BBO Discussion Forums: Encrypted bidding - BBO Discussion Forums

They stopped because it was so infuriating to the opponents. I mean, why should you ask at T1 when you can ask at t2?

pescetom, on 2018-June-10, 15:32, said:

But what you refer to is not encrypted bidding.

Encrypted bidding is based on a key which partner has access to by looking at his own hand. An example would be

2♣-2♠
3♠-4♣

Suppose you have the agreement that the 2♠ bid shows two of the three top honours while that 3♠ bid confirms the 3rd top honour. Now both partners know each other's honour holding in spades so they can agree that the 4♣ opening can have three different meanings, depending on which of the honours is held by opener. This means that both partners will know what the 4♣ bid means but opps don't.

This is not cheating. It is just a convention which may or may not be allowed, depending on local regulations.

OK, so this is not only legal, but presumably acceptable by national jurisdictions. The examples quoted here do not give anything other than obfuscation, as the same information may be given by open means, so could be deemed by Steve as nefarious ; except perhaps for my initial kings example, which does actually have a merit in that it is simpler than other methods if you want to find all kings. I infer from the replies given here that the defenders are NOT allowed to ask which king(s) are shown.

Law 16 seems in need of a revamp. The clauses are linked by ORs, and (d) effectively says "anything goes". (But not looking at the clock, because it has to be information gained before withdrawing the cards.)

Specifically 16A2 says you may use the traits of the opponents in this, so if you play against Steve you are allowed to use all the obfuscation you can muster, where you may play straight with others.

This discussion is interesting. I was aware that encrypted carding is generally not allowed by the EBU and it had never occurred to me that you might also be able to use encrypted bids. It would seem illogical to ban encrypted carding but not ban encrypted bidding and this suggests a few questions:
- Is the failure to ban encrypted bidding an oversight? Or because it is seen as less disruptive? Or because in practice nobody uses encrypted calls?
- It seems to me quite easy to design encrypted carding methods, which will be useful and easy to operate. I quickly thought of "defender with majority of the defensive high-card strength gives attitude, defender with minority of the defensive high-card strength gives count" (may or may not be effective but ...). It seems less easy to design encrypted bidding methods that are useful but do not disrupt the requirement to find the correct strain and level. I can't imagine such a method working in a competitive auction, or a part-score auction and possibly not in a game level auction. Maybe this is the reason that the EBU doesn't bother to regulate?

Tramticket, on 2018-June-12, 05:42, said:

I suspect that you are correct and that encrypted bidding is more of an interesting though exercise then something that can be applied practically

steve2005, on 2018-June-11, 14:54, said:

I think I've heard that Meckwell play reverse suit-preference signals simply because it's unusual and declarer will have to think about it more deliberately. There's clearly no technical reason why it's better than ordinary suit preference, so psychology seems to be the only possibility.

hrothgar, on 2018-June-10, 12:35, said:

helene_t, on 2018-June-11, 20:50, said:

Consider the auction

1N-4♣(15-17 BAL; asking for aces)
4♦-5♣(0 or 4 aces; asking for kings)
5♦(0 or 4 kings).

After 4♦ it's clear to both Opener and Responder, but not necessarily to any opponent, how many aces Opener has. And since a 15-17 NT must contain at least an ace or a king, 5♦ can equivalently be described as showing

* 0 kings if Opener has 4 aces
* 4 kings if Opener has 0 aces

So is 5♦ encrypted?

Btw, an example similar to Helene's is

4♠-4N (PRE; RKC(♠))
5♦-5N (1 key card; confirming all key cards).

Here e.g. 6♣ could be given five different meanings depending which key card Opener has.

Tramticket, on 2018-June-12, 05:42, said:

IMHO there are two reasons why encrypted bidding is usually not explicitly banned (although some specific encrypted methods may be banned due to other regulations).

1) There's no compelling reason to ban them, any more than there's a reason to ban Rusinoff leads or strong club systems. Most encrypted methods are slam conventions (the two-way king ask replies being a typical example), so it's not like they are particularly disruptive.

