[Free]

I've been playing transfer preempts for several years now, with GF (!!!) 55+ hands included. The results after 55+ hands are quite good, BUT the results after preempts don't seem to be much different from normal preempts. Ok, theoreticly, you have a huge disadvantage, but in practice it just works out fine, and the completely unknown hand plays. So the huge loss of preempts isn't thàt awfull.

What strikes me the most is that, whenever there's doubt about the method, Ben comes along with examples of the strong 55+ methods, where he usually has a GF hand. Where are the examples of preemptive hands, or examples where you just have an invitational hand with total misfit? It seems like you never get hung for a number, but we all know it's just an illusion... Give us some examples of hands where you open a strong two-suiter and responder has a great sacrifice bid just in case you have a preempt. Give us the same responder, but with a preempt opener. Give some examples of hands where responder is sooo weak that after this bidding you even end up too high. Give us ... (you get my point)

I think this thread has become a discussion about the method (great), but some people (both sides) are just a little too blind to see the situation from another angle (it's just theoretic vs practice).
Strong 55+ hands are difficult to bid, so there you get advantages, but weak preempt hands lose in the long run because you give opps a penalty Dbl and a lead directing Dbl (or something else great), even if the unknown hand plays! That is in imps, in mp's you might get away just fine. But there's one thing I've always HATED (speaking from experience): if you're in 3rd seat with a 3-level preempt, you have to transfer, and you can't make a NF bid!!! This is imo the biggest disadvantage of ALL in this method!

[inquiry]

Let me deal with a number of issues.

First to FREE, I am trying to discuss ways to deal with probing slam and grand slam after the strong opening. To be honest, misho wants the "MisIry cue bid" to simply deny the ACE. I prefer it denies both. I tired to get around the critique that I am hand picking hands by using ALL hands in vugraph, that fit the requirements. (AKA no "picking"). I will continue to do this. How to deal with the weak hands, is pretty much like any other preempt. The problem with that, however, is you have to guess what the defense will do.

Second, to Richard. I wonder if your math is off just a tad. If the chance that one hand will fit the requirement of 5+/5+ and the number of points is roughly 0.8, what affect does it have that there is not one hand, but rahter 4 hands on each deal. Does this increase this from 0.8 to 4 x 0.8 to a frequency somewhare around 2.4%?

Third, to Richarrd, on the vugraph hands I spefically picked hands that would have opened (not all hands, if someone else bids first, the hand isn't included). To follow this up, So let me continue with examples from the 1970 world championship book. Of course they "hand picked" the hands show so you can only do stats on the finals, which had 96 hands. There were two such 5+/5+ hands in the finals, out of 96 (2.1%).. Here are these hands.

Final II, deal 11, page 173

3D - 3H
3S - 6S

Easy, DQ, two aces, partner has 3 losers. Both finalist got to slam.

---

Final II, deal 11, page 173

3D - 3H
3S - 6S

Easy, DQ, two aces, partner has 3 losers. Both finalist got to slam.

---

Ok, no plus for system in the finals, in the rest of the same book, there was ____ more hands. The next one is a huge win for the MisiIry meithod...

Final II, deal 11, page 173

3D - 3H
3S - 6S

Easy, DQ, two aces, partner has 3 losers. Both finalist got to slam.

---

On this hand Italy got to 2NT making 12 tricks. Wolff and Jacoby got to 6NT going down. 6CLUBS is best contract, and can't go down.

Final II, deal 11, page 173

3D - 3H
3S - 6S

Easy, DQ, two aces, partner has 3 losers. Both finalist got to slam.

---

PS - PS - 3C - PS
3D - PS - 3H - 4D
5H - PS - 6H

Ok I wouldn't open 3C with this hand, as it has only 2 controls. But some may want to loosen the requirements, who knows. 5H ask for six if missing heart queen. On this hand EW can make 6D, NS can make 6H. There maybe ways WEST with diamonds can get into the auction earlier, but it is not clear.

---

There are some other hands in the book that "fit the strenth" requirement, but jsut like for vugraph, I only picked the hands where the strong hand would get a chance to open.

