smerriman

My main point was that a spade and an entry still isn't going to be enough to set it, unless declarer has exactly Qxx (where partner is more likely to have 2 anyway).

This is something I've always wondered. It's actually quite a different scenario, since if partner reverses into 2♥ over 1♠, he's still hopeful of finding a 4-4 heart fit, while reversing into a 4 card heart suit over 1NT is done with the knowledge that such a fit can't exist. It obviously doesn't promise any more than 4 hearts, but I'm not sure it should be done on exactly the same set of hands, or whether you should play the same lebensohl type followups.

But again, surely you also have to take into consideration the fact he didn't lead a spade. If I'm a conservative preempter holding AKJxxxx, I personally would definitely lead the king and look at what happens, rather than assume that partner can get in, lead his singleton spade, AND have declarer's queen drop in 3 (or get in a second time later). While if partner has 0 or 2 spades, a spade lead seems needed, and gives you the chance to see dummy too. Maybe others wouldn't for reasons I don't yet understand but it's still an important factor to consider.

Wouldn't you lead a top spade from that holding? Avoiding a spade seems to be playing partner for exactly a singleton and two tricks (or one trick and declarer to have precisely Qxx).

If you're playing XYNZ, then I see no reason at all to avoid the cheaper 2♥ response, since you have plenty of time to bid further. You mentioned yourself your only problem with 2♥ was the difficulties in keeping the bidding low to find slam, which is one of the main flaws of NMF in the first place.

Sorry, I deleted that like a minute after I posted it, you were too quick :P Still, it's not as if a 6-2 club break with that lead and play is going to affect the probability much.

Around 17-18, invitational.

If you look more carefully you'll actually notice that playing the queen does not cost a trick (double dummy at least).

I can see why this is more complex than it originally looked, but I'm not quite following the logic. You concluded that it's impossible for West to play the 9 from A9x, but that playing 9 from J9x gives you a losing option. What's the losing option? When we next lead low from hand, if West plays low, we know they couldn't possibly have played the 9 from A9x, so the only option is to duck, and it works. If West plays the J, we cover and set up our 8, so no losing option there either. It seems non-omniscient defenders *would* have to play the 9 from A9x at least some percentage of the time if they want to improve their chances with J9x too, gambling that declarer doesn't have K8x.

When you're hosting a table, you have additional options, such as removing a player. This has always been the case.

Low to the 10 works if West has singleton 9 or East has singleton J. Leading to the queen works if West has AJ. The former is more likely. However, Suitplay would also go with the percentages so there must be a catch..

My guess: South plays the queen, declarer figures this is from QJx and with the club ace likely onside, ends up finessing in trumps for a cold bottom? Would still play low myself.

What was 2♥? It doesn't appear to be 4sf given East's strange jump to 4♠..

With every single table getting a different result on a straightforward hand, I think you're going to have to expect pretty random results.

Correct. The biggest thing you'll discover is that while the function works fine, the deal program is designed to output deals, so you have to write your own function to accumulate the results and output averages / frequencies etc. But once you've done that, the part for checking conditions is really powerful and cool to play with.

Well, the OP said GIB opened with the OP's best suit, and they bid 4NT; if there were no intervening bids left out of the story, then GIB did. But yeah, there is definitely something wrong with the story, since if they were missing both AK of trumps then it's really not viable for slam to "depend on which honour" the bot had. So much more likely there would have been some misunderstanding on the human side here, rather than GIB getting the response wrong.

In case you missed it from another thread.. given the better alternative there's really no need:

Do you still have the hand record? I've seen GIB do a lot of weird stuff, but *never* get RKCB wrong.

Yep. As far as I'm aware, jumping in a new suit is undefined for GIB.

If a double suggests a spade lead, next time the deal will be exactly the same but with spades and diamonds reversed ;)

By the way, as I referred to earlier, whether you overcall or double isn't just about point count at all. If your RHO opens 1♣ and you have - KQJ98 KQ5 AKJ76, a 1♥ overcall is pretty clearcut (both with and without GIB), despite having 19 HCP (and a K&R value of over 22).

Playing with humans, if you're too strong to overcall, start with a double then override partner's choice to bid some number of your suit. What too strong to overcall means depends on the exact hand; if you're short in a major, expect partner will be bidding or even jumping in that suit, so make sure you can follow up properly. Playing with GIB, you have no hope of any sensible bidding sequence since it doesn't understand basic logic here. Probably just lie, show a weak hand by preempting to game and hope you didn't miss slam.

I guess the idea is that if you lead the 5 of spades, West is not as likely to duck. Playing queen to the king would be silly at MP but queen and low still works with the rest of the line. [edit] Not sure if that's true at all though, South could easily have Qxx and ducking seems right there too, if not more so.. [edit2] Oh right. Queen also works if *East* ducks the ace.

You really should learn to use Thomas Andrews' Deal. It's basically the same idea, but once you're used to it, you can do everything dealer can and much more. If the syntax is too complex, and you post a few examples of dealer scripts you use, I'm sure I or someone else can convert it to the TCL format, after which you just can copy and paste and adjust bits as you need.

Yikes, looks like I caught you on a bad day. Wasn't trying to have a go at all - initially read the first two mentions of minimum in the context you intended, but when I got to the point where you asked if 3♣ could be bid on a shapely minimum or whether it shows extras, figured you must have missed the extras bit and went back and reinterpreted the rest. To me (and the OP, based on earlier posts), 'extras' means at least 15 points, so referring to minimums and 14 counts didn't make sense to me. I apologise if I offended you.