[BLML] The Monty Hall trap

Alain Gottcheiner agot at ulb.ac.be
Fri Dec 10 17:06:26 CET 2010

Le 10/12/2010 16:30, Eric Landau a écrit :
> On Dec 7, 2010, at 6:47 PM, richard.hills at immi.gov.au wrote:
>> Richard Hills:
>>>> In Australia the Alert requirements are:
>>>> (a) 1NT overcall in the direct seat with no stopper
>>>>     = Alert!
>> Ron Johnson:
>>> All of this surprises me. Jeff Rubens has spent over
>>> 4 decades demonstrating that he can construct
>>> situations where a stopperless NT overcall will be
>>> the plurality choice (in partnerships that have not
>>> agreed that it's OK to overcall 1NT without a
>>> stopper).
>>> Many comments are on the order of "Showing my
>>> stopper" or "you got me", "but the alternatives are
>>> worse" or "after they run the hearts the play should
>>> be easy".
>> Richard Hills:
>> I agree with Jeff Rubens' analysis.  In my opinion it
>> is technically correct to immediately show pard that
>> one holds a strong balanced hand, thus enabling pard
>> to seize captaincy of the auction (maybe reaching a
>> laydown 4H on working values and a good fit, which
>> contract could not be achieved after a so-called
>> "textbook" Trap Pass due to lack of a stopper).
>> Thus Ali-Hills have the pre-existing explicit mutual
>> partnership understanding that any and all natural
>> bids of NT at the one-level neither promise nor deny
>> stopper(s).
>> However.....
>> Merely because a partnership agreement about a non-
>> conventional call is technically correct does _not_
>> necessarily mean that that natural call is not to be
>> Alerted.  If the vast majority of opponents expect a
>> technically incorrect partnership understanding, in
>> Australia the ABF has correctly deemed that the vast
>> majority should be protected with an Alert.
> An agreement that a 1NT overcall shows x to y HCP in a balanced hand
> (not suitable for a systemic takeout double) does not define one's
> agreement as to what action one will take with such a hand.  There
> are essentially three possibilities:
> 1. Bid 1NT with a stopper in the opponents' bid suit, otherwise
> pass.  Partner raises NT freely.
> 2. Bid 1NT with a stopper, otherwise seek a systemically flawed
> alternative call that approximately describes your hand (e.g. an
> overcall on a good four-card suit, a somewhat off-shape double, or a
> 1NT overcall with a "suitable non-stopper" such as Jxx or xxxx); if
> none is to be found, pass.  Partner raises NT freely.
> 3. Bid 1NT.  Partner does not raise NT without checking back for a
> stopper unless he has one himself.
> #1 is taught to beginners.  #2 is standard practice among competitive
> players.  #3 is an unusual agreement that should obviously require an
> alert.
> But requiring an alert to disclose that partner's 1NT overcall "does
> not promise a stopper in your suit" covers both #2 and #3, which,
> IMO, confuses more than it clarifies and is thus a bad idea.  One
> might argue that it is "common bridge knowledge" that there is a
> difference between "shows a stopper" (partner presumes a stopper, and
> raises freely) and "100% absolutely guarantees a stopper".  One might
> also argue that the only difference between these agreements is that
> partnerships with the latter call their stopper-less 1NT overcalls
> "psychs" rather than "minor tactical deviations".
AG : if I understand you, you mean that the possibility of "minor 
tactical deviations" is inherent to bridge (else there wouldn't be 
bidding contest columns) and therefore shan't be alerted, because you 
would have to alert every bid.
In that case, a local rule that you have to alert bids which can be 
deviated from would be totally impractical.

I'm fortunate enough to live in a land where most TDs understand this;

Best regards


More information about the Blml mailing list