So here is the next challenge I will do to address this issue. The ACBL runs 192 hands on BBO a day. Since Firday and Saturday hands have not been dealt yet, we will see how many of the hands fit these requirements (again as opener) and see how many such hands are found in the 384 tournement hands. Does that sound fair? Should we do next six days to get 1000 hands?

Ben

[inquiry]

Added note,

I ran bridgebroswer on an old database I ahve on CD (faster than online version).

Instead of looking at NORTH as richard suggested, I looked first at dealer, who had the 5/5 hand and 15+ hcp 0.77% of the hands. Then I ran the same data set where I only looked at hands were dealer passed, and asked if SECOND hand had the 5/5 hand (or better) and 15+ hcp. The chances of this was 0.63%. So just adding these two conditions together, the frequency of hands were either first or second hand will open 3C/3D/3H with strong pattern is already 1.4%. I think this shows the flaw in richard's math.. he applied it to a hand, not a deal, which has four hands. It looks like 3rd and 4th seat will have similar numbers (maybe getting slightly less as people have chance to open in front of hte strong hand). But the total will be somewhere around the 2% observed in the other test I did. (I still think it will be less than 2% because of the loser requirement).

Ben

[david_c]

A general question about transfer pre-empts - how should playing transfers affect your style of pre-empting? Should your pre-empts be more constructive, or more random than normal? I can see an argument either way:

Because you're giving them a penalty double, maybe your pre-empts need to be sounder than normal.

On the other hand, transfer pre-empts make it harder for partner to bounce, so maybe it makes sense to play a style where partner will not want to bounce very often (ie. a very random style of pre-empts).

[david_c]

Another note about transfer pre-empts in general:

Here in England, as of last week we are allowed to play a 2♦ transfer pre-empt in most competitions. I wouldn't be surprised if this becomes quite popular, because most other multi-meaning pre-empts are heavily restricted. So perhaps we will quickly become expert at defending against it!

You may be wondering, why aren't we allowed any other transfer pre-empts? It appears that the answer is partly that no-one asked to be allowed to play them. So if there are any English players who want to play this sort of thing, then go ahead and apply to the EBU - it ought to stand a decent chance of getting through, now that 2♦ has been accepted.

[hrothgar]

inquiry, on Apr 8 2005, 03:59 AM, said:

Second, to Richard. I wonder if your math is off just a tad. If the chance that one hand will fit the requirement of 5+/5+ and the number of points is roughly 0.8, what affect does it have that there is not one hand, but rahter 4 hands on each deal. Does this increase this from 0.8 to 4 x 0.8 to a frequency somewhare around 2.4%?

Traditionally, when I have seen people discuss frequency of an opening bid, they analyze 13 cards in isolation from anything else and repeat ad-infinitum. The reason that this formulation is used is to avoid all sorts of cross-dependencies. If we analyze an entire deal the probability that hand A gets a strong 5-5 hands is conditional on the holding in hands B, C, and D.

Ben is calculating a very different statistic and describing this as the frequency. Ben is looking across a large number of deals and making the following assumption. Assume that everyone at the table is playing their standard methods, with the exception that he "misiry" preempt substitutes for the standard three level opening structure ... Ben is then using BridgeBrowser to calculate the percentage of DEALs in which a Misery preempt would have occured.

This is all fine and dandy. It even produces an interesting statistic. However, you are misapplying standard vocabularly. The word frequency is used within a specific context. If you tell people that you are estimating the frequency of an opening bid, they are going to make the same assumption that I did, which is that you are talking about individual hands rather than deals.

From my perspective, the best way to present your results is to break out the information and present it as follows

1. The probability that you get deal a Misiry 3 level preempt in first seat...
2. The conditional probability that are dealt a Misiry 3 level preempt given a first seat pass
3. The conditional probability that you are dealt a Misiry3 level preempts given a 1st/2nd seat pass
4. I'm assuming that you ditch the weak adjunct in 4th seat...

Going back to my math...

>The combination of BOTH 5+/5+ shape and 15+ HCP is extremely rare.
>Roughly .7% of all hands would qualify for the strong hand type.

I'll simply note that when you look at one hand in isolation, the frequency was .77% which looks pretty damn close. (I suggested that you look at NORTH, you used DEALER - the specific hand that you CHOSE is completely immaterial...)

