akin to a Bikini, what is reveals is fascinating, what it hides is vital. Wink

PS: Just waiting for JHM to make a smart ass comment on this one. Laugh

by **JuliusHMarx** Tue 01 May 2012, 3:14 pm

I prefer it when they wear no bikini at-oll

by lags72 Tue 01 May 2012, 3:18 pm

Looks like you got your wish laverfan ....

JHM thumbsup

by Guest Tue 01 May 2012, 3:54 pm

Statistics reveal most people no longer fit in their bikinis.

by reckoner Wed 02 May 2012, 9:59 am

3.265% of statisticians wear bikinis under their work clothes.

by **laverfan** Wed 02 May 2012, 12:27 pm

reckoner wrote:3.265% of statisticians wear bikinis under their work clothes.

by Josiah Maiestas Wed 02 May 2012, 5:37 pm

In latest stats it has been revealed that homosexual people are less likely to have a baby than heterosexuals. :\

by **laverfan** Wed 02 May 2012, 5:44 pm

Josiah Maiestas wrote:In latest stats it has been revealed that homosexual people are less likely to have a baby than heterosexuals. :\

You forgot about surrogates. Laugh

by Josiah Maiestas Wed 02 May 2012, 5:47 pm

laverfan wrote:
Josiah Maiestas wrote:In latest stats it has been revealed that homosexual people are less likely to have a baby than heterosexuals. :\

You forgot about surrogates.

Are surrogates homosexual or heterosexual? I don't know what you mean. Very Happy

by Sponsored content

Has this article addressed any concerns of the legitimacy of the draws?