[1175]

I feel as if I need to start a thread of bidding explanations that appear "suboptimal" (meaning that they make no sense, and I hope they eventually change with a system update).

While some of these have appeared in other threads I have started, I will make this hand my first formal nomination:

---

This auction occurred at three tables, with 5♣ going down one or two. I will add a fourth table here, where (in this sequence) South simply passed the 3♥ bid by North (down five undoubled). What caused this? The treatment of the 3♣ bid: "My long suit - no major fit - 6+♣; 1-♠; 11-16 total points". If South had six or more clubs and 11-16 points, he would have acted over 1♥.

I will add this (similar) treatment of the 4♣ bid ("4-♦; 2-♠; 13- HCP; twice rebiddable ♣; 15+ total points"), which occurred at two tables (down three or four tricks):

---

Some Souths handled this sequence better than others. The winning call (made at only three tables):

---

That failed by one or two tricks, even though (double dummy) North-South can make 3♠. One South should have made his preference known a round earlier:

---

At four tables, South never retreated to spades (down three or four):

---

Finally, at two tables (mine included - a rare win in the duplicate IMPs), North-South produced a plus score with an "eccentric" bid by South:

---

One doesn't see both sides playing the same contract at the 3-level very often. This failed by two tricks (+100 and 9 IMPS). While I don't feel particularly proud of the 3♣ bid, it did have the effect of keeping North out of the auction.

[rhm]

You confuse bad treatments with bad judgement.

There is no cure against bad judgement except learning from experience.
Bad treatments require bad agreements, like playing something as artificial instead of natural or of playing something as forcing instead of non forcing or vice versa.
I doubt that the players in your example had clear agreements what the bids in your example showed.
An expert would have passed 3C in a flash not because of superior agreements but because of superior judgement

[1175]

 rhm, on 2024-June-23, 06:06, said:

You confuse bad treatments with bad judgement.

I don't agree.

This subforum concerns "BBO Robot Discussion." The Robot has no judgement, and (if you have read any of my other threads) you will see others making that point. Human players have to use the system, because the Robot will believe whatever its partner bid.

 rhm, on 2024-June-23, 06:06, said:

I doubt that the players in your example had clear agreements what the bids in your example showed.

The bid explanation showed exactly what the Robot expected from the South hand.

 rhm, on 2024-June-23, 06:06, said:

An expert would have passed 3C in a flash not because of superior agreements but because of superior judgement

I don't believe anyone (least of all myself) would classify the Robot as an "expert."

[smerriman]

 1175, on 2024-June-22, 12:59, said:

(meaning that they make no sense, and I hope they eventually change with a system update).

The description of this forum is "Abandon all hope, all ye who enter here".

Oh, maybe I misremembered..

Honestly, this one doesn't seem like a big issue to me; GIB has been told that if you want to ignore partner's strong request to pick one of their two suits and play in your own, you need a really good suit.

Sure, that can't exist by a passed hand, and no doubt it could have an improved definition - though be very careful, as weakening the definition most would likely cause GIB to start doing it opposite your Michaels with hands you'd rather it didn't..

But it's an exceedingly rare situation, compared to so many more common ones, and getting out of the auction as cheaply as possible with 2♠ doesn't seem that bad (can't fathom why anyone would pass, knowing you're guaranteed to be in a worse situation next bid..)

[1175]

I lost the hand here (not quick enough to save it), but I remember seeing the explanation and thinking "what the...?" I don't remember the vulnerability either (if that matters).

---

The explanation for the 4♥ bid said something like "4♠ and 6♥" (which I have never seen used anywhere else). I remember reading that several times. Please tell me that I haven't lost my mind.

[smerriman]

Was it actually GIB who made the 4♥ bid? As far as I'm aware, that bid does not exist for GIB, so it would certainly have been interesting seeing what it held.

If you, then it'll certainly say that description if you try to make it yourself; all undefined bids have a description attached, made up automatically either from the level of the bid, or where the rule that happens to match was meant to apply in other scenarios. That's not saying in any way that that was how the bid was intended to be used, just a side effect of impossible bids.