[inquiry]

david_c, on Apr 8 2005, 09:26 AM, said:

A general question about transfer pre-empts - how should playing transfers affect your style of pre-empting? Should your pre-empts be more constructive, or more random than normal? I can see an argument either way:

Because you're giving them a penalty double, maybe your pre-empts need to be sounder than normal.

On the other hand, transfer pre-empts make it harder for partner to bounce, so maybe it makes sense to play a style where partner will not want to bounce very often (ie. a very random style of pre-empts).

The biggest downside to the method I suggest is the inability to preempt to 3♣. My partner wants to use 2NT for this, but I am stubborn. We do play multi 2D, and use 2H/2S for lucas. I guess we could play 2H as major two suiter, weak, and 2S as tranfer to 3C, or even 2H as flannery. I guess we will wait until we decide to keep the transfer preempts before messing with the meaning of our two level bids.

Ben

[inquiry]

hrothgar, on Apr 8 2005, 10:02 AM, said:

Traditionally, when I have seen people discuss frequency of an opening bid, they analyze 13 cards in isolation from anything else and repeat ad-infinitum. The reason that this formulation is used is to avoid all sorts of cross-dependencies. If we analyze an entire deal the probability that hand A gets a strong 5-5 hands is conditional on the holding in hands B, C, and D.

Ben is calculating a very different statistic and describing this as the frequency. Ben is looking across a large number of deals and making the following assumption. Assume that everyone at the table is playing their standard methods, with the exception that he "misiry" preempt substitutes for the standard three level opening structure ... Ben is then using BridgeBrowser to calculate the percentage of DEALs in which a Misery preempt would have occured.

This is all fine and dandy. It even produces an interesting statistic. However, you are misapplying standard vocabularly. The word frequency is used within a specific context. If you tell people that you are estimating the frequency of an opening bid, they are going to make the same assumption that I did, which is that you are talking about individual hands rather than deals.

From my perspective, the best way to present your results is to break out the information and present it as follows

1. The probability that you get deal a Misiry 3 level preempt in first seat...
2. The conditional probability that are dealt a Misiry 3 level preempt given a first seat pass
3. The conditional probability that you are dealt a Misiry3 level preempts given a 1st/2nd seat pass
4. I'm assuming that you ditch the weak adjunct in 4th seat...

Going back to my math...

>The combination of BOTH 5+/5+ shape and 15+ HCP is extremely rare.
>Roughly .7% of all hands would qualify for the strong hand type.

I'll simply note that when you look at one hand in isolation, the frequency was .77% which looks pretty damn close. (I suggested that you look at NORTH, you used DEALER - the specific hand that you CHOSE is completely immaterial...)

Well, when I responded to your earlier comments about the frequency of hands that fit the "bill" for my requirement for an opening bid, I gave the frequency at which the bid would be opened, which came out to be above 2%, you were incredulous, saysing, "The data that you are presenting shows that the strong hand variant is four times are frequent across a reasonably large sample. Worse yet, you're only looking at Opening bids... The odds of this happening are absurdly low."

I think now, you will have to agree the odds are somewhat higher someone might be able to open with this bid than your calculations. The off by a factor 4 suggested to me where the diffeence came in.

Let me describe how I searched on bridgebrowser. I said, give me the opening hand where opener has 5-5 or better, and some number of hcp, I may have used 14, 15, or 16 (I vary these, in the trial above, it was 15 to be specific). The computer then looks at all hands where someone has to have that specific pattern (one of the four hands..but here, anyone of the four hands).. no one else would get to bid first... So, I suspect this simply is the chance one of the four hands has the right pattern. I even commented that from my studies, this number appeared a little high... but when I am looking individually, no one would have been able to bid in front of the strong hand. Also 15 hcp is not a ridgid requirement. With 6-5, 13 does quite nicely very often, raising the frequency a small amount.

Bridgebrowser can also limit searches by "absolute position" (by 1st, 2nd, 3rd, or 4th seat), If it can search by NORTH, I don't know how to do it. You can limit search by vul (any combitination), by imps versus mp, by controls, etc. So you can get quite sophisticated (how effective are 3D preempts vul at imps verus not vul if you like. Or how effective is a third seat preempt of 3D at this condition? How aobut 3 seat preempt iwth only 6 diamonds? How about thrid seat preempt with bad KJxxxxx diamond or worse? IT is easy to do.

