What's with 'fourth best'?

[oryctolagi]

Why do so many players rigidly stick to that lead in no trumps? Are we talking some sort of rule that 'must be obeyed'?

 

I was kibitzing a hand recently, declarer was in 3nt, and LHO was faced with:

[hv=pc=n&w=sq92ht62dq7cj7642]133|100[/hv]

Declarer and partner had bid both majors between them.

 

With my advantageous view of all four hands, I could easily see that a diamond lead would have killed the contract. But, predictably, out came the 4 ♣. And 3NT was made - easily.

 

With so poor a hand and only one potential side entry, surely West didn't expect to run his clubs! Wouldn't it be better to 'play for partner's strength'?

 

All I can say is, if I'm defending 3nt, don't expect a fourth best from me every time!

[Cyberyeti]

Depends on the exact auction, but:

 

So you lead a diamond and find partner with Jxx(x)/10xxx allowing declarer to make 4 tricks without losing the lead, that's likely to be fatal (although declarer might be fooled about the layout if partner has the J).

 

A club lead may give away a trick, but less likely to be 2 and might hit gold with partner holding A10x or Axxx.

[ggwhiz]

Geoff Hampson: I can't tell you how many imps I lost over the years from not leading 4th best from my long suit.

 

There are exceptions of course but that's good enough for me.

 

I played 1 matchpoint game with Lynn Deas and her last remark before we started was "Oh, and we ALWAYS lead 4th best against notrump. I did with 6 petunias in spades and out and caught her with Axx, 2-2 in declarers hands and with an entry in hers

[rmnka447]

The rationale for a 4th best lead pertains to the defense's strategy versus NT. With the declaring side having 25+ points, you aren't likely to beat the contract by taking high cards. Rather the objective of the defense is to set up and take long suit tricks before declarer can set up and cash 9 tricks. That depends on a couple things. First, you need a suit that has prospects of being set up. Second, you need an entry to be able to cash the long tricks.

 

Unfortunately, this approach to NT defense is rarely explicitly laid out to newer players. Instead, they are taught a hierarchy of leads starting with honor sequences, inner sequences, and leading down to 4th best and top of nothing. For more than a few players, the opening lead against no trump then simply gets down to running through this hierarchy and choosing the highest category of lead available in their hand. Often, it is 4th best.

 

A better approach for the opening leader is to ask "Whose suit are we trying to set up? My suit or partners?" as part of the process of deciding what to lead. To answer the question requires a little analysis.

 

First, opening leader can calculate partner's highest possible HCP by adding 25 to the points in his/her hand and subtracting from 40. The answer may tell if you are likely to be hit a suit which can be set up and if their might be an entry. Here, opening leader has 5 HCP, so partner has 9 or possibly 10 HCP at most. That's not a lot but enough to possibly have something in a suit and a side entry.

 

Second, if opening leader has a suit, opener needs to ask "Can my suit be set up? Do I have a potential entry?" Here again, the answer is ♠ Q might be an entry, but it'll take any awful lot to set up the ♣ suit.

 

Third, analyze the auction for clues as to what partner might hold. Sometimes what the opponents bid and don't bid can be very edifying. Here again the opponents are bidding the majors, so it seems like a minor suit lead should be made.

 

Putting it together, your best hope is to try hit partner with a ♦ suit and entry and lead the ♦ Q.

 

There is ONE BIG EXCEPTION. When the opponents have a "tortured" auction, find no fit and barely get to game, then you need to think about making a passive lead which gives away the least. Example - Opponents auction - 1 ♦ - 1 ♥ - 1 ♠ - 1 NT - 2 ♥ - 2 NT - 3 NT and you hold ♠ K432 ♥ 109 ♦ Q83♣ J962. They've barely limped into 3 NT with no great fit. What do you know about the hand from the auction. Well, partner has to have 4 ♥ - responder has 4 and opener 3 from his delayed raise. What lead potentially gives away the least? I found the ♥ 10 reasoning that with 4 ♥ to an honor it might help partner and probably wouldn't give much away. As it turned out partner held ♥ KJxx over Qxxx in dummy and helped us score 3 ♥, ♠ K and another to beat 3 NT.

