BBO Discussion Forums: Who will stop Italy? - BBO Discussion Forums

Jump to content

  • 5 Pages +
  • 1
  • 2
  • 3
  • 4
  • Last »
  • You cannot start a new topic
  • You cannot reply to this topic

Who will stop Italy?

Poll: Who will stop Italy? (51 member(s) have cast votes)

Who will stop Italy?

  1. Somebody. (33 votes [64.71%])

    Percentage of vote: 64.71%

  2. Nobody. (18 votes [35.29%])

    Percentage of vote: 35.29%

Vote Guests cannot vote

#21 User is offline   Gerben42 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 5,577
  • Joined: 2005-March-01
  • Gender:Male
  • Location:Erlangen, Germany
  • Interests:Astronomy, Mathematics
    Nuclear power

Posted 2007-October-05, 05:24

Why is USA 1 called the "Cinderella" team? Why is any team called that?

Anyway,

Italy: 62
USA1: 0

ouch!

In a KO system you always have to multiply probabilities, which makes it hard even for Roger Federer to win ALL tournaments he plays in.

Hup Holland hup! :)
Two wrongs don't make a right, but three lefts do!
My Bridge Systems Page

BC Kultcamp Rieneck
0

#22 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,402
  • Joined: 2004-April-22
  • Gender:Female
  • Location:Odense, Denmark
  • Interests:History, languages

Posted 2007-October-05, 07:31

I have fitted my model to the 18 first rounds. These are the probabilities that Italy will beat each of the teams in a 128-board match:

[1] "ARGENTINA 0.953012302526452"
[1] "AUSTRALIA 0.868122124683079"
[1] "BRAZIL 0.97464322042772"
[1] "CANADA 0.999338396916119"
[1] "CHINA SMEG 0.931933653485224"
[1] "CHINESE TAIPEI 0.991064605855261"
[1] "EGYPT 0.977016518343036"
[1] "INDIA 0.994100216807085"
[1] "INDONESIA 0.963919841634841"
[1] "IRELAND 0.990826825344023"
[1] "ITALY 0.5"
[1] "JAPAN 0.949946849954086"
[1] "NETHERLANDS 0.862697553887635"
[1] "NEW ZEALAND 0.99931127274922"
[1] "NORWAY 0.848930973373835"
[1] "PAKISTAN 0.989478435183908"
[1] "POLAND 0.964036923004087"
[1] "SOUTH AFRICA 0.918083617456822"
[1] "SWEDEN 0.904883807753579"
[1] "TRINIDAD & TOBAGO 0.9999968845097"
[1] "USA 1 0.80075194242031"
[1] "USA 2 0.942098374182473"
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#23 User is offline   hrothgar 

  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 15,725
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Natick, MA
  • Interests:Travel
    Cooking
    Brewing
    Hiking

Posted 2007-October-05, 07:55

helene_t, on Oct 5 2007, 04:31 PM, said:

I have fitted my model to the 18 first rounds. These are the probabilities that Italy will beat each of the teams in a 128-board match:

Silly question: My understanding is that you performed a simple linear regression based on total scores. I took a quick look at the results table that you looked at. From what I can tell, you have access to the total score for the match, as well as the number of boards. This gives you information about the mean result per board, but we don't have any informaion about the variance.

If you are generalizing your results to encompass longer matches, this variance assumption would seem to be critical. You can get arround this by assuming that the board results are normally distributed (hell, if you're using a linear regression, you pretty much made this same assumption). Even so, given how critical the variance assumption is to projecting forward to a 128 board match I'd be a bit leery.

I think that the monte carlo type techique that gerben and alex were using earlier this year is probably a better predictor than a simple extrapolation. I woder whether it will ever be possible to get board by board results.

(Justin, if you're reading this, send us all the board scores for the match between USA2 and Italy. We'll run it through the model and see what spits back out)
Alderaan delenda est
0

#24 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,402
  • Joined: 2004-April-22
  • Gender:Female
  • Location:Odense, Denmark
  • Interests:History, languages

Posted 2007-October-05, 08:09

I did not analyze this per board, only per match. I used the IMPs rather than the VPs because they are more accurate and because the linear model is more plausible for IMPs.

The residuals are normal distributed, but there could still be a number of issues related to the model, such as additivity and heteroscedasticity.

