[helene_t]

FWIW, I found the lead of ♣Q completely normal. The second example is a lot stranger, allthough "cheating" would not be my favorite theory (then again, while I suspect cheating appr. once or twice a weak at the local club, I have never encountered anything suspious on BBO, other than overt cheating such as double-shooting TD calls).

This just shows that for a statistical analysis the collected evidence must be based on objective criteria. Subjective criteria like "this is strange" could be confounded by a (possible subconscious) effect of the kind: "There was a kibber watching us so opps could have been cheating, I'll report this one".

Of course, objective criteria could be confounded as well (there are probably other differences between the WP Refugee and ACBL Tourneys than the allowance of kibbers) but if something statistically significant could be found at least we would know that something is going on, and could proceed to discussing possible confounders.

[hrothgar]

officeglen, on Feb 1 2007, 03:59 PM, said:

Here's an example from when I was watching:

If this is a representative example of the type of evidence you use to bolster your case, then you probably want to keep your mouth firmly shut.

I don't think that there is anything particularly strange about a club lead on the first hand. I'm not sure if I'd lead a club or a spade. (I suspect that I'd lead a Spade). However, I certainly wouldn't get bet out of shape if the opponents

(A) Lead a Club
(B) It happened to work

On the second hand, you don't provide any information about the bidding. For example, did the player open 3NT in first second seat or in fourth? (Big difference between these two opening bids) Moreover, it doesn't sound like the 3NT opening lead to a very good contract. 3NT might make on a heart lead and continuation, however, I suspect that it could have just as easily gone down on a different lead.

As Helene notes: The distinquishing characteristic of online tournaments is their relatively short length. In turn, this virtually requires players to adopt extremely high variance strategies. You need to roll the dice every chance that you get and then pray that the card gods are smiling. Personally, I think that this type of risk taking offers a much more convincing argument for both the bids in question.

If you want to make a serious case about cheating than you need to conduct a real statistic study. I very much suggest that you aquaint yourself with standard concepts like

"Confidence intervals"
"sample size"
"statistical significance"

This posting is brought to you by the number 1.96 and the letter "T"

[glen]

Be aware I have a degree in math, so you could consider the probability that I might know those terms (to assist with your probability, I should also reveal that I developed a statistical program purchased by Fortune 500 companies, worked extensively with SAS, created sampling PL/SQL in Oracle, blah blah blah).

I'm not making a "serious case about cheating" with two examples - how could one (also since one involved me)? I'm talking about perceptions and using two examples that might be perceived one way (e.g. see Helene on my 3NT) or the other.

"Moreover, it doesn't sound like the 3NT opening lead to a very good contract" - yes, it was a terrible contract, no play at all regardless of lead.

"this virtually requires players to adopt extremely high variance strategies" - also the fields have a high mix of skill levels - thus one will see a lot of stuff that appears strange. If Kibitzers are allowed, some will conclude that this could be the reason for some of the strange stuff. Not allowing kibitzers will rule out people taking this conclusion based on their perceptions.

[pbleighton]

Glen:

Unlike some posters who complain about cheating, I have found your posts pretty reasonable.

However, I think that you share a misperception with other posters, namely that some people who don't play "standard" bridge are "up to something".

I don't play much with pickup partners any more, but what I remember from their bidding and play makes me dismiss thoughts of cheating when opps do spectacularly weird things and get a good result.

I'm not saying that cheating doesn't exist, but I think that 99% (a made up number, to be sure) of these bids and plays are just the result of self-taught bridge.

I'm fixed, next hand.

Peter

[jtfanclub]

I know of one person (who reads the forums, so they'll post their name if they wish) who, when they see something suspicious repeatedly from the same person, will go through their history, and check for 'off' successful things that happen for the person and the overall result (somebody who repeatedly does strange successful things but even more often does strange unsuccessful things is just going anti-field). I've seen the evidence, and I am convinced, but we're still talking only a handful of people.

I don't think that banning kibbitzing will even slow these people down. Their most popular method seems to be Messenger, or something equivalent.

[jocdelevat]

I like to kibitz however I sugested before that in Yahoo bridge you have the option to hide your hand from kibitzers. I think this will improve the safety of players who are concern about kibitzers.

[uday]

What I've learned from all our experiments w/speccing:

1. Most of the players don't care (or don't tell us they care) whether specs are on/off

2. The people who do care care very, very, much. Some people stopped playing in games when we blocked specs. Others when we allowed some specs back in.

I will add to this a third thing that we all knew all along

3. Blocking specs only prevents a specific form of cheating (self speccing). Collusive cheating cannot be blocked.

and a fourth thing that that we probably suspected

4. In this debate, the wishes of the players carry much more weight with me than the wishes of the casual non-playing spectators


Someday, perhaps we'll split the ACBL tourneys into 2 halves; one with specs allowed and one (using different hands) with specs disallowed and see what happens.

Until then, there is likely no solution that will please the parties on both sides of the debate.

The current approach -- allow specs if the specs have won 5+ points on BBO -- seems as reasonable as anything to me for the moment. I might harden this requirement over time, requiring (say) 20+ points to spec or whatever but I have no immediate plans to do so.

[glen]

pbleighton, on Feb 1 2007, 09:22 AM, said:

However, I think that you share a misperception with other posters, namely that some people who don't play "standard" bridge are "up to something".

Why would I share this "misperception" given I opened 3NT on the hand above - wouldn't somebody who opens 3NT on the hand above, and does lots of other bids out-of-the-norm* actually be more accepting of non-"standard" stuff?

* I would do even more if psyches were not frowned on.

[glen]

uday, on Feb 1 2007, 12:38 PM, said:

Some people stopped playing in games when we blocked specs.

This I did not know - that customers were becoming non-customers if kibitzers were not allowed. Given this is the case, and that I will not become a non-customer if kibitzers continue to be allowed, then I retract my call to have kibitzers banned. Certainly I've had good entertainment value in watching friends play in the tourneys.