Re: The Big Fix?

[Sakari Lund]

I'll try posting the original post again, becauce it is really curious
everything else shows in Google Groups except the early part of this
thread...



I haven't believed in all the fixing theories in slam draws, but we
were approached at work a few days ago by someone who has studied the
subject, and she has quite interesting results, which have made me
reconsider it a little.

We have talked about a lot here how Federer and Djokovic always seem
to be in the same half. But I hadn't realized that, excluding FO, Fed
and Djok have been in the same half in every single slam in 2008-2011.
So it is 12/12. Same with Nadal and Murray. She has rather interesting
explanations of what might be behind this.

I wrote a story about it, which has been very popular and has been
commented a lot.

I have been in contact with the author during the last few days. She
wants publicity for it, so I post about it in here too. She has this
facebook page about it, click on the document at the bottom.

I don't know if I still believe there is a big conspiracy, but sure
seems to be strange.

http://www.facebook.com/pages/Tennis-biggest-public-secret/284730451547505?sk=info

[kaennorsing]

> http://www.facebook.com/pages/Tennis-biggest-public-secret/2847304515...

Don't know if the numbers she comes up with are correct, but I don't
think it's correct to combine these stats with the ESPN stats
regarding top players drawing unseeded players. That's a seperated
issue; The likelyhood (statistically) of anything happening that
actually happens is remote.

Overall though, of course there's been something fishy about the GS
draws... Quite clearly IMO.

[bob]

On Fri, 28 Oct 2011 13:26:32 +0300, Sakari Lund
<sakar...@welho.com> wrote:

>I'll try posting the original post again,

nah, wrong approach - better idea is see if you can hire TT to tutor
you on probability theory in his free time.

bob

[Sakari Lund]

They are not my numbers. I try to say in every post that I still have
doubts about there being a fix. But there are interesting points, if
anyone bothers to read it.

[Ted S.]

On Fri, 28 Oct 2011 13:26:32 +0300, Sakari Lund wrote:

> I'll try posting the original post again, becauce it is really curious
> everything else shows in Google Groups except the early part of this
> thread...

The original showed up just fine using a real news server.

--
Ted Schuerzinger
tedstennis at myrealbox dot com
If you're afraid of the ball, don't sit in the front row. --Anastasia
Rodionova

[Sakari Lund]

On Fri, 28 Oct 2011 08:38:18 -0400, "Ted S." <justac...@gmail.com>
wrote:

>On Fri, 28 Oct 2011 13:26:32 +0300, Sakari Lund wrote:
>
>> I'll try posting the original post again, becauce it is really curious
>> everything else shows in Google Groups except the early part of this
>> thread...
>
>The original showed up just fine using a real news server.

I know, but many people use GG, so they missed it. And I don't know,
but it seems strange that the early part of my thread didn't show up
on GG, but when I posted in other threads, and as far as I know what
everyone else posted in every other thread as well, everything showed
up just fine. A real big conspiracy...

Did you read Pijetlovic study? What do you think about it? For me,
what it did was that before I thought all the fix talk was ridiculous,
but now I am not so sure.

[Professor X]

> http://www.facebook.com/pages/Tennis-biggest-public-secret/2847304515...

Doubt it's a big conspiracy, but it is curious.

THe odds of each of top 4 being on either side of draw are 1/2 and 1/2
to the power 12 is 4096/1 as she says so that much is correct. Don't
understand how she gets from 4096/ 32billion to one though at all.
That simply isn't / can't be correct.


However yes 4096/1 is an unlikely event.... but not unlikely ENOUGH to
speculate about fixing i.m.ho

If in another year this is still the same in which case 1/2^16 =
1/65563

Then maybe we can phone the atp. :-)

[TT]

Yes, it's a separate issue. It's a matter of having odds of, say, 1/2048
AND 1/1000...rather than having an odd of 1/2048,000

One could as well claim that odds for John Doe having been on top half
of a slam draw was 50%...and thus multiplying the odds above by 2...

[Sakari Lund]

1/4096 is Federer and Djokovic being in the same half 12/12.