A further search turned up, that is third seat, the chance of two passes to you when you have 55+ and are strong is even lower still. The "frequency" of this pattern with two passes in front of you is only... 0.3%, down from 0.77%. This brings the total from first three positions to 1.6%. I haven't checked fourth seat yet, but I expect it lower still, so it will not reach 2%. I had estiamted it as under 2% in another thread, so the 2.5 % observed in the vugraph was a surprise. Not a large enough "n" obviously.

Can we agree that the rate of these hands where someone will have the opportunity to open them with strong holding is approxiametly 1.5%? This is less than this caluculated rate, but not all 55+, 15+ hands can open this, and some with less hcp not included here can.

Ben

[hrothgar]

inquiry, on Apr 8 2005, 05:42 PM, said:

Can we agree that the rate of these hands where someone will have the opportunity to open them with strong holding is approxiametly 1.5%? This is less than this caluculated rate, but not all 55+, 15+ hands can open this, and some with less hcp not included here can.

I'm not interested in the percentage chance that "someone" is dealt hand ABC, but rather, the percentage that my partnership is dealt a hand ABC... By assuming that everyone at the table is playing the methods, you're double counting. Now, take that 1.5%, divide by two and I think that you're in the right ballpark. (Of course, you're also back, once again to about .77%...)

If you really want to model the frequency with which you are dealt a suitable hand for your methods, you need to calculate the following:

(The percentage chance that my partnership is sitting in first seat/third) *
(The probability that I have hand type ABC in first seat)

+

(The percentage chance that my partnership is sittng in second/fourth seat) *
(The conditional probability that I have hand ABC given a first seat pass) *
(The percentage chance of a first seat pass)

+

(The percentage chance that mypartnership is sitting in first/third seat) *
(The conditional probability that I have hand ABC given a first/second seat pass) *
(The percentage chance of a first/second seat pass)

+

(the percentage chance that my partnership is sitting in second/fourth seat) *
(The conditional probability that I have hand ABC given a first/second/third seat pass) *
(The percentage chance of a first/second/third seat pass)

[david_c]

Quote

If you really want to model the frequency with which you are dealt a suitable hand for your methods

I like the word "really" here. When you're asking about the frequency of an opening bid, trying to consider all four seats in the same calculation is an odd thing to do. The first question to try to answer is whether transfer pre-empts are a good method in first seat. The answer to this does not depend at all on how often you expect it to come up in other seats. Once you've answered this question you can go on to ask whether the method is a good one in second seat - which is harder because it depends on the opponents' system. And so on. At no point is it useful to combine statistics for more than one seat.

[inquiry]

Ok, I made a mistake in the calculation of the third seat bid after two passes. I didn't have all four vul conditions picked. The numbers below reflect the correction for all seats...

This table is "frequency" on BBO of Opening bid of 3C/3D/3H or 3S with less than 12 hcp (first column), and frequency of hand with 5+5+ and 15 hcp (second column), the rows are by position (dealer, 2nd seat, 3rd seat, 4th seat).

1st   1.27   0.77
2nd   0.63   0.63
3rd   0.21   0.37
4th   0.19   0.20

A few things surprise me in this table, the first is that the third seat preempts at the three level are so low. I suspect this might be because of more frequent openign at four level if weak, and opening at one level with less than normal values instead of three level. The second is that the "frequency" of the weak bid is much closer to the strong bid than I would have ever suspected. Especially in seats other than the first. No doubt if people have passed to you, the chances you have a strong hand compared to a weak hand increase. Also note, while the total percent hands are 2.27 for preempt (all the same data set), and 1.97 for the chance to open the strong bid, a couple of issues here. The rquirement I used was "15 chp", not the correct 4 losers. so I think the 1.97% is still a tad high.

Let's compare for Richard... .

