[benlessard]

High win

Q9---A2 (eliminated)
Q65---A9
KQ62---A9 (however north has a 1S opening so its only good if East is dealer)

low win
A952---q6 (elim there is a double dummy endplay IRL 100% going down)

So for 4000 hands it end with 3-2 in favor of J wich simply mean that you wont get rich by leading 1 or the other.

Its really even.

[benlessard]

i also made a sim for

97,QJT54,A65,765

against (1Nt)--(3nt) and high H is clearly better

http://www.bridgebas...ead-vs-1nt-3nt/

[bluecalm]

Quote

These numbers surprise me.

Am I misinterpreting them. It seems that almost every time the card led was crucial since they add up to only slightly more than 1000.

This is "amount if hands where given card defeats the contract" not "amount of hands where given card is the best lead".

[bluecalm]

Btw:

NEC cup 2011, semi final:

---

Bocchi chose low ♥ and 1NT made instead of going 2 down.
Simulation (winning lead):

K♠ - 556
K♥ - 677
3♥ - 521
2♦ - 492
7♣ - 633

Clear enough. Even the best of the best cardplayers could learn something from those simuls (according to double dummy simul his choice was about 1imp/hand worse than the best choice so it's quite a lot)

[MrAce]

Personally I think Bocchi made the correct lead. It was mentioned b4, but i will say it again, double dummy analysis is extremely poor to reflect the real outcome, when it comes to low level contracts for defense. It is very hard for defense in real life, to figure out the best defense, they have very limited info b4 the lead and even after the lead about the shape and strength of opponents.

Some may argue that it is same for declarer, which i don't believe. Declarer, whether he plays correct or wrong, 2 hands he controls will be in cooperation with each other, while for defense thats not the case since it involves 2 people.

Simulations are great tool, no doubt. But i think they will do their real work when they are capable of performing the BEST % single dummy play and defense instead of double dummy play and defense.

KJTx vs A98x playing 7 of this suit everything else is solid and it is split 3-2 or 4-1, simulation will show it as "makes" % 100 of the time, but in real life without any other info, people will fail to make it % 50 of the time. This is a huge gap.

In fact, a human defender, if sure they play in a 4-4 fit, can lead trump from Qxxx and reduce declarer's chance to almost % 0

[han]

MrAce, on 2011-June-17, 12:23, said:

KJTx vs A98x playing 7 of this suit everything else is solid and it is split 3-2 or 4-1, simulation will show it as "makes" % 100 of the time, but in real life without any other info, people will fail to make it % 50 of the time. This is a huge gap.

A good declarer might play a high honor first, catering for a singleton queen.

[Zelandakh]

bluecalm, on 2011-June-12, 10:59, said:

I have noticed that especially American players lead too aggressively. I even remember that Lawrence in one his books wrote that leading from xxx is usually terrible preferring active leads from honors. This is wrong and modern players lead more passively, especially ones from Europe.

Hi blue,

I was meaning times when the "best" card in a given suit is against the tradional advice. For example, the tradtional lead from QJTxx (3 sequence) is the Q (or J playing Roman); similarly from QJ9xx (broken 3 seq) but low not from QJ8xx. So I am interested in times where this traditional advice is wrong. An example is the Bocchi hand where low from KQ8xx is the traditional lead but this was wrong here - but was it wrong in general?. What would be really nice is examples of when it is right to lead low from suits headed by KQJ, KQT, QJT, QJ9, JT9, JT8, HJT, HT9 and conversely when it is right to lead an honour from holdings like KQ9, KQ8, QJ8, JT7, something the Italians seem to do more than other nationalities despite the Bocchi example given.

As far as Americans leading aggressively and Europeans passively goes I think this varies. Again, looking at traditional leads, even Culbertson was quite fond of leading xxx in an unbid major. One thing that has changed is that invites have gone out of fashion on marginal hands. Once upon a time players (generally) led aggressively after 1NT - 3NT but passively after 1NT - 2NT - 3NT. Now when the first sequence is likely to be on 23-24 hcp this has gone by the wayside and leading passively makes more sense. That is precisely one of the advantages of this bidding style after all! I am not sure such style issues are so easy to classify using a double dummy analysis method but the question of the best card in a given suit (honour or small) seems to be something that we could easily classify and make new sub-rules for which could be of benefit even to low intermediates.

[MrAce]

han, on 2011-June-18, 03:50, said:

A good declarer might play a high honor first, catering for a singleton queen.

Ohh sorry that closes the gap by a lot indeed. What u said is no different than spell check. Next time i write % 47 time fail

[Cascade]

JLOGIC, on 2011-June-11, 19:11, said:

xx KQT7x Qxxx xx playing imps

1N p 3N all pass.

Basically I'm curious what is the cutoff point where we stop leading the HK. Or do we always lead it, even without the DQ? Obviously it wins on AJ doubleon or Ax+Jx when they don't have 9 runners which seems easy enough to quantify. The harder part to quantify is how often we have an essential entry, and how often we can beat it by just being passive. I feel like the DQ with nothing else is a pretty good cutoff, with a king or even some stray jacks I would feel good about the HK lead, and with just a jack or nothing I would feel fine about leading a low heart but this one felt tough.

100 hands single dummy using GIB as declarer/defender

♥K vs ♥7 (very similar for ♥x which did you intend as the choice)

284-70 IMPs

I am rerunning the simulation for ♦xxxx now.