1/131072 comes from the following. I don't know if it is correct, but
that is the claim:

"Under the laws of probability having the draw of 3rd and 4th seeded
players have identical outcome (i.e., Djokovic always in Federer's
half, and Nadal always in Murray's part whenever possible and if not
possible due to the seeding, Nadal-Federer finals were enabled by
drawing them into separate halves) in 12 out of 12 times is 1 in 4096,
or 0.02%. But Murray was not among 4 first seeded players at 5 of
those 12 tournaments (Australian Open 2008, 2010, 2011, Wimbledon
2008, and US Open 2008). This means he was drawn separately from
Djokovic, in the group with other seeded players. In each and every of
those 5 separate draws he was placed in Nadal's half. The same result
in 5 out of 5 draws can happen is 1 in 32. 4096 multiplied with 32 is
131072. THE PROBABILITY TO OBTAIN THE RESULTS AS WE SAW IN HARD AND
GRASS COURT GRAND SLAM TOURNAMENTS 2008-2011 IS 131072 TO 1."


1/32 billion comes from combining the above to ESPN research that top
two seeds get as easy opponents as possible at USO. That probably
doesn't make much sense.

[Joe Ramirez]

On Oct 28, 8:55 am, Sakari Lund <sakari.l...@welho.com> wrote:
>
> Did you read Pijetlovic study? What do you think about it?

It's not much of a study:
1. Asserting the improbability of the same two guys meeting each other
over and over: done often in many channels, including RST.
2. Noting that people like to see Federer-Nadal finals and that both
players are sponsored by Nike.
3. Listing the draws for us.

Any actual evidence of draw fixing? No. Any statements about possible
funny business by people involved in making the draws? No. Any
investigation of the actual mechanisms for conducting draws and
analysis of potential weak spots? No. Any money trail? No. Any
reaction from the ITF, ATP, Nike, the slams, or the players? No.
Merely running the numbers never has any long-term impact. It's like
using a calculation to support speculation about the probability of
life on other planets. After the initial, "Wow -- impressive!" people
just forget and go about their business.

[Sakari Lund]

On Fri, 28 Oct 2011 07:56:15 -0700 (PDT), Joe Ramirez
<josephm...@netzero.com> wrote:

>On Oct 28, 8:55 am, Sakari Lund <sakari.l...@welho.com> wrote:
>>
>> Did you read Pijetlovic study? What do you think about it?
>
>It's not much of a study:
>1. Asserting the improbability of the same two guys meeting each other
>over and over: done often in many channels, including RST.
>2. Noting that people like to see Federer-Nadal finals and that both
>players are sponsored by Nike.
>3. Listing the draws for us.
>
>Any actual evidence of draw fixing? No. Any statements about possible
>funny business by people involved in making the draws? No. Any
>investigation of the actual mechanisms for conducting draws and
>analysis of potential weak spots? No. Any money trail? No. Any
>reaction from the ITF, ATP, Nike, the slams, or the players? No.

Right. But there is a clip in the net of her presenting the case in
some conference, and there she says she is an academic and doesn't
have the resources to investigate this, and hopes the media does that.

[Joe Ramirez]

It's paradoxical but true that highly improbable things happen often
in the world. *Why* they happen is the all-important question. Few
organizations have the inclination or resources to go around
investigating things that are improbable simply because they are
improbable. There typically needs to be at least a nugget of fact
suggesting inappropriate conduct somewhere, or significant harm to
someone.

[Pelle Svanslös]

These look ok.

> 1/32 billion comes from combining the above to ESPN research that top
> two seeds get as easy opponents as possible at USO. That probably
> doesn't make much sense.

It's formally correct, but it's not the way to go about. It's easy to
calculate the expected ranking round by round and then compare the real
life draws to that.

Saying something is improbable doesn't say anything since there's tons
of events, whether they go with the match fixing or not, that are
improbable.

It's still a remarkable streak. We already have one recorded draw fixing
AND we have the guy responsible for that fixed draw on record lying 15
years later:

"What would the U.S. Open gain by fixing the draw in this way? I believe
the U.S. Open would gain nothing," Earley said. "I think that that would
be a risk that the U.S. Open would never take. Never."

