[1eyedjack]

kaydea, on Sep 26 2009, 05:50 PM, said:

So is such a system possible on BBO? Perhaps this question could be answered by the programmers.

It could be answered by the programmers. And has been. At length. Repeatedly.
Yes it is possible, but a decision was taken not to implement it.

BBO came along well after many competing providers, and they programmed it in the full knowledge of how the competition operates. It was a conscious decision to go down this road, not oversight, and carefully considered.

I doubt that they expected universal endorsement: they would never get everyone's agreement on this, whichever route they went down.

Lack of honesty in a self-rating system only becomes a problem if you mistakenly assume and then rely on honesty. No-one who has been on BBO for more than a few weeks pays any attention to these flags.

[babalu1997]

1eyedjack, on Sep 27 2009, 07:01 AM, said:

kaydea, on Sep 26 2009, 05:50 PM, said:

So is such a system possible on BBO? Perhaps this question could be answered by the programmers.

It could be answered by the programmers. And has been. At length. Repeatedly.
Yes it is possible, but a decision was taken not to implement it.

BBO came along well after many competing providers, and they programmed it in the full knowledge of how the competition operates. It was a conscious decision to go down this road, not oversight, and carefully considered.

I doubt that they expected universal endorsement: they would never get everyone's agreement on this, whichever route they went down.

Lack of honesty in a self-rating system only becomes a problem if you mistakenly assume and then rely on honesty. No-one who has been on BBO for more than a few weeks pays any attention to these flags.

The rating thread should be pinned, and every new member required to post on it as their introduction.

As one of our members here pointed out at a hot rating thread at another site, this decision is the likely the reason why you log on to BBO to find thousands of tables and many tournaments and, when you log on to other sites you find, wow, 100 players logged on, and yes A (as in one) daily tournament.

I remember playing one game for Lehmans being new at the site, the ratings got updated that same evening, to 49.80 from 50. On Monday, I could not find a table, sorry, 50 plus required, a +/- 0.000002 precision so as not to create another Chernobyl.

Play with the idiotic expert, the experience will give you a good light conversational topic.

I do not think any world class player logs on to wait for me, so those who invite are immediately suspect.!!!!1

[kaydea]

Lots of good arguments against a rating system. I know the Lehman rating system isn't perfecct either and some advanced players on OKB played against beginners to advance their points, not to mention bailing out of a game if they had a good score.

But re the singleton lead against a NT contract, the only situation that might make sense to me would be leading a singleton in a suit that my p had bid.

Anyway, I have met some great players I am happy to play with and against on BBO. Players who rate themselves honestly, and at the end of the day its about personal integrity.

[matmat]

i wouldn't want to see a rating system. in fact, i think the current set of categories is pretty much a shambles anyway.

a change that might not be the worst thing in the world would be to replace the current ratings with a list of reasons and places where a particular player likes and does to play.

less room for ego and possibly a simpler way to connect with others that look for the same thing...

do they like to play online only? clubs? tournaments? do they prefer a social game? a competitive game? pro game?

I'm also sure that this approach wouldn't solve most problems.

[pooltuna]

matmat, on Sep 27 2009, 11:51 AM, said:

i wouldn't want to see a rating system. in fact, i think the current set of categories is pretty much a shambles anyway.

a change that might not be the worst thing in the world would be to replace the current ratings with a list of reasons and places where a particular player likes and does to play.

less room for ego and possibly a simpler way to connect with others that look for the same thing...

do they like to play online only? clubs? tournaments? do they prefer a social game? a competitive game? pro game?

I'm also sure that this approach wouldn't solve most problems.

WAIT! Does this mean that bridge is not a euphemism for ego?

[quiddity]

Hate the idea of a rating system for regular play, for all the reasons awm mentioned. I do think rated robot-races could work.

[olegru]

kaydea, on Sep 27 2009, 10:13 AM, said:

But re the singleton lead against a NT contract, the only situation that might make sense to me would be leading a singleton in a suit that my p had bid.

