[ArcLight]

I've always wanted to use count signals. I think it would force me to pay more attention. But I almost never see them used below expert level. Perhaps one intermediate pard used them, maybe 10% of pick up "advanced" pards use them.

In BIL the emphasis is on things one can learn from books/CDs, such as conventions. I don't think there is enough emphasis on carding, counting, and other fundamentals. So in order to help generate potential partners for myself, I volunteered to help out in BIL. I came up with a number of hands (from various sources) that emphasized Count Signals, counting declarers tricks, figuring out distributions, and using the Rule of 11.

In BIL I took groups of 4, and had them play a few hands, privately quizzing them about what they had learned about the unseen hands ever couple of tricks.
In one hand, East had to duck their A, so as to kill dummies suit. In another hand, E must not duck; they had to take their tricks to set. The reason I used groups of 4 is I wanted the players to have to think hard, while playing, and not just watch others play and make comments like "notice West play high-low to signal count in Clubs". I think players learn a lot more by doing than watching.

All of this takes a lot of time. The dummy has to sit there with nothing to do for 10-15 minutes (as I privately chat with the players). With some hands, one of the defenders is weak, and contributes to the set by signaling. This means the other defender is the one that has to count and draw inferences.

Since only 4 players can play at a time, and over 33 people contacted me, it would take many sessions just to cover 4 hands. I have a lot more than 4 hands, and I don’t think a group is going to get much out of playing just 4 hands.

So I'm in a quandry.

1) I can continue offering these sessions the way they are, staying with just the same 4, and doing 4 sessions, then moving onto the next group. The problem is scheduling as they may not always be able to make it.

2) I can continue offering these sessions the way they are, rotating the groups, giving another 4 a chance. The downside is playing 4 hands (sitting out 1 as dummy) isn't really much practice. I think playing 12 hands would allow one at least a glimpse/taste of some counting decisions. Even if you miscount, at least you are now aware "Gee, I had the answer at my finger tips, next time I'll try and pay more attention".

3) have a mini-tournament where many players play some hands, and then we go over this. This will allow many hands to be covered quickly.
However I hate this for 2 reasons.
a - you can get this from books
b - By a mentor asking leading questions, you get into the habit of thinking. If you KNOW you are soon going to be asked how many spades are left, you will pay attention. You won’t play in automatic pilot. This wont be possible in a tournament/review session.


There is also a big disparity in the level of players. I had intended these to be geared towards Intermediate Plus players. No difficult techniques. There is no bidding; I tell you what to bid. You just play and count.

Any suggestions on how to structure mentoring sessions so as to make maximal use of time?
I want the players to get something out of this, but I think each one will need to play many boards to benefit. With so many interested, I can’t imagine being able to offer 30+ sessions in a timely manner.
How can I do this?

[Chamaco]

You can't perform in-detail lessons and reconcile them with large audience.

The task you are trying to realize is great , but I suspect it can work in a feasible manner for smaller groups only...

[Elianna]

you could have a list of questions in advance, and have several other people to act as mentors (thus duplicating yourself).

I don't know how useful this advice is, as I don't really know where you're going to find these volunteers. I know that I wouldn't want to.

I also don't really understand this emphasis on showing count signals. Yes, it's really important to count out and visualize a hand, but count signals aren't the only way to do that. You get a lot of inferences from the bidding (and lack of bidding), the opening lead and attitude towards that, any returns by the defending side, and from what declarer tries to do in the hand.

Adam and I have a somewhat experienced partnership, and we don't place much emphasis on count signals per se. Our primary signal is attitude, and then many following signals are suit preference, because we can usually work out approximate count on the hand.

This is not to say that we don't signal count, it's just that it has a low priority.

Anyway, sorry for the digression in your thread, but I think that it might be useful to have a discussion on what a useful primary signal is (ie, what helps partner figure out what to do most?).

[paulg]

Chamaco, on Jan 4 2006, 02:35 PM, said:

You can't perform in-detail lessons and reconcile them with  large audience.

The task you are trying to realize is great , but I suspect it can work in a feasible manner for smaller groups only...