--
http://memedepot.com/uploads/2000/2098_steamboat.gif

[bob]

On Fri, 28 Oct 2011 14:58:38 +0300, Sakari Lund

<sakar...@welho.com> wrote:

>On Fri, 28 Oct 2011 07:21:07 -0400, bob <stei...@comcast.net> wrote:
>
>>On Fri, 28 Oct 2011 13:26:32 +0300, Sakari Lund
>><sakar...@welho.com> wrote:
>>
>>>I'll try posting the original post again,
>>
>>nah, wrong approach - better idea is see if you can hire TT to tutor
>>you on probability theory in his free time.
>
>They are not my numbers. I try to say in every post that I still have
>doubts about there being a fix. But there are interesting points, if
>anyone bothers to read it.

i don't doubt a fix even if the #s were 1/100.
(but they're not 1/13000th or whatever's claimed).

bob

[bob]

scott has been hoping the media investigates the moon landing, jfk
killing and WTC#7 also.

bob

[number_six]

Ever notice that except for the matches he lost, Federer went
undefeated?

Stated in that form, the sham is transparent immediately. But in this
article, aren't we also looking at cherry-picked data?

Is the author really evaluating a series of like outcomes in 12
consecutive 50-50 throws, to which we can apply the normal method of
powers of two in the denominator?

Or is the string interrupted by an event (RG 2010) that *cannot* be
disregarded without having to change the method? Is the surface
distinction legit, or is it methodologically invalid? I think the
latter, but am I overlooking something? That would leave only six
consecutive like outcomes -- far, far less improbable than a run of
twelve.

By another measure, there is even a variance since RG 2010:

RG 2010 = 1/4/3/2
W 2010 = 1/3/4/2
US 2010 = 1/4/3/2
AO 2011 = 1/4/3/2
RG 2011 = 1/4/3/2
W 2011 = 1/4/3/2
US 2011 = 1/4/3/2

Look at it this way, and we are down to only *five* consecutive like
outcomes. Not that stunning. I was much more troubled by the stats on
the early round opponents drawn by the top 2 seeds at the USO.

[TT]

The trouble with life on other planets hypothesis is difficulty of
proving it...the distances are huuuge...

Guess the only argument against it... is our "uniqueness", and belief in
extraordinary...

[TT]

28.10.2011 18:44, Joe Ramirez kirjoitti:

> It's paradoxical but true that highly improbable things happen often
> in the world.

Like what for example?

[TT]

Yes, only thing written in stone is Denko's fixing. :)

> It's still a remarkable streak. We already have one recorded draw fixing
> AND we have the guy responsible for that fixed draw on record lying 15
> years later:

What cheapens the streak though is it being speculative..."not including
RG"....

>
> "What would the U.S. Open gain by fixing the draw in this way? I believe
> the U.S. Open would gain nothing," Earley said. "I think that that would
> be a risk that the U.S. Open would never take. Never."
>

Weren't they caught on fixing some years back?

[kaennorsing]

Take any particular draw and calculate the probability of all those
players ending up exactly where they did.

Or take your phone number and calculate the probability of you ending
up with that exact number.

Now combine the two and calculate the probability of both actually
happening. Then take the phone numbers of all your contacts in you
phone and calculate how probable it was they ended up with those exact
numbers they actually did end up with etc. etc.

[Sakari Lund]

On Fri, 28 Oct 2011 21:05:17 +0200, Pelle Svansl�s <pe...@svans.los>
wrote:

Cool, someone thinks so...

[Joe Ramirez]

Most complex contingent events, and even many simple ones, would be
considered extremely improbable if they could be viewed prospectively.
Generally, however, they are viewed only retrospectively, when
probability is no longer germane.

Like most people, you likely have experienced the "it's a small world"
phenomenon several times in your life. I.e., two people from the same
neighborhood or school, perhaps even old friends, happen to meet each
in a huge city far from home. "What are the odds ...?!" they might
say. But such events, considered cumulatively, are not especially
rare.

