An unusual problem (part 1)

[ais523]

I was experimenting with a highly artificial bidding system (not one I
usually play) in a practice session, when this unusual problem occurred:

Matchpoints, none vulnerable

S W N E
1C! - 1H! X
-! - ?

! = alerted bid
1C! = 10-12 balanced, or most 16+ hands
1H! = 4+ spades, any strength
X = not alerted, so presumably shows hearts
-! = 10-12 balanced, exactly 3 hearts

Sitting North, I held (SHDC):

A9852 Q73 AT96 3

With 10 HCP and decent shape, this hand looks like it's probably our
partscore: there's unlikely to be a game available, but we're also
likely to be somewhat stronger than the opponents (especially as West
was unwilling to bid anything).

There are three reasonable systemic calls here: 1NT, which asks North to
choose a partscore to play in (most likely 1NT or 2S, although with a
5332 hand they might choose 2 of a minor); 1S, to play; and 2H, which
forces partner to bid 2S (which could be passed to play spades from
partner's side). Choosing between those isn't the unusual problem
(although it's still an interesting judgement issue, I think I'd
probably prefer to play in a possible 5-2 spade fit).

The problem arises because there's also an unreasonable call,
which I was really tempted to make: although this is a live auction,
we just had a double and two passes, so it would be possible to pass to
play 1H. Even if it makes exactly, that's 160 points, which outscores
any likely partscore. Normally, intentionally playing a 3-3 fit is a bad
idea, but how hard can it be to make 7 tricks?

So, my first question is: is it best to go for a sensible 1S or 1NT, or
is the crazy idea of playing the 3-3 fit actually a reasonable choice
for once?

(My second question is "OK, so how do you even play a 3-3 fit", but I'll
delay that until I've had a few answers to the first question, so that I
can show what dummy would look like.)

[Douglas Newlands]

Post by ais523
I was experimenting with a highly artificial bidding system (not one I
Matchpoints, none vulnerable
S W N E
1C! - 1H! X
-! - ?
! = alerted bid
1C! = 10-12 balanced, or most 16+ hands
1H! = 4+ spades, any strength
X = not alerted, so presumably shows hearts
-! = 10-12 balanced, exactly 3 hearts
A9852 Q73 AT96 3
With 10 HCP and decent shape, this hand looks like it's probably our
partscore: there's unlikely to be a game available, but we're also
likely to be somewhat stronger than the opponents (especially as West
was unwilling to bid anything).
There are three reasonable systemic calls here: 1NT, which asks North to
choose a partscore to play in (most likely 1NT or 2S, although with a
5332 hand they might choose 2 of a minor); 1S, to play; and 2H, which
forces partner to bid 2S (which could be passed to play spades from
partner's side). Choosing between those isn't the unusual problem
(although it's still an interesting judgement issue, I think I'd
probably prefer to play in a possible 5-2 spade fit).
The problem arises because there's also an unreasonable call,
which I was really tempted to make: although this is a live auction,
we just had a double and two passes, so it would be possible to pass to
play 1H. Even if it makes exactly, that's 160 points, which outscores
any likely partscore. Normally, intentionally playing a 3-3 fit is a bad
idea, but how hard can it be to make 7 tricks?
So, my first question is: is it best to go for a sensible 1S or 1NT, or
is the crazy idea of playing the 3-3 fit actually a reasonable choice
for once?
(My second question is "OK, so how do you even play a 3-3 fit", but I'll
delay that until I've had a few answers to the first question, so that I
can show what dummy would look like.)

1S looks very normal and sensible since your first duty is to try to get
a plus score for your side.
It seems strange to have a system where pass shows 3 hearts.
Most transfer systems show something in responder's suit (spades) at
this point.

doug

[nrford100]

...(My second question is "OK, so how do you even play a 3-3 fit"...

It depends on the hands

Several months ago a bidding snafu left me in a 3-3 fit. We both had a quick trick or two and short suits. I cashed the winners and cross-ruffed with all the trumps for a very good score. I don't remember if it was an actual top. I was just happy to come out alive, much less with a top or near top.

[Lorne]

Post by ais523
I was experimenting with a highly artificial bidding system (not one I
Matchpoints, none vulnerable
S W N E
1C! - 1H! X
-! - ?
! = alerted bid
1C! = 10-12 balanced, or most 16+ hands
1H! = 4+ spades, any strength
X = not alerted, so presumably shows hearts
-! = 10-12 balanced, exactly 3 hearts
A9852 Q73 AT96 3
With 10 HCP and decent shape, this hand looks like it's probably our
partscore: there's unlikely to be a game available, but we're also
likely to be somewhat stronger than the opponents (especially as West
was unwilling to bid anything).
There are three reasonable systemic calls here: 1NT, which asks North to
choose a partscore to play in (most likely 1NT or 2S, although with a
5332 hand they might choose 2 of a minor); 1S, to play; and 2H, which
forces partner to bid 2S (which could be passed to play spades from
partner's side). Choosing between those isn't the unusual problem
(although it's still an interesting judgement issue, I think I'd
probably prefer to play in a possible 5-2 spade fit).
The problem arises because there's also an unreasonable call,
which I was really tempted to make: although this is a live auction,
we just had a double and two passes, so it would be possible to pass to
play 1H. Even if it makes exactly, that's 160 points, which outscores
any likely partscore. Normally, intentionally playing a 3-3 fit is a bad
idea, but how hard can it be to make 7 tricks?
So, my first question is: is it best to go for a sensible 1S or 1NT, or
is the crazy idea of playing the 3-3 fit actually a reasonable choice
for once?
(My second question is "OK, so how do you even play a 3-3 fit", but I'll
delay that until I've had a few answers to the first question, so that I
can show what dummy would look like.)

Playing a 3-3 fit requires at least the trump ace so you can increase
your odds of ruffing something before alert oppo take your trumps out.

A double dummy run with N fixed and S 10-12 balanced with 2-3 spades and
3 hearts suggests spades is a huge winner over Pass or 1N, even without
biasing the E hand which I assume has good hearts and should lead them.
(1S made 97% of the time and 2S 78%, 4S only 8%)