I agree and I believe that this is true for any subject not just carding. I believe you have to chose whether to TEACH a large number or MENTOR a small number - techniques are different.

Carding makes the task even more difficult as it is a subject where beginner and intermediates struggle to improve as it seems less interesting than bidding and declaring.

Paul

PS At one of my BIL sessions I forced everyone to use reverse carding as a method of getting them to watch and count the cards. It was a partial success - after a couple of hands most tended to stop watching and always play their lowest card; but those who stuck with it did improve.

[ArcLight]

>>I also don't really understand this emphasis on showing count signals.

It gets the players in the habit of counting! It helps to focus them. In the group email I told them that experts don't always signal count (they usually do so in hold up situations) but if they can start counting it will help get them into good habits.

> Yes, it's really important to count out and visualize a hand, but count signals aren't the only way to do that.

I never said they were. There are other inferences I ask the players about. Such as what they can gather from the Rule of 11. Who has the Ace of teh suit just lead. Etc. But Count will help get the players used to paying closer attention.


>Adam and I have a somewhat experienced partnership, and we don't place much emphasis on count signals per se. Our primary signal is attitude, and then many following signals are suit preference, because we can usually work out approximate count on the hand.

A while back I asked about Count signals. Fred wrote that expert declarers will also pay attention thus you dont always want to signal count. Bobby Wolff has a bridge column and I asked him and he pretty much said the same thing. However, the level of play in BIL is Beginner - Intermediate. Let them get started counting, then as their skill improves, and they establish strong partnerships they can revise their signaling.


>Anyway, sorry for the digression in your thread, but I think that it might be useful to have a discussion on what a useful primary signal is (ie, what helps partner figure out what to do most?).

I covered that at the start. Its not like I said "The only useful signal is COunt, use nothing else, every possible card played is a count signal, do not use judgement".
I explained Attitude and Count. I told them not to use suit preference signals, other than for ruffs, and deliberately selected hands that involve counting, usually hold up situations.


> However I think for beginners/intermediates, the signals get them in a good habit of counting and makes it easy to do. Counting the hand without getting any count signals takes a lot of practice and work. I'm all for teaching intermediates "all count all the time" because it will be very useful in their evolution as a player.

Exactly.


Here is a hand I used -
From the bidding, opps have 25-27 HCP ending in 3NT, you hold 13, meaning pard has how many HCP? 0-2, good.
Now pard lead the X, using the Rule of 11, how many cards do they have?
5? Good.

You hold the AJx, how do you play?

After all this, and getting the defender to think and answer correctly, they still misplayed. They played A, then J, then small, declarer held up, and made the contract.


Nothing earth shattering, all it takes is a bit of thinking. Now NEXT time that defender is in that situation, they will hopefully do better.

We had some hold up situations, and a case where counting told them if they could cash a Q on the third round.


At the end of the sessions I will suggest they buy Mike Lawrences DEFENSE CD.
(As well as all his other software. BM too)

[peaceman]

To count signal or not to count signal...good question...

I hav played for many years with a variety of pards, from intermediate to known expert names...generally short term casual play & lots of online play, some tournaments.

Among experienced players, ther r basically three types -- the attitude people, the compulsiv counters & the "signal what is important when u need it" people.

The most advanced defenders tend to be compulsiv count- oriented people. These r the types that can narrow down the possible hands declarer & pard has to two or three possibilities after one or two changes of suit...maybe at most 5 tricks, on some hands.

It is a persons visualizing power, not their experience in the game that determines if they r this sort of defender.

If u r this sort of defender, u will be very very reluctant to play anything but count signals. If yr pard is this sort of player, pard will benefit greatly from playing count signals & u guys will defend well, even if u occasionally do stupid stuff because u didn't "get" what was going on with the hand fast enough.

Yes, of course even compulisve count types recognize ther r times to false card, but they generally hav a good enough view of the hand to be able to factor this in -- especially when yr false signal fools declarer the way it was intended to.

But if pard is well abov yr level, usually best to give honest count except in situaltions when u hav either very few or a whole lot of the defensive values.

Many want to hav their cake and eat it to, so will play attitude for the first couple of tricks, then switch to count. But these types really belong in the third category listed above. The advantage the serious counters gets goes away if count signals dont start immediately.