If your partner did something you would never do there are at least two explanations possible.
First. Partner does not know what he is doing.
Second. You are not good enough to understand what your partner is doing.
No guaranties, of cause, but there is a method how to get some ideas which case is more probable. Press this link: http://online.bridge...hands/index.php
Type your nickname and look at your summary for a month. If in average you are losing ~0.5 imps per board - chances are there is something in this game you do not understand yet.
You can also check your partner’s stats to see if he is the person you would like to learn from or his ideas does not appear to be very successful.

BTW: Cases when bidding and cards require to lead singleton in unbid suit against a NT contract are exist.

[kaydea]

olegru, on Sep 28 2009, 09:23 AM, said:

kaydea, on Sep 27 2009, 10:13 AM, said:

But re the singleton lead against a NT contract, the only situation that might make sense to me would be leading a singleton in a suit that my p had bid.

If your partner did something you would never do there are at least two explanations possible.
First. Partner does not know what he is doing.
Second. You are not good enough to understand what your partner is doing.
No guaranties, of cause, but there is a method how to get some ideas which case is more probable. Press this link: http://online.bridge...hands/index.php
Type your nickname and look at your summary for a month. If in average you are losing ~0.5 imps per board - chances are there is something in this game you do not understand yet.
You can also check your partner’s stats to see if he is the person you would like to learn from or his ideas does not appear to be very successful.

BTW: Cases when bidding and cards require to lead singleton in unbid suit against a NT contract are exist.

The person who lead the singleton was a partner in an indy tourney so it is unlikely that we would ever play together again. In retrospect perhaps the only circumstances in which a ratings would be useful would be in tourneys and I would think most people who play in ACBL tourneys play with regular partners anyway.

But thanks for your comments, I will certainly pursue NT leads which for me up until now have consisted of top of a sequence, 4th highest of longest and strongest or my p's suit. Obviously I still have a ways to go.

[Trinidad]

There are situations where I would lead my void against NT, if it weren't for the fact that...

Rik

[billw55]

eh, thread too long to read all posts.

But I can describe how my own BBO self-rating process worked:

Started with "intermediate" which seemed reasonable. Was overwhelmed by the astonishing quantity of gross incompetence by players rated "advanced" and "expert". Noticed that most "intermediates" seemed fairly weak. Adjusted myself to "advanced" on basis of comparison to others in the BBO population. "Expert" was tempting but I felt that would make me part of the problem.

Even now, a couple years later, I still mark ops and/or pickup partners as "not advanced" or "not expert" almost every time I play. IMO it is way too rampant to be mostly honest errors.

[1eyedjack]

kaydea, on Sep 28 2009, 03:53 PM, said:

I will certainly pursue NT leads which for me up until now have consisted of  top of a sequence, 4th highest of longest and strongest or my p's suit.

I realise that it is rather moving away from the subject of the thread, but have you not just answered your own point here? Emphasis added by me, but the point being - if it is your partner's suit, it is OK to lead it even if it may be singleton. Sure, there may be other factors in play.

[Quantumcat]

I don't understand why people feel the need to self-rate above their actual skill level. You'd think it would be better for people to be pleasantly surprised at your ability rather than disappointed in it.

Another advantage to self-rating below your skill is that if you say you are a beginner when you're not, is that your false-cards will work. An expert won't think a beginner capable of making it and then fall into your trap.

Another example could be this. Imagine defending a slam where you win the first trick and have to decide the next card. If declarer has a large range of hands compatible with the bidding, you'll be squeezed later. You have to knock out an entry so the squeeze can't work. But if declarer has one or two particular hands there's no squeeze, and you will give away the contract by leading that card. If you know the person is capable of executing the squeeze, there are more hands he can have where the squeeze happens so you'd try to stop it. But if declarer is a beginner you'd just do your best to not give away the contract since they probably will just cash 11 tricks and be disappointed when they realise there's no twelfth. So if you are not really a beginner you will make the hand!