First and third seat, total is 1.27 and 0.21 for weak preempt, for 1.45. Forstrpmg ppair, +1.14%

For second and fourth seat, weak hand are only 0.82% and strong 0.83%,

Again, lets be clear on how these numbers were obtained. I looked at 2 million hands, 16 plays per hand (main room). I asked how many times 3C/3D/3H/3S was opened in each position (first bid). And calculated the total. On some hands, only one player opened 3C, everyone else passed. That hand counted as 1 out of 16... etc. Then I did same thing, I said based on position did someone have 5+5+ and 15+ hcp. In first chair, all hands counted. In other chairs, the bidding had to be all passes before that hand at the table, or it did not count.

As I said, this represents over 2 million hands of bridge and more than 125,000 actual different deals, and it includes the variabilyt of what different people do with each hands (preempt or not light, open light in front of your 5+5+ hand or not), psyce a bid before you get a chance to open or not.

I now put this question to rest, for me. The opening strong version occurs at least at a frequency worth using it if you don't have some other major objection (like not being able to open 3C).

Ben

[hrothgar]

inquiry, on Apr 8 2005, 07:24 PM, said:

Ok, I made a mistake in the calculation of the third seat bid after two passes. I didn't have all four vul conditions picked. The numbers below reflect the correction for all seats...

This table is "frequency" on BBO of Opening bid of 3C/3D/3H or 3S with less than 12 hcp (first column), and frequency of hand with 5+5+ and 15 hcp (second column), the rows are by position (dealer, 2nd seat, 3rd seat, 4th seat).

1st   1.27   0.77
2nd   0.63   0.63
3rd   0.21   0.37
4th   0.19   0.20

A few things surprise me in this table, the first is that the third seat preempts at the three level are so low.  I suspect this might be because of more frequent openign at four level if weak, and opening at one level with less than normal values instead of three level. The second is that the "frequency" of the weak bid is much closer to the strong bid than I would have ever suspected. Especially in seats other than the first. No doubt if people have passed to you, the chances you have a strong hand compared to a weak hand increase. Also note, while the total percent hands are 2.27 for preempt (all the same data set), and 1.97 for the chance to open the strong bid, a couple of issues here. The rquirement I used was "15 chp", not the correct 4 losers. so I think the 1.97% is still a tad high.

You still need to divide by 2 to reflect that fact that your partnership does not sit North/South and East/West. With this said and done, we now have relative freqencies suggesting how how often you have a weak 3 level preempt relative to the strong hand type.

The next step is to estimate expected gains and losses for adopting the methods... How much (if anything) does one lose by opening 3♦ with a weak heart preempt? How much does one gain by immediately describing a strong 5-5 shape and adding definition to our 1 level openings...

[inquiry]

Friday's ACBL Event.

Earlier in this thread, I said we would look at friday and saturday's ACBL event on BBO to examine how the bids would work (before friday). The friday's hands have been played. And here are the weak three and strong two suited hands from those events. There was only 180 ACBL hands played on Friday. If you bid 6C, it makes on any defense other than trump and trump.

Scoring: IMP

---

North opened at most tables (he would not get a chance if I was WEST). My auction would be...(with E/W passing)
3C-3D-3N-?

Responder know, not quesses, that 5♣ is playable. The question is can he stick in 3NT? Since this is matchpoints, he can consider pass, or he can try for slam. If partner is missing the heart Queen, the slam will be a fair bet (his losers in that case will presumably be somehting like two off suit aces, heart queen, and one more, maybe a club honor or another heart honor). South might pass, or might bid 5C or 6C. No science here. If he bids 6NT, will EAST find the club jack lead? No one in 6C got the club lead, so I doubt it. And an auction that goes 3C-3D-3N-6C is hard to defend against. I will not guess where this auction will end (I would shoot 6C).

Result      Points      Score
3NS+1      630      100.00%
5CN+1      620      90.91%
3NN=      600      72.73%
3NS=      600      72.73%
3NS=      600      72.73%
3CN+3      170      54.55%
3CN+2      150      45.45%
3CN=      110      36.36%
6CN-1      -100      18.18%
6CS-1      -100      18.18%
5CN-1      -100      18.18%
3CN-2      -200      0.00%

The next hand of interest is a preempt hand. It was board 12 from #155 Pairs .----------- ACBL Fri 5pm ---

Scoring: IMP

---

Ok, I would open 3♦. I doubt this will have any affect of the auction, other than if north can DBL to show diamonds. Their getting to auction is probably not good for them.. As EAST I would bid 3♥ (remember at this vul, my preempt shows five undertricks, so heart ACE, and club AK cover three of those losers, I need to cover two more just to make 3H.. so I would never consider bidding 4H. Now the bidding goes back to north, do you really think he can stay out of the auction now? If NS bid, they will regret it. And I can make 3♥. So I don't think I am hurt here by the the transfer preempt.

