[gwnn]

rhm, on 2011-January-07, 15:59, said:

What about speaking English for a change and calling both 3♥ bids raises, but one of these raises is a sign-off while the other one is invitational to 4♥?

Rainer Herrmann

What about answering my question that you ignored first? I will quote it for your convenience:

gwnn said:

Please just answer me this: if 3♣ is played as a raise which could include a 10 point hand with no shortness, what do you think 3♥ is played as? How often do you think it is accepted?

[rhm]

gwnn, on 2011-January-07, 19:49, said:

What about answering my question that you ignored first? I will quote it for your convenience:

Oh yes now I understand.

If you were serious you would have bid 3♣.
Your 3 ♥ bid is just a courtesy raise, which is almost never accepted unless over-caller has a super maximum.
Besides it is so much more fun to go down in 3♥ after a courtesy raise when you might have had a fighting chance in 2♥.

Rainer Herrmann

[Fluffy]

Oh my god this is so funny, much funnier than explaining restricted choice to foo. Please keep going, I remain in the dark, what can I bid to invite partner to 4♥?

[gwnn]

rhm, on 2011-January-08, 02:49, said:

Oh yes now I understand.

If you were serious you would have bid 3♣.
Your 3 ♥ bid is just a courtesy raise, which is almost never accepted unless over-caller has a super maximum.
Besides it is so much more fun to go down in 3♥ after a courtesy raise when you might have had a fighting chance in 2♥.

Rainer Herrmann

Nobody said I would have bid 3♣. Remember I even said that pass would not be horrible, only that I think it's irresponsible vulnerable at teams.

I asked you if 3♣ in opening poster's system can be a 10 count, what is 3♥? The answer is:

Spoiler

it is a bid that shows a hand weaker than a 10 count, hence not accepted 50% of the time.

The problem with this thread (and I will try not to make this sound personal) is that the discussion derailed from what would you bid from East/West to what should 3♥ show? It is crystal clear what 3♥ shows in the opening poster's system since 3♣ is clearly not a gameforcing relay, it shows a "good hand with fit". I don't agree with 3♣ on this hand, but the fact that East bid 3♣ makes it clear that it is not a game forcing bid with a fit, it is an invitational bid with a fit.

[rhm]

gwnn, on 2011-January-08, 07:00, said:

Nobody said I would have bid 3♣. Remember I even said that pass would not be horrible, only that I think it's irresponsible vulnerable at teams.

I asked you if 3♣ in opening poster's system can be a 10 count, what is 3♥? The answer is:

Spoiler

it is a bid that shows a hand weaker than a 10 count, hence not accepted 50% of the time.

The problem with this thread (and I will try not to make this sound personal) is that the discussion derailed from what would you bid from East/West to what should 3♥ show? It is crystal clear what 3♥ shows in the opening poster's system since 3♣ is clearly not a gameforcing relay, it shows a "good hand with fit". I don't agree with 3♣ on this hand, but the fact that East bid 3♣ makes it clear that it is not a game forcing bid with a fit, it is an invitational bid with a fit.

Reminds me of a story I read somewhere years ago.

Declarer, an expert, went down in a contract because dummy overbid.
In the postmortem dummy explained: " I had to show you my 3 kings" upon which the expert replied "But why could you not wait till you put your hand down."

Rainer Herrmann

[gnasher]

rhm, on 2011-January-07, 10:03, said:

AWM claimed to have done a simulation. Different to me he did not specify his assumptions, he did not say how many deals he generated and looked at. This is very dubious to start with.

I agree with Rainer that providing the simulation results alone is of doubtful value. If you don't provide the details of both the method used and of any subjective decisions made, it makes it hard for people to reproduce your results, or to judge how reliable they are.

If your simulation involves no manual intervention, it's sufficient to provide the constraints (though I usually include the actual code), because that's enough to make it reproducible and criticisable. If, however, you make a subjective decision about each hand, you should provide actual the hands, so that other people can decide whether they agree with your decision.

Look what happened here: for both simulations we were given details of the constraints, and very properly both sets of constraints were discussed and criticised.

Rainer used double-dummy analysis, and some people questioned the reliability of double-dummy analysis for this purpose, so he directed them to an analysis of the consequences of using double-dummy analysis. We can now factor that into our evaluation of Rainer's simulation.

Adam used his judgement, but didn't show us the hands, didn't tell us how many hands he looked at, and didn't tell us what proportion of the hands favoured each action. Hence we have no idea whether there was any inadvertent bias in his analysis, or how accurate his analysis was, or whether he analysed enough hands to make his results reliable.

Quote

Since he claimed he looked at them by hand, hopefully carefully, it can not have been many for the reasons above. This method is far more subjective and biased than large sample double dummy analysis by software.
Even if his analysis by hand is correct, which is difficult to accomplish, and the few deals have been generated randomly according to the specifications, the margin of error will still be high.

It may or may not be more reliable than your double-dummy analysis - it depends upon how many hands he analysed, how good his analysis is, and how strong the effects he noticed were.

[cherdano]

gnasher, on 2011-January-08, 08:15, said:

Rainer used double-dummy analysis, and some people questioned the reliability of double-dummy analysis for this purpose, so he directed them to an analysis of the consequences of using double-dummy analysis. We can now factor that into our evaluation of Rainer's simulation.

That's not accurate. Just because on average, 4-level contracts make the same number of tricks in single-dummy and double-dummy play doesn't mean this will also be true for this particular hand.
For example, on this particular hand RHO has few hcp and thus maybe only one chance, on opening lead, to take LHO off endplays or to set up a trick in a suit that can only be broken from his side. Meanwhile we know that clubs, the suit he is most likely to lead, is probably not that suit.

[gwnn]

actually now i understand. edited out.

[gnasher]

cherdano, on 2011-January-08, 09:05, said:

That's not accurate. Just because on average, 4-level contracts make the same number of tricks in single-dummy and double-dummy play doesn't mean this will also be true for this particular hand.
For example, on this particular hand RHO has few hcp and thus maybe only one chance, on opening lead, to take LHO off endplays or to set up a trick in a suit that can only be broken from his side. Meanwhile we know that clubs, the suit he is most likely to lead, is probably not that suit.

Indeed, there may be several valid criticisms of Rainer's reliance on double-dummy analysis. But my point was that he's given us sufficient information to enable us to identify and discuss such problems, whereas Adam hasn't.

This isn't particularly a criticism of Adam - I know it's quite common on these forums for people to provide simulation results in the way he did. I just don't think they should.

[whereagles]

Well, I think you can get very valid conclusions by generating 10000 hands with constraints and double-dummying them.