[kenberg]

I have recently agreed to play udca with a partner. I am a bit of a skeptic but I am not grumbling. Some situations we are discussing:


A. Defending a suit contract partner leads the ace, presumably backed by the king, three spots hit the dummy, you hold JTx. The J, or the T, will not be a success if declarer holds Q9x so you play small. This is right?

B. Same contract, same lead, dummy has QT5, you hold J62. Declarer will read partner for the king and establish his queen, but there is no point in helping him. If you play the J declarer has no need to come to his hand to lead towards the board, he will just play the Q or T. So you play the 6, not the Jack, right?

C. Same as B, but this time you hold J6. Given A/B, it seems that you should play the J. That is, with Jx you revert to standard carding. I am not so sure this applies to Tx. Probably not.


Partner suggested that third hand treat the J as being "lower" than all spots in choosing the play. This works in B and C (the 6 now being "high" in B and the J being "low" in C) but it doesn't apply in A.

My view is that A and B are covered by the dictum that you don't signal with a card that you cannot afford to part with. That still leaves C however. Playing J from Jx seems like a logical consequence of A and B, but it also seems to require explicit agreement. I can well imagine holding J6 and thinking "Sure, the 6, it's udca". But I think the J is the way to get the ruff.


Your thoughts? I assume udca regulars have thought this through long ago.

[Codo]

A: I would have to look at the complete hands to decide whether I need to make a correct signal or whether I cannot aford that card.
So, if I guess that partner is continuing the suit to give me a ruff if I show an even number, I better play the honst card.

B: With J62 I signal an odd length or my dislike of this suit with the six.

C: Easy one, I show that I like the suit or my even legth by contributing the 6.

Partner cannot always read the six in case B and C? Right. This happens if you signal with the six and is true for udca and for standard.

The idea that a a jack always shows a low card is interessting but not my cup of tea. I accept the frequent problems I have with any given signalling method.

[hanp]

No signalling method will work best on all hands. I know people who signal standard on the lead of an ace or king at trick 1 only.

With hand A I'd play the jack if I have a side entry. This indeed loses if declarer has Q9x, can't win them all. Usually I try not to give away tricks by signalling but here it seems too likely that partner is going to continue if you play small. A lot of people solve this problem by playing small very slowly.

Quote

B. Same contract, same lead, dummy has QT5, you hold J62. Declarer will read partner for the king and establish his queen, but there is no point in helping him. If you play the J declarer has no need to come to his hand to lead towards the board, he will just play the Q or T. So you play the 6, not the Jack, right?

Yeah, play the 6 because you don't want a ruff. The jack is definitely also possible, it is a clearer signal. It can cost on some layouts though, for example if declarer has a singleton and play the suit to pitch two losers. Unlikely but possible.

Quote

C. Same as B, but this time you hold J6. Given A/B, it seems that you should play the J. That is, with Jx you revert to standard carding. I am not so sure this applies to Tx. Probably not.

Play low of course, you want a ruff. You can't say "with a jack you revert to standard signalling" because partner won't usually know we have the jack.

[billw55]

I guess I am too weak to understand the problem. I would just play as:

A. I have the JT but not the Q. J.
B. I don't have the Q or a doubleton so I discourage. 6.
C. I do have the Q or a doubleton so I encourage. 6.

No signal method is perfect on every layout. You can always construct hands (and find them in real play too) where a particular method works the best, and some other method fails.

[kenberg]

I am not actually trying to build a case against udca. I accept that every system fails from time to time. Rather I am, or rather we are, trying to decide the best course in certain reasonably common situations.

For example, a holding of JTx is hardly uncommon. Partner leading the ace is not uncommon. When this happens I think we place partner with the king and declarer with the queen, because if partner does not have the king the hand is already lost and if partner has AKQ he will cash them regardless of my signal. So give partner the king and declarer the queen. I guess half the time declarer has the 9 to go with it.


I guess I can think as follows: If I play the x, partner will continue and we build a trick for declarer 100% of the time. If I play the jack, we only build a trick for declarer on the 50% of the time (conditioned on Q with declarer, king with partner) when declarer has the 9.