Result      Points      Score
5HxW-2      300      8.77
5HW-2      100      5.23
5HW-2      100      5.23
5HW-2      100      5.23
5HW-2      100      5.23
4HW-1      50      3.77
A--      0      0
A--      0      0
4SxS-1      -200      -1.69
4HE=      -420      -6.46
4HW=      -420      -6.46
4SxS-2      -500      -8.08
4SxS-2      -500      -8.08
4SxS-2      -500      -8.08

The next is hand six from #100 Pairs .---------- ACBL Fri 6:30pm -

Scoring: IMP

---

Playing MisIry, south opens 3D, then rebids over 3H (or 3S), 3NT. This shows minor 2 suiter, 4 losers. North knows DQ is cover, and figures SA is a cover and heart King might be. The spade King is another possible cover. My auction would be...

3D-3H-3N-5N-6D-Pass

Here 5N was pick a slam. If partner picks clubs, I will put the contract in 6NT. If he picks diamonds, he will have six. But if I respond a forcing 3♠, the auction will now be...

3D-3S-3N-5N-6D-Pass... but if partner had bid 6C, instead of bidding 6N, I would bid 6D giving choice between 6D and 6S. PArtenr would choose 6D.

Result      Points      Score
3NN+2      460      0.88
3NS+2      460      0.88
3NN+2      460      0.88
3NN+2      460      0.88
3NS+2      460      0.88
3NS+2      460      0.88
3NN+2      460      0.88
5NS=      460      0.88
3NS+2      460      0.88
3NN+2      460      0.88
3NN+2      460      0.88
4SN+1      450      0.88
4SN+1      450      0.88
5DS+1      420      -0.12
5DS+1      420      -0.12
5DS+1      420      -0.12
5DS+1      420      -0.12
6NxS-1      -100      -11

This is board 9, for the same tourney (100) above.

Scoring: IMP

---

I show this one because Some Did preempt 3C on the north hand. The way I play MisIry now, I have no way to preempt in clubs below the four level. So I would have to pass. But pass or bid 3C, the results are the same on this hand.

Result      Points      Score
6SxW-2      500      13
3HE-3      300      10.76
4SW-1      100      8
4SW-1      100      8
4SW-1      100      8
4SW-1      100      8
4SW=      -620      -4.12
4SW=      -620      -4.12
4SW=      -620      -4.12
4SW=      -620      -4.12
4SW=      -620      -4.12
4SW=      -620      -4.12
4SW+1      -650      -5.06
4SW+1      -650      -5.06
4SW+1      -650      -5.06
4SW+1      -650      -5.06
4SW+1      -650      -5.06
4SW+2      -680      -5.76

Finally, I show board 6 from #187 Pairs .----------- ACBL Fri 2am ---

Scoring: IMP

---

A fair number of people opened 3D on this hand. Horrible. The vul was wrong, the suit quality was wrong, and you had a side four card major. Biddign 3D here would not occur to me. But even If I did, I doubt it would have had too much of an effect on the outcome versus a 3D opening bid. I will not rate this, because it is not a preempt I would use anyway.

4SS+1      450      5.62
4SN=      420      5
4SS=      420      5
5CW-4      400      4.25
5DE-3      300      1.88
4DE-2      200      -0.88
4CW-1      100      -3.38
4SxS-1      -100      -7.75
4SxxS-1      -200      -9.75


[B]Summary: 180 hands. Two good strong hands, reeach laydown 6D on one, have shot for normal 3NT/5C or lucky 6C on the ohter. Really only one preemptive hand I would preempt on, and no differences on thiese that I can see. The rate is a little low here for both opening preempts at the three level and I think the strong hand. But Saturday is another day. Even if you accept the horrible preempts on 3D and 3C, the above, that is only 3 preempts at the three level.