As for extrapolating the variance, I computed the residual variance from the 16-board matches. Now if a 128-board match can be seen as the aggregate of 8 independent 16-board matches, the variance and the IMP saldo are both expected to grow linearly, so the CV must be corrected by a factor sqrt(1/8). That's what I did.
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#25 User is offline   cherdano 

  • 5555
  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 9,520
  • Joined: 2003-September-04
  • Gender:Male

Posted 2007-October-05, 08:54

Helene, did you take carry-over into account?
(It is 1/2 of the RR result if the higher-ranking team won, 1/3 if the lower-ranking team won, and a maximum of 16 for 96-board matches, a maximum of 20 for 128-board matches.)
The easiest way to count losers is to line up the people who talk about loser count, and count them. -Kieran Dyke
0

#26 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,402
  • Joined: 2004-April-22
  • Gender:Female
  • Location:Odense, Denmark
  • Interests:History, languages

Posted 2007-October-05, 08:55

Oops no, forgot that.
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#27 User is offline   drinbrasil 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 115
  • Joined: 2003-December-12

Posted 2007-October-05, 09:26

helene_t, on Oct 5 2007, 01:31 PM, said:

I have fitted my model to the 18 first rounds. These are the probabilities that Italy will beat each of the teams in a 128-board match:


Construtive comment. I think this model can be improved. Like we do in models to make odds in soccer, we use some other aspects who in bridge i imagine could be at least the recent h2h (very better information, because in soccer we have many player changes in teams). So you need use weight for actual form (here you used just the RR) and weight for H2H. I imagine h2h is important because we know in bridge our game fit more or less against one or other opponent due to systems or styles for example.
Occam's razor: "When you hear hoof beats, think horses, not unicorns."
0

#28 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,402
  • Joined: 2004-April-22
  • Gender:Female
  • Location:Odense, Denmark
  • Interests:History, languages

Posted 2007-October-05, 09:32

Yes, my brother who is in fisheries research often uses adaptive filters for similar problems, basically it means exponentially decaying weights.

As for H2H I dono how to estimate it. Is it simply taking the RR-result of Italy vs. X into account? It's difficult to see how I could estimate the weights for the H2H without repeat measurements. Of course if used per-board info I would have repeated measurements. I would have to write a script to download the scoreboards then, don't like the idea of copy-pasting each board.
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#29 User is offline   drinbrasil 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 115
  • Joined: 2003-December-12

Posted 2007-October-05, 10:52

For H2H you should use all last lets say X* year(s) results between teams plus some weight for RR 16boards result this year. Of course perfect would be board a board all time, but general results can be used.

*This isnt fixed number, in really you should use many in past datas to make model who fit more for bridge. Of course i am only saying about ideal models, i cant use my time to do this actually, but if you have time to take previous meetings between all teams in past events, you much probabily will get better estimations. Maybe someone has this work partial done around, like data from 2003-2005 and used for 2006 torneys.
Occam's razor: "When you hear hoof beats, think horses, not unicorns."
0

#30 User is offline   cherdano 

  • 5555
  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 9,520
  • Joined: 2003-September-04
  • Gender:Male

Posted 2007-October-05, 11:00

Btw, if you think you understand those odds better than others, you can make a little money on betfair.com :) (Actually not really, the margins between the odds you can get for backing or have to give for laying are way too big.)
The easiest way to count losers is to line up the people who talk about loser count, and count them. -Kieran Dyke
0

#31 User is offline   drinbrasil 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 115
  • Joined: 2003-December-12

Posted 2007-October-05, 11:09

little money isnt interesting :)
Occam's razor: "When you hear hoof beats, think horses, not unicorns."
0

#32 User is offline   skjaeran 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 3,727
  • Joined: 2006-June-05
  • Location:Oslo, Norway
  • Interests:Bridge, sports, Sci-fi, fantasy

Posted 2007-October-05, 12:15

helene_t, on Oct 5 2007, 03:31 PM, said:

I have fitted my model to the 18 first rounds. These are the probabilities that Italy will beat each of the teams in a 128-board match:

[1] "NORWAY 0.848930973373835"

84.9% probability that Italy will beat Norway in a head to head match?
That's ridicolous. They've met in matches like this several times before. Those matches has as a rule been very close. Since 1995 the two teams have met three times at the KO stage of the BB. Norway have won twice, Italy once.
Italy probably are favourites if they meet again, but not by much - maybe 55-60%.

And your probabilitiies are likewise way too high for several other teams too.
Kind regards,
Harald
0

#33 User is offline   joker_gib 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,384
  • Joined: 2004-February-16
  • Location:Belgium

Posted 2007-October-05, 12:25

I agree with Harald, this Norvegian team is a very good one !

