[blml] 40B3, etc.

[Eric Landau]

On Mar 25, 2008, at 3:12 PM, David Burn wrote:

> [EL]
>
> On Mar 25, 2008, at 7:41 AM, David Burn wrote:
>
>> [RH]
>
>> However, Law 23 does not bar better bids that you discover after
>> the fact, known as "rub of the green".
>
>> For example, if you make an insufficient bid which cannot be  
>> untangled, so
>> partner must call an enforced pass, and you choose a rational  
>> guess of
>> punting 3NT, you keep your top if the field has had a scientific
>> auction to 6NT failing by one trick due to a 5-0 break.
>
>> [RH]
>
>> "...Damage exists when, because of an infraction, an innocent side
>> obtains a table result less favourable than would have been the
> expectation
>> had the infraction not occurred..."
>
>> [DALB]
>
>> What is truth, and what is fable? Where is Ruth, and where is Mabel?
>
> It is all truth; the fable is where we fut our cards.  Ruth is in the
> expectation, where 6NT is superior to 3NT; Mabel is in the  
> actuality, where
> 3NT is superior to 6NT.
>
> The law-writers are to be commended for making it unambiguously  
> clear that
> it is Ruth, not Mabel, that counts here, by including the reference to
> "expectation" (cited by Richard above) in the new L12.
>
> [DALB]
>
> It is not clear to me whether you would then adjust the score to  
> 6NT down
> one, which was the expectation had the infraction not occurred, or  
> leave it
> at 3NT making five, the actual table result. Richard seems to me to  
> consider
> on the one hand that the 3NT bidders keep their top, and on the  
> other that
> the innocent side has suffered damage and is entitled to redress.
>
> One would expect the Australian answer to "Where is Ruth and where is
> Mabel?" to be "Drinking - strewth! Under the table."

I let them keep their top.  David has convinced me that the citation  
from L12 (by RH, above) is ambiguous if read literally, as one cannot  
quantitatively compare an actual "table result" to a probabilistic  
"expectation".  Thus I have been (and will continue) reading "the  
table result" as "the expected value of the table result at the time  
of the infraction".  Since it was presumably not known at the time of  
the infraction that the key suit would break 5-0, the expected value  
of reaching 3NT (consequent to the infraction) is (assuming the one  
key suit is the only issue, with the OS is NV, and using binomial  
approximation) roughly 487.5 (15/16 of 490 + 1/16 of 450), whereas  
the expected value of reaching 6NT (stipuated as likely absent the  
infraction) is roughly 925.  The "rub of the green" principle says  
that we do not adjust if we determine that the infraction reduced the  
OS's expected result, regardless of what it did to their actual result.

IOW, I rule that the NOS was not damaged by the OS's infraction,  
since the infraction reduced the OS's expectation below that "had the  
infraction not occurred".  They were admittedly unlucky to be damaged  
by a 5-0 break in the key suit, but that was not due to anything the  
OS did.


Eric Landau
1107 Dale Drive
Silver Spring MD 20910
ehaa at starpower.net