 

One final example - opponents auction 1 NT - 2 ♣ - 2 ♥ - 3 NT and you hold ♠ 9 ♥ 1063 ♦ Q7643 ♣ J952. What do you know? With 3 HCP, partner can hold as many 11-12. My suits aren't likely to be set up and more importantly there's no entry. So, it looks like we need to find partner's suit. From the auction, responder has 4 ♠ and opener probably 3 at most. That gives partner at least 5 ♠. It wasn't going to be my suit, so I led ♠ 9 partner's known suit. (So much for not leading stiffs at NT.) Partner actually held 6 ♠ and 11 points and with the ♠ lead became able to beat 3 NT by 2 tricks. Any other lead would let opener make.

[mgoetze]

All I can say is, if I'm defending 3nt, don't expect a fourth best from me every time!

Of course it's not always best to lead from a 4+ card suit, but I'm not sure your (quite incomplete) example is a great demonstration of this.

[zillahandp]

They do it, or play 3/5th ( rules of 10,11,12 ) cos it works best but they also listen to the auction. I suggest you read Goren, better bridge for better players, chapter two. And you will be.

[rhm]

Geoff Hampson: I can't tell you how many imps I lost over the years from not leading 4th best from my long suit.

 

There are exceptions of course but that's good enough for me.

 

I played 1 matchpoint game with Lynn Deas and her last remark before we started was "Oh, and we ALWAYS lead 4th best against notrump. I did with 6 petunias in spades and out and caught her with Axx, 2-2 in declarers hands and with an entry in hers

Did Geoff Hampson and Deas tell you how many IMPs they lost by leading fourth best?

I bet they don't know either.

Humans have a selective memory and even experts are human.

If they choose an unorthodox lead and it backfires it hurts and memory will record the result.

If they choose fourth best from longest and strongest and another unorthodox lead would have fared better, the reaction is shrug your shoulder and get to the next board.

 

Not every notrump game centers around the issue, which side can first establish a suit. Often declarer has not enough tricks and needs help from the defense.

 

Do not get me wrong. I do not recommend a lead from shortest and weakest against notrump.

The subject is complex and no matter what's your strategy, it will often fail.

 

Instead of recommending simplistic rules, which are not really backed up by statistics, I recommend

 

Bird/Anthias book Winning NT leads

 

You can find a discussion and a review at http://www.bridgebase.com/forums/topic/62749-birdanthias-books-on-opening-leads/

 

Rainer Herrmann

[fourdad]

I hope I am declarer someday in a NT contract with you on lead! :rolleyes:

[PhilG007]

Why do so many players rigidly stick to that lead in no trumps? Are we talking some sort of rule that 'must be obeyed'?

 

I was kibitzing a hand recently, declarer was in 3nt, and LHO was faced with:

[hv=pc=n&w=sq92ht62dq7cj7642]133|100[/hv]

Declarer and partner had bid both majors between them.

 

With my advantageous view of all four hands, I could easily see that a diamond lead would have killed the contract. But, predictably, out came the 4 ♣. And 3NT was made - easily.

 

With so poor a hand and only one potential side entry, surely West didn't expect to run his clubs! Wouldn't it be better to 'play for partner's strength'?

 

All I can say is, if I'm defending 3nt, don't expect a fourth best from me every time!

The reason why 4th highest is led at NTs is simple...for partner to use the so called "Rule of Eleven" but it should be remembered not to

apply the procedure too rigidly. Only an automaton plays the same way all the time(!) Other ways of leading could be leading a suit partner has bid,

or leading 'top of nothing' It should also be remembered that a rule,ANY rule no matter how old it is is just a guide,not a muzzle. Even a top darts player

doesn't hit the bullseye every time(!)

[fluff]

For our part, we lead 3/5 all the time. Always leading 4th best vs NT is perfect in theory, but, sometimes, with the bidding pointing you towards the unbid suit, you will lead small from Hxx. So why not systematize 3/5 ? Partner, seeing your smallest, will now right away your count. Obviously, you will lead top from xxx.

[GrahamJson]

You say that declarer and partner had bid both majors between them. In which case it seems obvious to lead partner's suit (assuming a normal auction) especially as partner may have bid in order to direct a lead. This is particularly the case if he has overcalled.

 

I see that a diamond lead is the winner but, unless there were strong inferences from the bidding, I think you would need second sight, or a second PC, to come up.with that.

[ggwhiz]

Did Geoff Hampson and Deas tell you how many IMPs they lost by leading fourth best?

I bet they don't know either.