[1175]

smerriman, on 2024-June-28, 03:50, said:

Was it actually GIB who made the 4♥ bid? As far as I'm aware, that bid does not exist for GIB, so it would certainly have been interesting seeing what it held.

If you, then it'll certainly say that description if you try to make it yourself; all undefined bids have a description attached, made up automatically either from the level of the bid, or where the rule that happens to match was meant to apply in other scenarios. That's not saying in any way that that was how the bid was intended to be used, just a side effect of impossible bids.

Obviously (at least to me ), the explanation came up when South (meaning the human player - me) hovered over the explanation for 4♥. I held a hand 5-5 in the majors just below an opening bid (many might have even opened the hand). I don't even remember if I actually made the 4♥ bid (I might have seen the explanation and said, "oops, don't have that, have to find a different bid"). I wanted to drive to game in whatever 5-3 fit we had, and (again, if I remember) most alternative bids (say 3♥) allowed the Robot to pass.

[smerriman]

The systematic way to show that would be to start with 2♦; 4♥ wouldn't exist for me in a human partnership either. Sometimes it would be better for a bid to be described as impossible, though GIB also has to sub in for humans who have bid illogically on prior bids, so it has to have rules for any situation, no matter how weird they are.

What actually happened here is that GIB has been given a comprehensive set of rules for all hands it can hold after the 1nt rebid; 4♥ isn't even on the list. But there's an unrelated section which talks about generically bidding a new major in an auction at the 4 level showing a 6 card suit, which is where the description comes from.

[1175]

I have decided to describe this hand in this thread, because I blame a strange explanation for what happened.

---

At a couple of tables, South doubled the 1♥ bid, and ended up far too high in clubs (the remainder of those auctions don't merit repeating), doubled and down a gazillion.

At one table, South led a heart against 3NT (making 4):

---

At one table North played 3♠ (down 3):

---

At four tables (including mine), South overcalled 2♥ directly (down 2 in most cases, and down 3 at one table, though I should have held it to down 1). This scored poorly, which slightly surprised me:

---

Because East-West found 5♦ at only two tables (the auctions differed slightly, but began with the same 11 calls):

---

The remaining six tables played 4♠ on the auction shown in the hand diagram, failing by a trick in every case. I want to highlight the 3♦ bid by East, which apparently, doesn't show diamonds ("1+ ♦; 4+ ♥; 13+ total points", as I would have expected. Not really sure why West bid 3♠ (the 2♠ bid already showed the 6-5 distribution), but East (despite the treatment not showing diamonds) deserves the blame for this debacle by placing the contract in a 5-1 fit instead of a known 6-4 fit.

[smerriman]

It could be worse. In the older version of GIB I usually test first with, you wouldn't like the definition of 2♠.. it *deletes* one spade from the 4 it showed last round, and describes it as showing exactly 3 spades instead

But if I edit the definition to fix that to match the current GIB without doing anything else, then it replicates the rest of the auction. So I guess BBO fixed that one definition, without looking at the definitions of any followups. GIB sees 2NT as a club stopper, 3♣ as 4+ clubs, 3♥ as 6+ hearts, and 3♠ as 3+ spades, which together imply 3♦ can't be less than 3 diamonds. But GIB doesn't form any definitions by negative implications, only what the bid is specifically programmed to show, and the only matching rules are default rules in situations where the bid is a last resort to show a 7 card fit, thus the 1+ diamond.

Things just get worse after that, told that if nobody has been able to show an 8 card fit, resort to 4 of partner's major.

[1175]

This hand just came up:

---

I wanted to make a balancing double in this position, but the system literally has the point requirements for a balancing double ("3+ ♣; 3+ ♦; 3+ ♥; 2- ♠; 14-16 total points") higher than those for a direct takeout double (which makes no sense to me). I suspect that doubling at this point would get something like 3NT out of the Robot (not what I want to see). Does some way exist to go back at played hands and see what actions the Robot would have taken on alternative auctions? Letting East-West play 2♠ does not represent the best outcome for North-South.