Back in the 1950s, RAND Corporation published a book entitled, "A
Million Random Digits," for use in mathematics and statistics
projects. Although it was produced with the best random number
technology available at the time, and *as a whole* doesn't appear to
have a discernible pattern, there are plenty of small-scale patterns
within the overall sequence. E.g., you might see something like
"0101010101" that in isolation would appear quite improbable.

How difficult do you think it would be for one person to win 12
straight coin tosses of a fair coin? Would that be a "normal" sort of
event, or a freakish, mysterious occurrence that would have to
indicate either ESP or a "fix"? Well, all you need to do to identify a
person who can win 12 straight coin tosses is to go to a medium-size
rock concert. Hold a 12-round coin flip tournament with the audience
members (you need only 4,096), and the champion will be a person who's
just won 12 coin tosses in a row. (You'll also have multiple people
who won 11 straight tosses, and 10 straight, etc.)

[Pelle Svanslös]

That's 4K experiments in total though. Doesn't matter if it's 4K people
doing one 12-tossing sequence each or if it's one person tossing 4K
12-toss sequences.

In slam draw terms that would mean 12000 years' worth of slam draws to
search for the 12 streak.

LOL.

--
http://memedepot.com/uploads/2000/2098_steamboat.gif

[TennisGuy]

Very poor examples.

[Joe Ramirez]

Yes, so?

>
> In slam draw terms that would mean 12000 years' worth of slam draws to
> search for the 12 streak.

No. You're focusing on the number of trials required to make an event
*certain* (or highly probable). The point here is that the mere fact
that an event appears very improbable from one perspective does not
mean that a "fix" is to be inferred. Much of our experience consists
of a series of improbabilities.

[TennisGuy]

On Oct 28, 11:44 am, Joe Ramirez <josephmrami...@netzero.com> wrote:

> It's paradoxical but true that highly improbable things happen often
> in the world.

How often is often? A little more than seldom? 40% of the time? 60%
of the time?

[Pelle Svanslös]

I just wanted to make the more or less parallel situation into a
sequential which is the case with the slam draws.

>>
>> In slam draw terms that would mean 12000 years' worth of slam draws to
>> search for the 12 streak.
>
> No. You're focusing on the number of trials required to make an event
> *certain* (or highly probable).

Not really focusing on anything, it's your example I'm going by.

The point is that "It's paradoxical but true that highly improbable
things happen often in the world." is only true given the numbers. In
your example the number would be 4K experiments for one streak.

If we had those numbers in the slam draw context and happened to bump
into a streak, nobody would think anything of it. But we don't and
eyebrow raising is legit. It doesn't have to mean anything further if
that's what you're concerned about.

--
http://memedepot.com/uploads/2000/2098_steamboat.gif

[Court_1]

> The point here is that the mere fact that an event appears very improbable from one perspective does not
> mean that a "fix" is to be inferred.

By the same token the mere fact that an event appears very improbable
doesn't mean a fix is NOT to be inferred.

Whether or not tournaments fix their draws I don't know, but one would
be naive not to consider the business and econmonic side of fixing
draws. Where there is big money there is often corruption of some kind.

[Joe Ramirez]

No one knows -- that's why reliance on numerology is usually futile in
situations like this one.

[Joe Ramirez]

Huh? That's a perverse way of formulating the issue.

>
> Whether or not tournaments fix their draws I don't know, but one would
> be naive not to consider the business and econmonic side of fixing
> draws. Where there is big money there is often corruption of some kind.

Yes, which is why I say that if there is any actual evidence of draw
manipulation, let's investigate full-throttle. But no evidence exists
here, as far as I am aware. The world's investigative resources are
not unlimited, and possibly statistically anomalous tennis draws make
a very weak case for applying them.