Humans have a selective memory and even experts are human.

If they choose an unorthodox lead and it backfires it hurts and memory will record the result.

If they choose fourth best from longest and strongest and another unorthodox lead would have fared better, the reaction is shrug your shoulder and get to the next board.

 

In the Lynn Deas case I think her reasoning was that I'll win on bidding and mid hand defense so I want to make the same leads as the field.

 

To not do so is swinging, fine if you have a decent reason or your game needs a boost but I have played against a few pairs that almost always lead short suits and they are palookas.

 

I also remember reading that Hampson (and Levine and Weinstein) are regarded as the types that will beat anyone in a KO but are not great at racking up big matchpoints against random fields so it comes down to what style of player you want to be.

[mgoetze]

For our part, we lead 3/5 all the time. Always leading 4th best vs NT is perfect in theory, but, sometimes, with the bidding pointing you towards the unbid suit, you will lead small from Hxx. So why not systematize 3/5 2/4?

FYP.

[Zelandakh]

I see that a diamond lead is the winner but, unless there were strong inferences from the bidding, I think you would need second sight, or a second PC, to come up.with that.

A couple of coughs ought to be sufficient. ;)

[zillahandp]

In autshell because it needs the least number of cards from your p to run the suit. In case you dont have the book or it has not occurred to you to look at defence this way

[barmar]

There seem to be two questions here:

 

1. Should I lead from my longest suit or not?

 

2. When I lead from my longest suit, should I lead 4th best?

 

Most of the responses seem to be for #1, but some are for #2.

 

The title implies that #2 is the question the OP was concerned with, but the mention of leading a diamond in the text indicates that #1 is his real question.

 

There are no hard and fast rules, but a reasonable guideline is to lead your longest suit if it's a strong suit or you have a side entry. But if your hand is very weak, even if you can set up the suit you may not be able to get back in; in that case, you often try to find your partner's suit, since he'll have the side entry to take advantage of it.

 

The hand in the question looks to me to be on the borderline of the criteria, but I think I'd lead a club. It's possible that declarer will have to take a finesse into one of your queens, so trying to set up your clubs could work. And sometimes you find partner with Axxx in that suit, and the suit sets up right away.

[1eyedjack]

The reason why 4th highest is led at NTs is simple...for partner to use the so called "Rule of Eleven"

Declarer is equally capable of applying the rule of 11, so it is not a one-way street even if that is the reason.

Furthermore, if playing 3rd and 5th leads, you just apply the rule of 10 and 12 instead.

All that is required is some consistency and agreement regarding what card you would lead from a holding.

[oryctolagi]

The point is certainly worth making, that all defensive conventions (including things like HELD and McKenney) might give information to declarer as well as to your partner. I don't suppose there's any getting away from it: you can't (legally) exchange any information with partner without giving declarer a few clues...

 

Unlike psyching or other unusual bidding, I don't think there are any rules laid down in the big tournaments, about falsecarding during play (assuming you don't have an undeclared agreement with partner). Or are there?

[oryctolagi]

Of course it's not always best to lead from a 4+ card suit, but I'm not sure your (quite incomplete) example is a great demonstration of this.

You're right, I didn't give full details of this hand because I wanted to have the topic discussed on more general terms (as it has indeed turned out).

 

Bear in mind I was only kibitzing, not playing this hand. As it happens, East held AJ10xx in ♦ plus a side entry, while declarer had Kxx. But East hadn't bid, so it would have been down to guesswork. I'm not a great fan of the person who was declarer, so I was willing West to come out with that Q♦ :rolleyes: . But it didn't happen...

[Caitlynne]

I agree that the Queen of diamonds lead is pretty clearly indicated on this particular hand, even with a side 5 card suit, simply because the weakness of the hand makes it a better bet to try to hit partner's suit.

 

However, with a five (or six) card suit, there are many ways to win, so a club lead is by no means hopeless when headed by an honor (so it won't block if established). It is possible that the opponents cards in your 5 or 6 card suit are divided 2-2, for example.

 

Indeed, the attractiveness of the club lead would be considerably greater if the suit were a little better, e.g., Qxx, xxx, xx, QJxxx and the 4th best club would probably be my choice.

 

It should be mentioned that the diamond lead would not be so attractive without the Queen in that suit. That is, it is important that you have a honor to contribute in trying to hit and establish partner's hoped for suit. To this end, Kx and Qx are the most desirable "hit partner's suit" leads when leading against 3NT with a weak hand.

