Another "deal with preempt" poll

[cherdanno]

Ulf, would it change your opinion if you re-read the original post and noticed it was a clubs preempt, not a diamonds preempt?

[gnasher]

I don't agree that partner will raise with a fairly balanced 8- or 9-count and three-card support. I'd be much more worried about the hands where partner isn't going to raise me.

 

Edit: Sorry - got rid of the nonsensical example.

Edited May 4, 2009 by gnasher

[jdonn]

Whereagles! Let's see the actual deal so you can laugh at me.

I'm scared to, I strongly fear bidding will have worked out well!

[ulven]

Ulf, would it change your opinion if you re-read the original post and noticed it was a clubs preempt, not a diamonds preempt?

Oops!

 

Definitely changes things. For some reason I read it as a 3D opening, as evidenced by my post.

 

Thanks for pointing that out.

[eyhung]

Count me in the camp of not even close to bidding 3H. I don't see why we should be stretching to bid at the 3-level at either IMPs or MPs when we have length in the enemy suit, and a weak notrump. It's hard to imagine game on even if partner has a fit if partner is also balanced with 3 clubs, or weak with short clubs. Whenever partner has balanced 8-11 and hearts, he will raise and we will likely be too high. And when partner is broke or heartless, he will pass and we will still be too high.

 

The general rule of thumb I like is : If you have 3 or more enemy trumps, you shouldn't be eager to enter over a preempt. You should have a "classic" hand for your action. However, if you are short, then you can be aggressive. Change the hand to A42 AJ8543 A94 2 and now 3H seems right -- most hands with 8 working HCP and 3 hearts will now provide us with reasonable play for game, and partner is unlikely to balance since you are short in clubs.

[Fluffy]

Nobody pointed it out yet, but 3♥ leads to 4♥ many times, when it is very likelly that your best game is in 3NT wich we will bid whenever partner reopens with 3♦/♠.

 

I am nowhere close to bid with this, About a King short for a bid, but being MPs makes it more likelly to bid than the contrary (those 1400 aren't that painy after all).

[ulven]

I asked Tjolpe Flodqvist to make a sim with his program, which is able to generate HTML.

 

Here it is:

http://www.syskon.nu/sims/Preemt_poll.htm

 

I haven't checked the deals yet as I'm hitting the sack now (1 am here).

[cherdanno]

At matchpoints, this may be much closer than Justin and Josh are making it. Looking through the simulations, there were many gains from 4H-2 undoubled (or 3H-1/-2) against 3C making. It seems to depend a lot on how sound RHO's preempts are (if he is very classical as in good suit, little outside strength that favors bidding), how aggressive partner is in balancing, and how aggressive LHO is in doubling (might be a good idea to double aggressively against Ulf!).

At IMPs, I think bidding is suicidal.

[awm]

Looking at these, it seems like there are a lot of hands where 3♥ goes two down but might be hard to double, or where 3♥ is likely to get raised to 4♥ for two down, but might be hard to double. These are all good "saves" against 3♣ making. My quick scan didn't see many good games that wouldn't be reached anyway.

 

So it appears that bidding 3♥ might be good at MP scoring, since all the -100 results actually score up pretty well. But at IMPs it seems suicidal, since opponents have the occasional big number and many chances at +300 opposite air.

[xcurt]

I don't buy the argument that 3H is ok since we can go -100 vs -110. minus scores that beat par matchpoint poorly if we aren't in wc field. Bidding lays big mp odds. I'm passing smoothly.

[xcurt]

I asked Tjolpe Flodqvist to make a sim with his program, which is able to generate HTML.

 

Here it is:

http://www.syskon.nu/sims/Preemt_poll.htm

 

I haven't checked the deals yet as I'm hitting the sack now (1 am here).

It looks from the bottom of this page like it only took 472 deals to generate 102 hands. If this is the case, I'm skeptical about the accuracy of the probability distributions until proven otherwise.

[Codo]

Thanks Ulven,

 

I love it when an opinion is backed up with facts.

 

Most contributors like to back up their opinion just with strong fealings.

[gnasher]

It's an overstatement to describe the results of a simulation as fact. In this case, I think it's quite a long way from the real world.

