Rating Players Basic theory
#41
Posted 2010-June-10, 11:03
#42
Posted 2010-July-24, 21:42
#43
Posted 2010-July-25, 03:51
Find out which rating system is most accurate, by means of cross-validation. If you remove a single board* from the database, a good rating scheme is one that can predict the IMP or MP score on that board, from a statistical analysis of the remaining boards.
There is an issue with partnership rating vs individual rating. So an alternative project could be: remove all hands played by a particular partnership and try to predict their performance from an analysis of the remaining boards (in particular those that they played with other partners).
Such a study would give clues to how accurate a rating based on total points would be relative to one based on IMP relative to PAR or to IMP relative to the hand played by robots, or to IMP relative to the BBO field.
*by "board" I mean board x played by players a,b,c,d, i.e. not all the times board x was played, just one of them. You do this for many (ideally all) boards in the database, then compute the average prediction error.
#44
Posted 2010-July-25, 10:01
#45
Posted 2010-July-25, 11:38
#46
Posted 2010-July-25, 12:48
djeast, on Jul 25 2010, 07:01 PM, said:
Comment 1: There are 15K people logged in right now. Feels pretty viable to me...
Comment 2: BBO doesn't cost anything, so its hard to understand how knitting is going to save to any money
Comment 3: You seem to be laboring under the belief that there is some relationship between "attendance points" and skill. If this is the case, then you are probably beyond hope.
Comment 4: BBO ain't the ACBL
#47
Posted 2010-July-25, 13:15
#48
Posted 2010-July-26, 02:26
djeast, on Jul 25 2010, 12:38 PM, said:
LOL
Now, i am not sure about this, but i was under the impression that bridge was a partnership game.
Excluding pairs that play and practice together (THE VAST VAST VAST VAST VAST VAST VAST VAST (that's a lot of VASTs) majority of who DO NOT CHEAT) is ridiculous. it randomizes the field, causes bad and high variance bridge and rewards good luck rather than good actions.
This and your previous post make it pretty clear that you are exactly the sort of person I prefer not to have to meet at a bridge club; someone who loves to result, blame others, and argue. how's that masterpoint attendance award coming along?
#49
Posted 2010-July-26, 08:40
djeast, on Jul 25 2010, 12:01 PM, said:
Promises, promises.
So many experts, not enough X cards.
#50
Posted 2010-July-26, 09:48
bid_em_up, on Jul 26 2010, 07:40 AM, said:
djeast, on Jul 25 2010, 12:01 PM, said:
Promises, promises.
LOL
#51
Posted 2010-July-28, 08:39
#52
Posted 2010-July-28, 09:30
sallyd, on Jul 28 2010, 09:39 AM, said:
Play with people you know.
#54
Posted 2010-July-28, 14:22
sallyd, on Jul 28 2010, 03:00 PM, said:
Sally,
Yup, it's a jungle out there. The response from TimG is correct I think. You either have to make a list of 'friends' on BBO or take your chances in the Main Bridge Club.
When I drop in the Main Bridge Club, I look at pard's profile first. If no information, or no system profile, or pard is -50 IMPS, I leave.
Good luck.
#55
Posted 2010-July-28, 15:28
Suppose that I have access to a large number of results from matchpoint events at a local club. I would like to apply some semblance of a sensible rating system to this data. I've seen a number of rating systems proposed (the Lehman system being prominent, but I'm aware of others).
Where can I get software which will, given the event results in some reasonable format, compute the ratings for each player according to some system? Or do I need to write this software from scratch?
a.k.a. Appeal Without Merit
#56
Posted 2010-July-29, 14:17
awm, on Jul 28 2010, 04:28 PM, said:
Suppose that I have access to a large number of results from matchpoint events at a local club. I would like to apply some semblance of a sensible rating system to this data. I've seen a number of rating systems proposed (the Lehman system being prominent, but I'm aware of others).
Where can I get software which will, given the event results in some reasonable format, compute the ratings for each player according to some system? Or do I need to write this software from scratch?
Hi,
not sure, if this answers your question, and is helpful at all, but the scoring
program ruderyv has a rating module for club tournments:
http://www.rudersyv.de
It seemed to be reasonaable, our club uses the program, but we did not activate
the rating option.
=> Other similar programs may also have such a feature.
With kind regards
Marlowe
Uwe Gebhardt (P_Marlowe)
#57
Posted 2010-July-29, 14:22
sallyd, on Jul 28 2010, 03:00 PM, said:
Hi,
one possible option is to join the IAC club tournaments.
http://pigpen.org.uk/IAC/iac.php
In general, if you play in tournaments with a fairly well defined player set,
the niveau will be better than the niveau in the MC.
You will also have a better chance to find the same people, and if you start
a tournament you will play with your p the whole tournment (most of the time),
i.e. you will have a chance to get to know each other.
With kind regards
Marlowe
Uwe Gebhardt (P_Marlowe)
#58
Posted 2010-August-07, 08:14
#59
Posted 2010-August-07, 17:11
awm, on Jul 28 2010, 01:28 PM, said:
Suppose that I have access to a large number of results from matchpoint events at a local club. I would like to apply some semblance of a sensible rating system to this data. I've seen a number of rating systems proposed (the Lehman system being prominent, but I'm aware of others).
Where can I get software which will, given the event results in some reasonable format, compute the ratings for each player according to some system? Or do I need to write this software from scratch?
Doesn't answer your question directly but you may want to contact Chris Champion -
http://www.colorados...idge.com/pr.htm
Where were you while we were getting high?
#60
Posted 2010-October-29, 16:59
I do think, however, that bridge knowledge is at least an indicator of performance at the table.
What if a standardized test(s) were developed that could be taken online and scored automatically? Such a test could cover the basic bidding of various common systems ..... in my opinion to be rated an expert, a player should have at least a rudimentary knowledge of Standard American, SAYC, Acol, 2/1, and Precision.
Such a test should also include a section on play of the hand including basic leads, defensive signals and carding, how to play basic card combinations, etc.
Also the test should cover the more common conventions.
Rating would then be based on overall score acheived on the test.
The test should be available to be taken online and players should be allowed to retake it as often as they wish.