Of course the vast majority of bridge players never do aquire this sort of visualizing skill (it tends to show up early, & especially if the game is learned quite young). If u r a compentent average player, attitude signals will probably get best results for u & pard.

Experienced players with average visualization skills will always count out a hand & not, say, pitch a good card toward the end on a pseudosqueeze, because by that point they know what declarers remaining cards are.

Even average players can count up to thirteen as they go, but that is not the skill i am referring to at all.

In short, if pard insists on count signals, play them & be grateful.

Count signals should of course be tried out by intermediate level partnerships, just to see if u can "trigger" the sort of advanced visualization i am talking about.

[Free]

I've been playing mixed signals for a while, and our defense has improved significantly! We try to show which suit to play first, but that's not always possible. Next we show count. This way we get visualisation of distribution and possible honour cards, and we try to decide what to do with lots of information.

Imo suit preference is still more important in the first stage of a game where you might need tempo, but count only needs to come in after a while for the endplaying. Imo it's quite silly to keep showing your suit preference. If you showed ♣ once, don't show it again (unless for ruffs or something similar).

[Badmonster]

But how do you learn to signal well? To judge which signal is needed? To know how to signal it? To know what it is p is signalling?

[han]

I agree with what Elianna and Justin say here. I don't think that you need to teach count signals to get your students to focus on counting the hands, but it can be useful. I prefer not to give a lot of count when playing but count signals can be very useful and all good players know how to give count and when it is absolutely necessary to give count. I think that giving frequent count signals is a good way to get comfortable with the signal.


BTW, I see count signals used below the expert level all the time. I also see a lot of self proclaimed experts on BBO who don't know what a count signal is but that is a different matter.

But how do you learn to signal well? To judge which signal is needed? To know how to signal it? To know what it is p is signalling?

This goes hand in hand with learning to defend (and declare) well. If you know what's going on at the table then you know what signal partner needs. When I'm playing with a really good player and I have no idea what to shift to then I assume that my partner has just told me. Similarly, if I need to know whether to hold up or which suit to hold on to then I assume that my partner has just given me count.

When I'm sharp and focussed then I need fewer signals because I will more often find the necessary clues from what the opponents do. However, no matter how good you get there are always hands when you really need partner's signal.

[keylime]

I normally use count in very specific situations.

I also admit I use a modified Obvious Shift method at trick one as well. I think suit preference based carding might be a touch more superior than count carding.

[Free]

Badmonster, on Jan 22 2006, 09:33 PM, said:

But how do you learn to signal well? To judge which signal is needed? To know how to signal it? To know what it is p is signalling?

Imo, before learning to defend well you need to declare well. You'll recognize what declarer is doing, or will do, so you'll know how to defend better than if you're just looking at some signals. That's why count is imo very useful in a later stage of the game to visualize the hands, where suit preference is still the number one at the beginning for me because then you need tempo.

How you do it exactly depends on the hand. Best is to make pretty good agreements on this. For example, if you play trumps and there's a singleton in the lead suit, just show suit preference if it's possible (when you can't cover). It doesn't matter much if use 'obvious switch' or lavinthal or whatever, as long as your partner understands the signal. Once partner knows what suit we want, you can just show count (normal or upside-down or something else) and look what partner plays.

[bearmum]

ArcLight, on Jan 4 2006, 11:42 PM, said:

I've always wanted to use count signals.  I think it would force me to pay more attention.  But I almost never see them used below expert level.  Perhaps one intermediate pard used them, maybe 10% of pick up "advanced" pards use them.