 

On at least six of the first ten deals I would have opened RHO's hand at the one level. I know my views on what constitutes a preempt aren't quite middle of the road, but even so I don't believe that a majority of good players would open 3♣, first at adverse, on a ten-count with a side ace.

 

The conditions of the simulation require that North have six losers. That almost guarantees that he'll have a side ace or king. If that is their agreement, it will be a lot easier for LHO to double us, so on many of the hands where it looks like we'll get away with 3♥-2 or 4♥-2 we will, in fact, be doubled.

 

Alternatively, if they don't have the agreement that the preempt promises six losers, the simulation is of little value, since it excludes a large number of deals which should have been included.

[Codo]

Hi Andy,

 

the given hands are facts. How they are generated is a fact.

 

I would never argue that all these hands are 3 club openings, nor that these hands are the ultimate proove that 3 HEart (or pass or any other bid) are the one and onyly way to win.

 

But these are 102 deals, where anybody can get his own conclusion whether or not bidding 3 HEart on these deals had been right or wrong and whether the conditions of the simulation matches reality as good as possible.

 

Ulf stated himself that his analyses are influenced by his opinion, same is true for all of us, so which conclusion you take from these hands is your personal thing- and no fact.

 

I agree that 6 losers (for east) is quite regressive and makes it easier for them to double. And at least to me it is far from clear what will happen if you allow lower limit openings, I belive that it makes bidding more dangerous, but what do I know.

[whereagles]

Well, I must confess I wasn't expecting such a lively discussion. The actual deal was probably quite far from any simulation:

 

[hv=d=e&v=e&n=sj842ht4da986cj73&w=sq765h962dkq752ck&e=skt3hkq7djcqt9865&s=sa9haj853dt43ca42]399|300|Scoring: MP

East South

3♣ ...??[/hv]

 

I posted this because it seemed to me like nice practical deal to illustrate that very important principle of post-preemtptive bidding, which is to ask oneself the question:

 

"Do I need to bid NOW?"

 

to which the answer is, in his case and in my opinion, "Probably not. We are long in clubs, so pard is likely to be short there and will balance if he has some cards."

 

Ok, you can argue East's hand is no where near a red 3♣ opener, but these things happen and you have to deal with them, whether you like it or not. For the record, I held the South hand and passed (& have witnesses to prove it.. lol). It went -2 for most of the matchpoints. Bidding would have led to a really bad score.

 

I think it would have been a much more difficult problem if we held club shortage.

[hotShot]

I think the decision is closer than one might think:

 

East restrictions: HCP: 5-11, 6- loser, shape: ♠0-3, ♥0-3, ♦ 0-4, ♣7+

South hand set to: A9 AJ853 T43 A42

 

Looking at 1000 simulated deals:

4+♥ makes: 377 times

3♥ makes: 203 times

2♥ makes: 177 times

opps 3♣ only makes: 468 times

Failing in 3♥ while 3♣ won't make: 159

 

So pass will give you a positive score about 53% of the time (and if partner has a good hand, he will reopen).

 

If you manage to bid 3♥ and partner will only bid more with good cards, you will score positive about 57% of the time.

Down 1 even dbled is still better than 3♣=.

 

Only in about 16% you will go down, while opps won't make.

 

Since this is MP, it does not matter much if you score -300 or -1400 as both will be (close to ) zero. So if my agreements include such a hand over a 3 level preempt, I'll bid 3♥, knowing my partner will not go wild with a weak hand.

At IMPs big numbers can hurt your overall result much more, so I would probably pass.

[hotShot]

Just rerun the simulation allowing opener to have only 6♣ and this changes a lot.

 

Now opps only make their 3♣ (or better) 42% of the time, so in 58% of the cases you are better off passing.

You can only make 3♥ or better in 52% of the boards and both sides going down goes up to 23%.

This makes bidding a lot less attractive.

[whereagles]

Oh, I didn't think the person in case could open on a 6 card suit, but he did. In any case 6 cards certainly are the exception rather than the rule. Especially since people who open red on 6 cards usually have a pretty good suit and extra shape, e.g. 6-4.

[Fluffy]

What did you lead Nuno?, diamond doesn't look best :)

[whereagles]

I led the ♠A and played another spade.

[wclass___]

"I have observed many bridge players and came to the conclusion that many of them, including some very good players are very conservative when it comes to overcalls, but they are extremely aggressive once their partners make an overcall.'' - Melih Ozdil

 

This is a hand with many positive and negative factors to consider, therefore i think that hand simulation with analyses is right path to go.