2) System regulations create legal havoc since it is difficult to define exactly which methods would fall under the regulation. A ban on methods that are explicitly called "encrypted" would just lead to pairs disclosing their methods in different ways. OTOH, a ban on anything that is encrypted in the broad sense that it relies on partner's holding as a (secret) key, would ban lots of vanilla methods. For example:
(1♥)-1♠-(2♥)-pass
(pass)-X

In some styles, opps can be assumed to be in exactly in an 8-card fit as both would have an obligation to bid 3 if they had known of a 9-card fit. Hence, this double is a so-called Polish double: "Penalty or take-out, you will see in your own hand which it is".

Basically, the default position should be that everything is allowed. Only disallow methods that are genuinely harmful to the game AND can be defined sufficiently clearly that a ban is meaningful, easy to communicate and won't affect non-harmfull conventions.

Tramticket, on 2018-June-12, 05:42, said:

I think it all hinges on law 16A1d. "A player may use information in the auction or play if it is information that the player possessed before he took his hand from the board and the laws do not preclude his use of this information". This information already known is the name of partner's dog. No law I see precludes using this information.

Therefore it can easily be used in free bidding. If the name of the dog is Lucy, you play weak NT, otherwise strong NT. Or 1♣ is either a strong unspecified type bid or a transfer walsh 1♣, and a 1♥ reply is either controls or shows spades, etc. It must be very difficult to compete against this.

Used in competition, you can say that an overcall is either natural or the suit higher (encrypted), and this would certainly throw my bidding off if they did it against me.

What is the point of this clause (d)? This is the one law that deals with unauthorised information, and it is saying that the name of the dog is legal information.

fromageGB, on 2018-June-13, 03:52, said:

You might be able to get away with that, however, given that this this is now a matter of agreement, the other side can compel you to disclose the name of the dog which kind renders the whole plan moot

5.(a) When explaining the significance of
partner’s call or play in reply to an
opponent’s enquiry (see Law 20) a player
shall disclose all special information
conveyed to him through partnership
agreement or partnership experience but
he need not disclose inferences drawn from
his knowledge and experience of matters
generally known to bridge players.

fromageGB, on 2018-June-13, 03:52, said:

Yes, that information is authorized and the laws require you to make such information and how it affects your calls etc. known to your opponents as part your partnership understandings.

fromageGB, on 2018-June-13, 03:52, said:

I think the purpose of that clause is to prohibit having discussion about agreements after drawing your hand from the board. It's not saying that non-bridge information is legal because you knew it beforehand, it's just specifying the distinction between before taking your cards from the board and after.

E.g. you take your cards out, notice you have a 2♣ opener, and realize that you never discussed what response methods you're using. It's too late now.

fromageGB, on 2018-June-13, 03:52, said:

Does the name of this dog change randomly over time? No? Then I don't see the point. There's no "encryption" here.

blackshoe, on 2018-June-13, 10:26, said:

But there could still be a concealed partnership understanding.

pran, on 2018-June-13, 12:20, said:

How, in a bridge sense, does "we play weak NT here" differ from "we play weak NT here because partner's dog's name is Lucy"?

blackshoe, on 2018-June-13, 13:42, said:

I don't know and I don't care.
But if the dog's name (or whatever extraneous information) has any impact at all on the partnership understanding the effect must be fully disclosed (in advance).

My point, since you seem determined not to get it, is that the answer to my question is that there's no difference at all.

pran, on 2018-June-13, 14:34, said:

I think the question is whether the effect must be disclosed, or does the name of the dog has to be disclosed?

Playing Devil's Advocate: Clause (d) says that you can base your actions on information you had before taking cards from the board. Where does it say you have to give that information to the opponents? You're only required to disclose agreements. The name of the dog isn't an agreement, it's just a fact about the world. The agreement is only "If partner's dog is named Lucy, we play weak NT."

barmar, on 2018-June-14, 09:18, said:

Posting once again for Barry, who doesn't bother to read threads

5.(a) When explaining the significance of
partner’s call or play in reply to an
opponent’s enquiry (see Law 20) a player
shall disclose all special information
conveyed to him through partnership
agreement or partnership experience but
he need not disclose inferences drawn from
his knowledge and experience of matters
generally known to bridge players.