Ben

[awm]

Just a question about the 6♦ slam hand. Your suggested auction was:

3♦-3♥-3NT-5NT-6♦

Apparently 3♦ followed by 3NT shows a minor two suiter with four losers, and 5NT is pick a slam. But wouldn't the following hands for south also rebid 3NT:

x
Ax
AKxxx
AKxxx

x
x
AKxxx
AKxxxx

-
xx
AKxxx
AQJxxx

-
Ax
AKxxx
ATxxxx

Honestly I would have no great desire to be in slam with the north cards opposite any of these hands. With 2-1 in the minors it seems awfully aggressive by north to force to slam despite holding good cards. Unless there's some way your methods can find the SIXTH diamond, which is really what makes slam so good, this one seems unconvincing. Curiously, my relay auction starting with 1♣ has a fair chance to be unimpeded and find 6♦ on this hand.

[awm]

The analysis on the 3♥ preempt hand is also interesting. Certainly you get something of a win here because partner supposedly will not bid 4♥. This is especially interesting because 4♥ will make if one of the two heart honors is onside, or if the club finesse is on. Even with both heart honors off and a 6-1 diamond break, you will make if ♣Q is on. It's an excellent game that you happened to avoid (and find the magic lie of the cards such that avoiding it is right).

However, regardless of east's raise or non-raise, playing natural preempts it will be very hard to pass with the north hand. The board results are good evidence of this. If north doubles, we could easily see:

3♥-X-4♥-4♠

And now the good easts are doubling for a top board.

On the other hand, playing my preferred defense to transfer preempts:

3♦-Pass-3♥-Pass
Pass-X...

North passes initially, knowing he will get another chance to bid. The double at the second chance is a balancing double, but denies sound values. With a GOOD takeout double, north doubles 3♦ initially to show "cards." Knowing that partner has only a balancing double and not sound values, south becomes much less likely to compete to 4♠.

It seems like finding hands and presenting hypothetical auctions can really go either way. And the field in ACBL tourneys is not that strong which may also bias the results. I guess there's no way to find out for sure except through actual play.

[inquiry]

awm, on Apr 13 2005, 02:54 PM, said:

Just a question about the 6♦ slam hand. Your suggested auction was:

3♦-3♥-3NT-5NT-6♦

Apparently 3♦ followed by 3NT shows a minor two suiter with four losers, and 5NT is pick a slam. But wouldn't the following hands for south also rebid 3NT:

x
Ax
AKxxx
AKxxx

x
x
AKxxx
AKxxxx

-
xx
AKxxx
AQJxxx

-
Ax
AKxxx
ATxxxx

Honestly I would have no great desire to be in slam with the north cards opposite any of these hands. With 2-1 in the minors it seems awfully aggressive by north to force to slam despite holding good cards. Unless there's some way your methods can find the SIXTH diamond, which is really what makes slam so good, this one seems unconvincing. Curiously, my relay auction starting with 1♣ has a fair chance to be unimpeded and find 6♦ on this hand.

The 6♦ hand shows, indirectly, the problem with the MisIry bid, that needs to be considered" That is, when you transfer with the strong hand (say five-five), your partner's suit can get lost. This doesn't happen when you have the suit your partner doesn't transfer too, if you don't want it too. You can reject the transfer and bid your suit (or jump in your suit to show solid type suit). But on the 5♦ hand, north is not strong enough for a forcing 3♠ bid.

Here was why I shot 6♦. Over 3NT, north has to take a gamble. He can pass or bid some number of diamonds. At this point, however, everthing else is a slam try in a minor. That is, if he failed to bid 3♠, there is no reasonable way to get to spades now. So to pass 3NT or to bid 5♦ both gamble that 4♠ is not the right spot. 3NT surely is playable, as you have extra strength here, and is a normal result as the result show. However, opposite as little as

x
Ax
AKxxx
AKxxx

You drew up, even with a 4-1 diamond split, you have 1C ruff, 4D, 2H. 2S. amd 2C, so there is chance for either an elopment with your fifth trump, or to set up a long spade, or a squeeze.

As an aside, this analysis wasn't so much for what it the right way to bid the found hands. it was more of a frequency thing.. (weak versus strong) in response to richard's questions. I lacked the time to do Saturday;s yet.. I had better do them quick before they get lost... otherwise, we can look at next saturdays.