 

I have no clear preference, so i will bid 3♥

 

(knowing opponent style is important, but i think i know how looks 1st seat vulnerable 3♣)

[jdonn]

I think the decision is closer than one might think:

 

East restrictions: HCP: 5-11, 6- loser, shape: ♠0-3, ♥0-3, ♦ 0-4, ♣7+

South hand set to: A9 AJ853 T43 A42

 

Looking at 1000 simulated deals:

4+♥ makes: 377 times

3♥ makes:  203 times

2♥ makes: 177 times

opps 3♣ only makes: 468 times

Failing in 3♥ while 3♣ won't make: 159

 

So pass will give you a positive score about 53% of the time (and if partner has a good hand, he will reopen).

That part in parentheses in the last sentence is really important! Also "So pass will give you a positive score about 53% of the time" can not be concluded from your results at all.

[hotShot]

I think the decision is closer than one might think:

 

East restrictions: HCP: 5-11, 6- loser, shape: ♠0-3, ♥0-3, ♦ 0-4, ♣7+

South hand set to: A9 AJ853 T43 A42

 

Looking at 1000 simulated deals:

4+♥ makes: 377 times

3♥ makes:  203 times

2♥ makes: 177 times

opps 3♣ only makes: 468 times

Failing in 3♥ while 3♣ won't make: 159

 

So pass will give you a positive score about 53% of the time (and if partner has a good hand, he will reopen).

That part in parentheses in the last sentence is really important! Also "So pass will give you a positive score about 53% of the time" can not be concluded from your results at all.

If opps make 3♣ 468 times of 1000 simulations, they go down in 532 of 1000.

If they go down I get a positive score.

532 of 1000 is about 53%.

 

I think that is a valid logical conclusion from my data.

[Echognome]

I think the decision is closer than one might think:

 

East restrictions: HCP: 5-11, 6- loser, shape: ♠0-3, ♥0-3, ♦ 0-4, ♣7+

South hand set to: A9 AJ853 T43 A42

 

Looking at 1000 simulated deals:

4+♥ makes: 377 times

3♥ makes:  203 times

2♥ makes: 177 times

opps 3♣ only makes: 468 times

Failing in 3♥ while 3♣ won't make: 159

 

So pass will give you a positive score about 53% of the time (and if partner has a good hand, he will reopen).

That part in parentheses in the last sentence is really important! Also "So pass will give you a positive score about 53% of the time" can not be concluded from your results at all.

If opps make 3♣ 468 times of 1000 simulations, they go down in 532 of 1000.

If they go down I get a positive score.

532 of 1000 is about 53%.

 

I think that is a valid logical conclusion from my data.

Without considering any other conclusion, if you write "3♣ only makes" then that would imply the opponents can make exactly 9 tricks with clubs as trumps. The key word being only. If you had written 3♣ or more makes, then we'd start to get there.

 

However, you are missing one key point. When we pass, it is not going to always go P-P-P. Sometimes LHO will raise and they will go down. Sometimes LHO will bid a new suit and they will find an alternative contract that may or may not make. Sometimes partner will take an action. Etc. etc.

 

So no, I do not think you have made a "valid logical conclusion."

 

No offense of course!

[jdonn]

I think the decision is closer than one might think:

 

East restrictions: HCP: 5-11, 6- loser, shape: ♠0-3, ♥0-3, ♦ 0-4, ♣7+

South hand set to: A9 AJ853 T43 A42

 

Looking at 1000 simulated deals:

4+♥ makes: 377 times

3♥ makes:  203 times

2♥ makes: 177 times

opps 3♣ only makes: 468 times

Failing in 3♥ while 3♣ won't make: 159

 

So pass will give you a positive score about 53% of the time (and if partner has a good hand, he will reopen).

That part in parentheses in the last sentence is really important! Also "So pass will give you a positive score about 53% of the time" can not be concluded from your results at all.

If opps make 3♣ 468 times of 1000 simulations, they go down in 532 of 1000.

If they go down I get a positive score.

532 of 1000 is about 53%.

 

I think that is a valid logical conclusion from my data.

If we pass, that doesn't mean partner will pass.