Alain
Alain
0

#34 User is offline   Gerben42 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 5,577
  • Joined: 2005-March-01
  • Gender:Male
  • Location:Erlangen, Germany
  • Interests:Astronomy, Mathematics
    Nuclear power

Posted 2007-October-05, 12:31

Copy and paste argument for Netherlands :)
The 1993 final replay sounds appealing!
Two wrongs don't make a right, but three lefts do!
My Bridge Systems Page

BC Kultcamp Rieneck
0

#35 User is offline   Fluffy 

  • World International Master without a clue
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,404
  • Joined: 2003-November-13
  • Gender:Male
  • Location:madrid

Posted 2007-October-05, 13:05

Hannie, on Oct 4 2007, 03:43 PM, said:

Who will stop Italy?

No one, Go Italy!. Kick their arses.
0

#36 User is offline   helene_t 

  • The Abbess
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 17,402
  • Joined: 2004-April-22
  • Gender:Female
  • Location:Odense, Denmark
  • Interests:History, languages

Posted 2007-October-05, 13:25

skaeran, on Oct 5 2007, 08:15 PM, said:

84.9% probability that Italy will beat Norway in a head to head match?
That's ridicolous.

You're probabbly right, but why does my model say 84.9% then? Which of my assumptions are more likely to be wrong?
The world would be such a happy place, if only everyone played Acol :) --- TramTicket
0

#37 User is offline   skjaeran 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 3,727
  • Joined: 2006-June-05
  • Location:Oslo, Norway
  • Interests:Bridge, sports, Sci-fi, fantasy

Posted 2007-October-05, 14:27

helene_t, on Oct 5 2007, 09:25 PM, said:

skaeran, on Oct 5 2007, 08:15 PM, said:

84.9% probability that Italy will beat Norway in a head to head match?
That's ridicolous.

You're probabbly right, but why does my model say 84.9% then? Which of my assumptions are more likely to be wrong?

I'm no statitican, and I don't really know how you have computed your probabilites, nor what assumptions you've made. But knowing that the score in head to head competition in the KO stage in BB is Norway 2 - Italy 1 and that Norway actually beat Italy (narrowly) in their outing in the RR, common sense alone will tell you that your probability must be way off.

And even after Italy beat USA1 63-0 in the RR I can't believe the odds for an Italian victory in an hypothetical semifinal/final encounter with a 16/20 IMP carry over could be more than 60% at most.

The Italians are very good at crushing the weaker teams - just look to the European Championships (and there are "weak" teams relatively speaking in the BB too) - and thus always do well in round robins. But the knock out stage is something different. Even in long matches other strong teams have a reasonable chance to beat them, they are far from unbeatable, even if they are a really formiddable team.
Kind regards,
Harald
0

#38 User is offline   cherdano 

  • 5555
  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 9,520
  • Joined: 2003-September-04
  • Gender:Male

Posted 2007-October-05, 15:49

helene_t, on Oct 5 2007, 01:25 PM, said:

skaeran, on Oct 5 2007, 08:15 PM, said:

84.9% probability that Italy will beat Norway in a head to head match?
That's ridicolous.

You're probabbly right, but why does my model say 84.9% then? Which of my assumptions are more likely to be wrong?

Are you assuming that the Roundrobin is an accurate reflection of the strength of the teams? Anyway, it could also be that Harald's intuition is wrong for 128-board matches.
The easiest way to count losers is to line up the people who talk about loser count, and count them. -Kieran Dyke
0

#39 User is offline   Halo 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 875
  • Joined: 2006-June-08

Posted 2007-October-05, 16:03

Once you define rules of competition, there is little point whining about the consequences.
0

#40 User is offline   Echognome 

  • Deipnosophist
  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 4,386
  • Joined: 2005-March-22

Posted 2007-October-05, 16:03

helene_t, on Oct 5 2007, 11:25 AM, said:

skaeran, on Oct 5 2007, 08:15 PM, said:

84.9% probability that Italy will beat Norway in a head to head match?
That's ridicolous.

You're probabbly right, but why does my model say 84.9% then? Which of my assumptions are more likely to be wrong?

I will have to agree with Harald and Arend here. It wasn't exactly clear *how* you are modeling.

Let me give a modest proposal. Suppose that a team's performance in a head-to-head matchup is based on some combination of their overall performance in the RR and their performance in their H2H matchup (also in the RR). Then we have a factor for how well the teams match-up with each other in addition to controlling for them having a bad match (their performance in other matches will make up for it). Start with that combination at say 50% (you can try various values of course) and then see what your numbers say.

There are obviously more complex ways to do this, but I'm trying to offer a relatively simple calculation.

It would also be interesting to see what your model says about their probability of winning a 16-board match vs a 128-board match to see if the model is internally consistent with the results from the prior 16-board matches.

Were using probit to model this?
"Half the people you know are below average." - Steven Wright
0

  • 5 Pages +
  • 1
  • 2
  • 3
  • 4
  • Last »
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users