[whereagles]

Hi all, see if you get this one right.

Venue: open team trials, training session. Thus top opponents, obviously. IMP scoring.

---

Auction:
1♣ = either (1) natural 4+ cards unbalanced or (2) preparatory with balanced 12-14 H (can be 2+ clubs).
1♦ = 4+ hearts.
1♥ = 3 hearts, normally min hand.
2♣ = GF relay.
2♠ = nat.
3♦ = stopper ask.

Yer lead?

[the hog]

small D

[eagles123]

spade

[Jinksy]

Small ♦ for me. A passed P's unlikely to have good enough ♣s to set the contract, and with LHO sounding balanced, P shouldn't have more than 3Ss, just in case I was feeling fanciful. If the D lead carves a trick, we probably weren't ever setting it.

[Lord Molyb]

small diamond for me too

[1eyedjack]

Hello

I chose that:

Declarer had 13-18 HCP. Outside that range he would either have made some further slam move or not game forced.
Declarer had fewer than 5 Spades and precisely 4 Hearts, or would have played in known major fit. Not guaranteed but close enough.
Declarer had fewer than 2 of top 3 honours in D (to justify asking)
Dummy had at least 1 of top 3 honours in D (to justify accepting)
Dummy had 12-14 bal with 4 Spades and 3 Hearts.
Partner completely random in context.

Dealt 10000 matching deals (took 15812246 deals to achieve it). D/D result of alternative leads are:

Lead ........ % set ......... Expected tricks

D8 ...........8.18% ......... 2.70
C9 ......... 6.16% ............ 2.59
DK ........ 6.02% ............ 2.32
S4 ........ 5.66% ............ 2.60
S6 ........ 5.64% ............ 2.60
ST ........ 4.51% ............ 2.51
H5 ........ 3.38% ............ 2.35
H2 ........ 3.37% ............ 2.35
H8 ........ 3.00% ............ 2.30
HQ ........ 1.15% ............ 1.93

[whereagles]

@1eyedjack: your constraints are a bit confusing. Opener is the declarer.. I think you got it the other way around.

[1eyedjack]

whereagles, on 2014-September-26, 14:09, said:

@1eyedjack: your constraints are a bit confusing. Opener is the declarer.. I think you got it the other way around.

You are right. My bad.

I have swapped declarer and dummy in the previous example and re-run.

Dealt 10000 matching deals (took 15674724 deals to achieve it). D/D result of alternative leads are:

Lead ........ % set ......... Expected tricks

D8 ...........6.15% ......... 2.61
C9 ......... 5.34% ............ 2.57
S4 ........ 4.93% ............ 2.57
S6 ........ 4.86% ............ 2.56
DK ........ 4.50% ............ 2.20
H2 ........ 3.30% ............ 2.41
H5 ........ 3.23% ............ 2.40
ST ........ 3.00% ............ 2.33
H8 ........ 2.68% ............ 2.35
HQ ........ 1.60% ............ 2.18

[Lord Molyb]

@1eyedjack: in which scenarios would the ♠4 lead gain over the ♠6?

[1eyedjack]

Lord Molyb, on 2014-September-26, 23:03, said:

@1eyedjack: in which scenarios would the ♠4 lead gain over the ♠6?

Hmm, I could have exported the output into a pbn file, loaded it into DDS, filtered the 10K hands as much as possible and then hunted through the remainder, for what appears to be the single hand that makes a difference, based on the spread of %.

Not going to attempt that. Not sure that I can reproduce the data if I fail to pbn it first time. That is a more interesting point, actually, and will look into it. Need to be able to save the random number seed in the .ini file.

My guess is that declarer had 4 card Spade suit including the Spade 5. Leading the 6 somehow promoted the 5.

[whereagles]

So that seems to have settled it: diamond is the winner.

And it is so in practice, because opponents had ♦Jx in dummy opposite declarer's "stopper" which consisted of ♦Qx tight

If you were to "try" for something more fancy, like leading the club 9 on grounds that pard "could have doubled 3♦ with ♦AT9xx", you'd end up seeing declarer making an overtrick Two things warn against this: (1) pard is a passed hand, so you're probably not going to beat this on a passive lead, and (2) partner had a chance to double the 2♣ relay with a bunch of those.

Thx all.