[Joe Ramirez]

Your conclusion is shaped by what I consider an artificially narrow
definition of the sample set. But I can't blame you for that, because
the author of the study chose to go in that direction. Others in this
thread already have pointed out the unwarranted omission of the French
Open. Why was it left out? *To tailor the database to promote the
desired "suspicious" result.* For that matter, why were Masters events
left out? Again, to tailor the database to promote the desired
"suspicious" result. The smaller your set, the easier it is to cry
foul when *any* kind of pattern is detected.

In considering these kinds of slightly paranoid statistical analyses
in general, I would say that the fundamental flaw is the implicit
belief that the "improbabilities" we encounter in the world should be
more or less evenly distributed throughout our experience. What this
means, to put it simply, is that even the decision to limit the number-
crunching to *tennis* may be philosophically illegitimate. I.e., maybe
the existence of a wacky tennis streak is "compensated for" by an
unusually high level of randomness to be found in the weekly
assortment of vegetables one encounters at the local supermarket. This
doesn't mean that the calculation of probability here is
mathematically incorrect, merely that by itself it is not particularly
meaningful. As I noted, it's a "Wow" -- then forget about it.

[TT]

30.10.2011 3:55, Joe Ramirez kirjoitti:
> Others in this
> thread already have pointed out the unwarranted omission of the French
> Open. Why was it left out? *To tailor the database to promote the
> desired "suspicious" result.*

Yes.

Djokovic was on same side as Federer on 10/12 slams during 2008-2011.

The odds for Djok being on same side 10 or more times of 12... is
0.01929 aka 1/52, according to cumulative binomial distribution.

...Or twice the odds if one thinks that Djokovic being as much on
opposite side of Federer being equally impressive. Aka 1/26.

And that's not even including that this "series" could have begun earlier.

[TT]

Fine... But actually Djokovic was on same side 14/16 slams! Duh.

That makes the odds of 0,00209...which is 1/478
...or 1/239 considering Djokovic could have been 14/16 on opposite side
of the draw.

That's not so small. Still during whole history of tennis it's not that
big either.

[Court_1]

> a very weak case for applying them.-

I don't think my statement is perverse at all. Just because we can't
see something does not mean it is non-existent. Many people believe in
God but what proof do they have God exists?
My personal opinion is that even though we have no concrete evidence
of draw fixes I am fairly certain they exist in the big tournaments
where tennis is essentially a big business and certain match-ups mean $
$$$$$.
I think to think otherwise goes against all business sense.

[Joe Ramirez]

Certainty without evidence is the domain of either metaphysics or
delusion. Either way, it ends the discussion.

> I think to think otherwise goes against all business sense.

"Business sense" is not the only factor that determines whether
illegal conduct occurs.

[Superdave]

On Sun, 30 Oct 2011 20:56:33 -0700 (PDT), Joe Ramirez
<josephm...@netzero.com> wrote:

>On Oct 30, 9:49�pm, Court_1 <Olympia0...@yahoo.com> wrote:
>> On Oct 29, 8:36�pm, Joe Ramirez <josephmrami...@netzero.com> wrote:
>>
>>
>>
>>
>>
>> > On Oct 29, 7:53�pm, Court_1 <Olympia0...@yahoo.com> wrote:
>>
>> > > On Oct 29, 6:56�pm, Joe Ramirez <josephmrami...@netzero.com> wrote:
>>

Correct.

BIG payoffs, Secrecy, Murder Threats and other things (such a crooked lawyers)
also play a part.

[Superdave]

Of course there are fixed draws !

Never mind business sense.

Mathmatics does not lie !

It can be statistically proven that draws are fixed if somebody is willing to
pony up the cost of proving it.

I mean lawyers like Joe have used LESS "proof" to send people to the electric
chair ! Yet, they claim it's NOT enough to prove draw fixing?

Duh?

[TT]

Or prolongs it, as we've seen before with court1. Dingbats don't need
proof, they're better off without!

[Court_1]

On Oct 31, 8:49 am, Superdave <DaveHazelw...@remail-it.net> wrote:

Not sure that will ever happen in reality--i.e. the proof we are
looking for will come. I won't hold my breath.