(Edited quote)

Interesting comment re count signals -- my regular pertner (husband ) and I have used them for about 30 years -- since we first switched to "Shenken Club" system. We now play Precision but still employ count signals and find them useful
. BTW I would consider us "Intermediate + " or maybe "Advanced -" players

And with one partner I used to play online (when I lived in the US) played attitude when following to partner's lead and count when following to opps lead -- and many times we got the vital extra trick in defence

[zasanya]

The following comments although regarding a different game and in a defferent context seem appropriate here.
When former world chess champion Emannuel Laskar was asked to comment about the evolution of a chess player He said," A youngster plays chess because he loves to play and if he has raw talent and analytical ability he wins but then he comes across better players and he discovers that there is something called chess theory.If he is sufficiently interested he learns the theory, applies it and becomes a chess expert.Then he comes across still better players and discovers that to beat them one must not only play according to theory but on occasions play seemingly antitheoretical moves.Thus he becomes a chess master."
When former world chess champion Anatoly Karpov was asked whether he always played according to theory he replied," It is true that I know basic chess theory but when I play I am not worried about theory.I find moves to increase my advantage and build a winning position.The theoreticians study my games and make theory out of them."
Coming to bridge and signals, count or others, a beginner relies on instinct and reasoning without signals ,the intermediate/ advanced use signals,the experts knows when NOT to use them and world champions contribute and enrich theory by their practical play.

[peaceman]

Jlall, on Jan 22 2006, 01:21 AM, said:

Could not disagree more with the above post.

I would say that most top experts are in the "signal what partner needs" category, and that is rarely count.

A little clarification:

I was not intending to imply that people who do not play count signals can't be experts, altho perhaps i did accidentally do so.

I am uncertain as to what the exact proportion of count-preference very good players to attitude preference & mixed preference vgps ther are out ther -- i hav certainly seen examples of all.

I would grant that for a lot of players --even those who win tournies regularly -- pure count (esp. if it always "tells the truth") is probably a less than optimal method.

I was trying to make the claim that for a CERTAIN type of high-visualizing mind, count is by far the best option, and that if u don't hav this sort of mind, count is probably worth learning, but not worth using unless pard insists. Having a scarily good visualizing mind is by no means sufficient for consistent good results - but i always like it in my pards ! Getting the information u need earlier than someone else in yr seat wld pick up on it is a key sign of a quality player, but the information that people look for, & therefor the best tools to aquire it, definitely vary according to the type of mind u hav. Many expert players wld rather play with a weaker pard than another expert who has an incompatable style, method or temperament. Also, ther is still plenty of room for innovation in signalling systems. Granovetter, for example, some years ago came up with a method that varies according to cards in dummy that seems quite reasonable. I would be interested to learn if anyone has tried it & found it good or bad.

If YOU don't hav the sort of count - oriented mind i was talking about, experts will happily switch their signalling methods to what suits you best (esp. if u r paying them to play!). Everybody tends to bend over backwards to make it easier on the weaker pard don't they? (not intending to imply any particular poster here is not an expert, nor that i am.)

I like the analogy made from the world of chess, and agree that as players move up through the various levels of the game their strategies & tactics mutate.

Creativity (knowing when to break the rules --whatever u perceive the rules to be) is a key element at the top level of any human activity, i believe.

[mikeh]

Hi Peaceman

I think you are very wrong.

Those who possess great visualization are the very players who probably play count the least

Visualization comes in many ways, but I like to think of the main technique as being a form of reverse engineering of the bidding and the play.

An improving player learns to draw inferences from the bidding. An advanced player may draw inferences from the cards played.

The good player draws inferences from the calls not made during the auction, as well as those actually made, and from the cards not led or played as well as those he actually sees. Breaks in tempo, either in the auction or the play, go into the mix, as well as plays in tempo: altho against really good players, the card play is often at the pace of a metronome.

All of this is independent of any actual signalling method that the defenders utilize. Good defenders will add to this sea of information when they feel, based on experience and partnership style, that signals will convey information likely to be of more assistance to partner than to declarer.

Now, a beginner cannot be expected to engage in this reverse engineering. Beginners and intermediates are still learning what their own bids are supposed to mean, and have enough trouble remembering what cards have been played without worrying about the cards or bids not played or made. Further, they are likely playing with and against comparable players, and strangely, inexperienced players are the ones against whom reverse engineering works the least well.

If one is missing the AKQ of a suit and opening leader leads another suit, it is reasonable to infer that LHO lacks the AKQ, almost certainly (depending on the auction) lacks the AK or the KQ. However, I have seen poor players lead a side suit against 6N when holding an AK combination (wel,, I only saw it once).