[matmat]

i led a singleton against NT yesterday in a suit p didn't bid... it led to declarer going down 4!

[matmat]

matmat, on Sep 29 2009, 01:57 AM, said:

i led a singleton against NT yesterday in a suit p didn't bid... it led to declarer going down 4!

it just struck me that this is actually an argument that only bad players do this.

[billw55]

Fluffy, on Sep 27 2009, 05:34 AM, said:

Rating systems are not good, people don't like to fight for a %. I won't like beginners being hunted down by advanced plaers to improve their rating.

Several people mentioned this. From my experience in other games (chess, Go) I would expect this plan to backfire. Consistently playing against substantially weaker opponents will make oneself play weaker. And the small rating gain from trouncing the lowbies may well be less than the loss from subsequent bad play against one's former equals.

[kaydea]

1eyedjack, on Sep 28 2009, 02:46 PM, said:

kaydea, on Sep 28 2009, 03:53 PM, said:

I will certainly pursue NT leads which for me up until now have consisted of  top of a sequence, 4th highest of longest and strongest or my p's suit.

I realise that it is rather moving away from the subject of the thread, but have you not just answered your own point here? Emphasis added by me, but the point being - if it is your partner's suit, it is OK to lead it even if it may be singleton. Sure, there may be other factors in play.

Not necessarily..how would my p know this was my suit if we haven't taken part in the auction?

[matmat]

kaydea, on Sep 29 2009, 07:39 AM, said:

1eyedjack, on Sep 28 2009, 02:46 PM, said:

kaydea, on Sep 28 2009, 03:53 PM, said:

I will certainly pursue NT leads which for me up until now have consisted of  top of a sequence, 4th highest of longest and strongest or my p's suit.

I realise that it is rather moving away from the subject of the thread, but have you not just answered your own point here? Emphasis added by me, but the point being - if it is your partner's suit, it is OK to lead it even if it may be singleton. Sure, there may be other factors in play.

Not necessarily..how would my p know this was my suit if we haven't taken part in the auction?

x
xxxx
xxxx
xxxx


1N-2C;
2D-3N

where are the spades?

[gnasher]

With a properly designed rating system (which is what the Lehman system is claimed to be) playing against weak players shouldn't help at all. If the system predicts that you will score 60% against a given pair, and you do score 60% against them, your rating won't improve.

To artificially inflate your rating, you have to find players who are currently overrated. One category of such players is beginners who are also newcomers to the rating system, and have been given an average rating to start with. The solution to that is easy: after a new member has been playing for a few weeks, recalculate the ratings of everyone he has played against, but assume that his rating at the time was the same as his current rating.

[Fluffy]

gnasher, on Sep 29 2009, 01:13 PM, said:

To artificially inflate your rating, you have to find players who are currently overrated. One category of such players is beginners who are also newcomers to the rating system, and have been given an average rating to start with.

Or beginners who play against beginners.

[pooltuna]

gnasher, on Sep 29 2009, 08:13 AM, said:

With a properly designed rating system (which is what the Lehman system is claimed to be) playing against weak players shouldn't help at all.  If the system predicts that you will score 60% against a given pair, and you do score 60% against them, your rating won't improve.

To artificially inflate your rating, you have to find players who are currently overrated.  One category of such players is beginners who are also newcomers to the rating system, and have been given an average rating to start with.  The solution to that is easy: after a new member has been playing for a few weeks, recalculate the ratings of everyone he has played against, but assume that his rating at the time was the same as his current rating.

StepBridge has the best (most usable?) rating system I have seen. Basically when you join you are assigned a rating of 1-1(MP-IMP) with rating corrections after every 20 hands (increase or decrease 1) and no bunny bashing here if you had a 5-4 rating playing with a like partner against 2 newbies you would be spotting 1.5 IMPS/hand (difference between IMP ratings * 0.25) and you need to average more than 0.5 IMPs/hand to increase.