I have played udca off and on over the years, and on most hands it doesn't much matter as long as partner is playing the same thing. But if I am going to play it more regularly I want to think about some of the things that surely will arise.

[hanp]

Yes, that's what we said. So what's your question?

[kenberg]

None, I guess.

[peachy]

kenberg, on Sep 15 2010, 07:53 AM, said:

C. Same as B, but this time you hold J6. Given A/B, it seems that you should play the J. That is, with Jx you revert to standard carding. I am not so sure this applies to Tx. Probably not.

With J10x, play the J that also shows the ten. If you encourage with the spot card, partner will think you have the Q.

With Jx, play the spot. If you play the J, partner will think you have J10x or a singleton.

With J62, play the 6. If you play the two, partner will think you have a doubleton or you have the queen or king

Finally, do not revert to standard signals when you have agreed on UDCA. What is the point of that?

[kenberg]

Perhaps I should not have phrased it as reverting. I just talked to partner and here is what we have agreed to as our way of handling these matters..

Partner leads A.

Three spots hit the table. From JTx I throw the ten, suggesting a shift, hoping partner has the 9. It's true that partner is unaware of my jack. He is unlikely to care.


QTx hits the table, I have J62, I throw the 6 trusting partner can read it. On some occasions, the extra work for declarer, coming to hand to lead toward the table establishing the queen, will help us. No, I don't throw the 6 slowly.

QTx hits the table, I have J6, I throw the J. Since I would play the 6, not the jack, from J62, this should (?) be readable.

Perhaps we will regret this.

[JLOGIC]

Meckwell play standard on the lead of the ace at trick 1 for this reason (and a lot of other pairs now also).

The problem can be the same on the king lead with dummy having 3 small and you having T9x (if declarer has AJ8...).

I would strongly recommend with JTx you play the jack. Yes you blow a trick to Q9x, but if you play low partner will never figure it out anyways so just chalk it up.

[JLOGIC]

BTW playing the J from Jx is something I have discussed with my dad when there is QTx on the board. I think it's good. It seems clearly suboptimal for the J to only be stiff J.

[kenberg]

Thanks all.

I like the idea of playing std on T1 A/K leads. I would have to convince partner. Assuming we stick with udca, I am now convinced it must be J, or T, from JTx. I have a mild preference for the ten. I realize playing an honor third hand denies the honor directly above it, but this is a situation where I wish to disguise, not clarify, my honor card holding. Maybe randomly J or T. [Oops, scratch that, I do get dealt doubleton JTs]. Doesn't really matter. If dec has Q9x he will be leading to the 9. Win some, lose some.

Thanks again.

[zenko]

kenberg, on Sep 15 2010, 03:02 PM, said:

I am not actually trying to build a case against udca. I accept that every system fails from time to time. Rather I am, or rather we are, trying to decide the best course in certain reasonably common situations.

For example, a holding of JTx is hardly uncommon. Partner leading the ace is not uncommon. When this happens I think we place partner with the king and declarer with the queen, because if partner does not have the king the hand is already lost and if partner has AKQ he will cash them regardless of my signal. So give partner the king and declarer the queen. I guess half the time declarer has the 9 to go with it.


I guess I can think as follows: If I play the x, partner will continue and we build a trick for declarer 100% of the time. If I play the jack, we only build a trick for declarer on the 50% of the time (conditioned on Q with declarer, king with partner) when declarer has the 9.


I have played udca off and on over the years, and on most hands it doesn't much matter as long as partner is playing the same thing. But if I am going to play it more regularly I want to think about some of the things that surely will arise.

Regardless what are you intention you are barking at the wrong tree IMO, what you should really focus on is how dependent is your signaling method on declarer playing a honest card. With limited number of small cards, playing UDCA you are more exposed to declarer messing up your positive/even length signal (by hiding the smallest card), same way by playing standard, he can spoil your negative/odd signal which is kind of a wash.

When it comest to playing "small" small card its a wash, but the key advantage of UDCA is "high" small card play situation, with worthless holding it is usually safe to give the clearest negative signal possible by playing the highest small card you have, while playing standard with valuable holding you often have to compromise an not play the highest small card you have, to avoid weakening your holding too much, which sends muddled signal to your pard.