[Court_1]

Says the "extremely reasonable individual" who obsesses over two very
important people in the world--i.e. the teenybopper idol Rafael Nadal
and the dead pedo freak Michael Jackson!!!! ;)

[Court_1]

> illegal conduct occurs.-

No of course business sense is not the only factor that determines if
illegal conduct occurs but in the case of a possible tennis draw fix
economic reasons would indeed be THE determining factor.

[Pelle Svanslös]

There already is an existence proof. USO 96.

The IMG and Teddy Forstmann, the 100% honourable huggy bear, control
everything in tennis.

> Never mind business sense.
>
> Mathmatics does not lie !

Some people using that shit do.

> It can be statistically proven that draws are fixed if somebody is willing to
> pony up the cost of proving it.

Even when the event is 1/500, somebody always pops up saying "yeah, but
we don't know in what *order* they come up". Sure sure sure.

--
http://memedepot.com/uploads/2000/2098_steamboat.gif

[Iceberg]

it's fairly obvious to me that Fed and/or the Fedfans have paid off
the drawsmen for years, esp against Nadal.

[Iceberg]

pretty much all run by 'business men' and crooker lawyers.

[Iceberg]

yes, that's hilarious. it's like Tony Bliar saying he didn't know the
Iraq dossier was all bullshit.

[jdeluise]

On 2-Nov-2011, Iceberg <iceber...@gmail.com> wrote:

> it's fairly obvious to me that Fed and/or the Fedfans have paid off
> the drawsmen for years, esp against Nadal.

So when they were seeded 1 and 2, the "drawsmen" were paid off to put each
player on the opposite half? LOL

[Pelle Svanslös]

There's actually 4 draws in the study that could have put Rogi and Rafa
on the same side of the draw. You guessed it, none of them materialised.

If you forget about names for the moment, out of the 6 possibilities
that could have resulted in a lopsided draw, three of the fab 4 in the
same side, only one materialised. They narrowly, 5/6, escaped the
nightmare draw from the financial POV. One of the big 4 dropping out in
the QFs and one certified chump in the SFs.

After the only lopsided draw, RG2008, it's 5/5.

--
http://memedepot.com/uploads/2000/2098_steamboat.gif

[Iceberg]

On Nov 3, 4:02 pm, Pelle Svanslös <pe...@svans.los> wrote:
> On 2.11.2011 22:39, jdeluise wrote:
>

> > On  2-Nov-2011, Iceberg<iceberg.ru...@gmail.com>  wrote:
>
> >> it's fairly obvious to me that Fed and/or the Fedfans have paid off
> >> the drawsmen for years, esp against Nadal.
>
> > So when they were seeded 1 and 2, the "drawsmen" were paid off to put each
> > player on the opposite half?  LOL
>
> There's actually 4 draws in the study that could have put Rogi and Rafa
> on the same side of the draw. You guessed it, none of them materialised.
>
> If you forget about names for the moment, out of the 6 possibilities
> that could have resulted in a lopsided draw, three of the fab 4 in the
> same side, only one materialised. They narrowly, 5/6, escaped the
> nightmare draw from the financial POV. One of the big 4 dropping out in
> the QFs and one certified chump in the SFs.
>
> After the only lopsided draw, RG2008, it's 5/5.

jdeluise always reckons we should give everyone, even OJ, the 'benefit
of the doubt'. This draw fixing thing can be summarised that it's
about as likely that the draws weren't fixed as that Djoker has become
by far the world's fittest man, way beyond ANYBODY else, purely by
using a 'gluten free' diet.

[Joe Ramirez]

You and Scott need to bottle and sell the secrecy potion used by draw
fixers, lunar hoaxers, and 9/11 hologrammers. It would sell very well
in the rest of the world. (Brand name: "Jimmy Hoffa Juice"?) Here in
the United States, a trade association exec can be accused of sexual
harassment, and even if he pays off the accusers in cash, makes them
sign confidentiality agreements, and buries the whole thing with the
aid of the group, the accusations will *still* surface and create a
scandal when he runs for president 15 years later! Somehow our
political conspirators have lost the knack of covering things up
effectively, unlike tennis moguls and astrophysicists.