So teaching count is an excellent idea since it is a relatively easy introduction to visualization: sort of like training wheels on a bicycle. Once one learns to ride with ease, one reduces the use of the training wheel.

(okay, a bad analogy, since we still use count occasionally, and we never put the training wheels back on... but it is late and I am still stuck at the office )

[ArcLight]

>So teaching count is an excellent idea since it is a relatively easy introduction to visualization: sort of like training wheels on a bicycle. Once one learns to ride with ease, one reduces the use of the training wheel.


Wow, I actually got 1 positive response!


In BIL what I have been doing is asking the kibitzers to only look at one hand (plus dummy obviously) and try and figure out whats going on and what to do. The hands I've selected have generally involved count signals. I ask the players questions about the unseen hands based on what we can infer about the signaled suit, and from the bidding. The players respond privately.

Ex: If declarer opened 1 Spade (SAYC) they should have 5+ spades, They ruffed the 2nd round of hearts, and pard is signaling an even number of Clubs, how many diamonds does declarer have? What is declarer and pards distribution?

Ex: Using 4th best leads, pard leads the 5, she may have 5 cards, we win the ace and return the 8 from A 8 4 showing pard we started with 3.

The people in the session (remember this is Beginner Intermediate) have generally not seen this before or been asked to think along these lines.


Instead they have been receiving lectures on Transfer Lebensohl and lots of other conventions they will surely misuse, especially with pick up pards.

[yoder]

I strongly believe that count is at the heart of visualization, and agree with previous posters that learning count signals can trigger visualization in a way that nothing else does. I know, because it happened exactly that way with me, many years ago.

I know a bit about cognitive style and it's very true that people differ greatly in how they learn and how they process information. I happen to be a fairly visual person. As a beginner, I knew how to count suits, cards played, etc, but I played with other beginners who had, like me, been taught to give attitude signals.

Then one day long ago when I was still pretty novice, a very excellent player condescended to play a session with me. He asked that we play nothing but count signals, something that I had not done before.

That one session of bridge opened up the entire world of bridge to a new level for me. Suddenly I could visualize what the declarer had, I could figure out what card to save at the end, I could actually COUNT out the hand!! It was an epiphany for me in bridge.

To this day, if I can find a partner who likes to play mostly count, I treasure him/her. Count enables visualization, and as I have also learned to make inferences from play of suits, etc, I find that I can figure out where all the high cards are very easily from the play, so count is by far the most valuable thing that partner can tell me.

If you have not played count signals, find a partner who likes them and do it. You may be amazed what windows it will open into the hands of the opponents.

[kenberg]

I may be reading more into Yoder's comments than s/he intends, but I recommend teaching that count/attitude signals be given in accordance with the pretty much standard approach of attitude when we crack a suit, count when declarer is playing on a suit, skipping at least for the moment the many exceptional cases. I recently led a queen against a suit contract, partner produced a somewhat high spot, and when I was in again I led the suit again only to find that partner for some reason thought I needed to know at trick 1 whether he had an even or odd number of cards in my suit. This was not a successful defense. Players can make a choice to do this if they think it right, but don't expect partner to understand this approach if it has not been discussed. Quite possibly Yoder is suggesting no such thing.

But mostly I have a thought about the teaching format. You mention you have a large number of folks out there and it's tough to organize the comments. How about this?


First do the teaching as you have been doing it. Then have them play some similar hands where the sequence of events is:

They play, without any comment.

Afterwards, at each table, the four players discuss the hands asking themselves such questions as
If everyone played well, would the hand have made or gone set? Who, if anyone, should have done something different? How could they have known to do so? They should try to distinguish between reasonable choices they should have got right and things like underleading an AK to partner's Queen. (Yes, that is sometimes reasonable, but mostly it's not.)


After giving them a little time for discussion at the table among the four players, ask for conclusions to be shared among all the players. You may need to schedule comments from the tables to prevent everyone from talking at once.
Perhaps they will all agree on the proper line for declarer and defenders, perhaps not. Then you could come in with comments of your own.

This may or may not appeal to you and may or may not work, but in theory it seems good.

Ken