[manudude03]

I've always believed that when people are being taught, it is just simpler to teach 4th best. You will always have a suit with at least 4 cards in it, but you are not going to be guaranteed to have a 5 card suit, and so if playing 3/5*, you need to know what to lead from a 4 card suit, and then if partner doesn't know how to apply the rule of 10 and 12 properly, they may not be able to play the right card in 3rd seat. As a 3rd best lead is usually a lower card than a 2nd best lead, it is more likely there will be ambiguity regarding if it was a 3rd best lead from a 4 card suit or a 5th best lead from a longer suit.

 

*Assuming the normal 3/5 as opposed to the 3/low which is sometimes played.

[barmar]

I've always believed that when people are being taught, it is just simpler to teach 4th best. You will always have a suit with at least 4 cards in it, but you are not going to be guaranteed to have a 5 card suit

Although often there will be reasons why leading from your 4-card suit may not be best, like if the opponents bid the suit. So there's still some ambiguity.

[1eyedjack]

I am curious to know with what frequency all y'all consider the rule of 11 (or as the case may be the rule of 10 and 12) to be relevant to the defence,

 

I have a vague recollection of a time some decades ago when I first learned the game and was struck by the elegance of the rule of 11, and I would religiously go through the mechanical exercise of working it out on every hand possible, as part of the discipline of the overall plan.. But these days I cannot remember the last time that I actually applied it at the table. That could be because the potential for its adding value is infrequent (ie dummy wins trick 1, or dummy only has low cards etc). Or it could be because the right card to play is kind of obvious, without doing the maths (but that may just be an instinct honed by experience).

[Mbodell]

I also remember reading that Hampson (and Levine and Weinstein) are regarded as the types that will beat anyone in a KO but are not great at racking up big matchpoints against random fields so it comes down to what style of player you want to be.

 

I don't know about Levine, but I'll back Hampson, Weinstein, and Bobby Levin all putting up good match point scores in pretty much any field!

 

Robotic 4th best is obviously terrible. The double dummy best lead is frequently not 4th best, but frequently making more speculative leads also makes defending the hand more difficult to read. Some form of Smith echo can be very helpful then in my opinion.

 

There are also reasons to not lead 4th best in a suit you lead, like you are leading rho's suit and have 5 of them, often lead 5th since the spots might matter. Or you have AKxx, lead the attitude A/K, not 4th best.

[rhm]

I am curious to know with what frequency all y'all consider the rule of 11 (or as the case may be the rule of 10 and 12) to be relevant to the defence,

 

I have a vague recollection of a time some decades ago when I first learned the game and was struck by the elegance of the rule of 11, and I would religiously go through the mechanical exercise of working it out on every hand possible, as part of the discipline of the overall plan.. But these days I cannot remember the last time that I actually applied it at the table. That could be because the potential for its adding value is infrequent (ie dummy wins trick 1, or dummy only has low cards etc). Or it could be because the right card to play is kind of obvious, without doing the maths (but that may just be an instinct honed by experience).

This is the crux of the matter.

The times when fourth best really helps my defense are few and far between, while as declarer I find it often useful to know how the suit is likely to break.

That's why people sometimes depart from fourth best to fool declarer.

But fourth best seems to me a signal, which breaks even at best. I have given up zero or two higher honors (sometimes called coded leads) for the same reason. It seems to help declarer more than partner.

Not all signals are created equal in this respect. On balance some seem to benefit partner more and some declarer.

 

Attitude leads are an alternative against notrump contracts.

Attitude leads are clearly superior when you decide to lead passive. Choosing second best from a poor suit is not quite the same.

The way I play attitude leads I have a lot of leeway.

The card led is not only determined by the length and quality of the suit led, but by the whole hand and how much I hold in other suits.

If I have a second suit, to which a switch might be welcome, I might choose a different card than when not.

Declarer gets the same information, but it seems less useful to him.

Whether I lead high or low, he never knows how the suit is breaking, while partner is rarely in the dark. It seems to me declarer has not the same benefit from this lead than partner.

Of course there is a learning curve for this lead.

For example I am reluctant to waste the 9 or even 8 say from a "worthless" three card or longer holding and will often choose a lower card.

 

Rainer Herrmann