[ASkolnick]

Everyone keeps talking about LTC. But even in LTC, there is an adjustment which says if your hand is all Aces, deduct one loser, If queens, add 1 loser. So, this is really an 8 loser hand, not a 9 loser hand. As ffor slam being off 2, without partner's hand, the fact that you held a "minimum" limit raise doesn't mean you should be in slam. If he/she needed more info, you can always ask.

As for LTC(with adjustments for Q's and A's) and Zar Count, I find they tend to be close enough that most of the time they tend to the same answer. But either of these methods are much better calculator's then HCP alone.

[effervesce]

NickRW, on Nov 3 2008, 10:39 AM, said:

effervesce, on Nov 3 2008, 04:54 AM, said:

3♥. Aces are undervalued,  and you have 4 card support. The hand has an amazing 24 ZAR-2♥ may work out, but is too pessimistic for me.

23 Zars - unless you're relying on opener having 5♥s and counting a bonus for an extra trump - but Acol players sometimes open minimum 4=4=4=1 shapes with 1♥ - unlikely of course, but possible. But, whatever, even at 23 Zars it is definitely pushing into the 3♥ response range.

The thing that makes me feel slightly conservative is that hands with relatively high loser counts like this one are usually better off having the aces in the side suits and the quacks in partner's suit(s) where we know they will be useful. I'd be a whole lot happier with

♠Axxx
♥Jxxx
♦xx
♣Axx

Alternatively I'd also be happier with a 3♥ call if we were vul at imps where pushing for the game pays real dividends.

I suppose this would be a good one to run a sim on, but I can't be bothered just now.

Nick

You get +1 ZAR for each honor in partner's suits up to a max bonus of 2 per suit. Thus it is 24 ZAR.

[thebiker]

P_Marlowe, on Nov 1 2008, 02:37 PM, said:

Playing Acol, I would bid 1S.

The bid buys me time, maybe after partner makes
another bid, I can make a more informed decision.

If partner opened 1S, I would go with 2S, the hand
is not worth a limit raise.

With kind regards
Marlowe

I cannot see how responding 1S helps when playing an Acol style system

If partner rebids a minor you still have the same problem

Partner rebids 1NT, rasing hearts only shows 3 hearts , and jumping directly to game now historically shows a good hand, a delayed game raise.
Ditto if pard rebids 2NT

Acol works best by making limit bids, if you don't raise hearts immediately then you deny four hearts or end up promising a big hand that you havent got


regards

Brian Keable
alias thebiker

[NickRW]

effervesce, on Nov 4 2008, 01:03 PM, said:

You get +1 ZAR for each honor in partner's suits up to a max bonus of 2 per suit. Thus it is 24 ZAR.

You may be right - I vaguely recall something to that effect somewhere in his writings - but, imo, while Petkov's work is in principle very sound, he got a bit carried away with bonuses. In particular counting a bonus for an ace of partner's suit, which is already carrying over half an extra point by Milton Work standards, is overkill. In comparison, minor honours in partner's suit, which are carrying less points by Milton Work standards, are definitely worth - well - at least regarding them as an asset and upgrading in an otherwise close decision.

Perhaps I'd better do the sim - perhaps instead of a bunch of us quibbling about whether this hand is worth 2 or 3♥ we should look at some actual results. I'll do one - later.

Nick

[NickRW]

Well, I did a sim. What I got from looking at cases where opener is not strong enough to go on over a 2♥ call, but *might* be strong enough to bid 4 after a 3♥ call was this:

About 65% of the time opener doesn't have a 4♥ call even if you bid 3. In these cases you go down in 3♥ about 17.5% of the time and will be a MP loser against a field that is simply calling 2♥ on the hand in the OP. The rest of the time 9 or more tricks are available and you draw with the field.

About 35% of the time opener does have enough to call 4 over 3. You win 58.5% of these cases as game does in fact make - but obviously you lose in the remaining 41.5% of the cases when the breaks don't happen or the hands don't fit as well as hoped for.

So (.35*.585) - (.65*.175) - (.35*.415) = -.05425 = you lose match points by bidding 3♥.

Obviously these results are extremely sensitive to exactly what you think is good enough for an opener in the first place and so on - so your mileage might vary a bit.

Also these results are close enough to make one think this is really a 50/50 guess non vul at IMPs and it is probably a winner to bid 3♥ vul at IMPs - though I didn't actually do the IMP calculation.

Further, my assertion earlier in this thread about the hand being better if the ♥J and ♣A swapped suits is without any significant grounds. I think I stand by my assertion about quacks in side suits versus quacks in partner's suit in the general case - but in this specific case the ♥J is worth no more than the ♣J - probably because in most cases opener has 5+ hearts and the minor honour in hearts is not as much help as it would have been if most fits were only 8 cards.

Nick

[NickRW]

thebiker, on Nov 4 2008, 01:55 PM, said:

P_Marlowe, on Nov 1 2008, 02:37 PM, said:

Playing Acol, I would bid 1S.

The bid buys me time, maybe after partner makes
another bid, I can make a more informed decision.

If partner opened 1S, I would go with 2S, the hand
is not worth a limit raise.

With kind regards
Marlowe

I cannot see how responding 1S helps when playing an Acol style system

If partner rebids a minor you still have the same problem

Partner rebids 1NT, rasing hearts only shows 3 hearts , and jumping directly to game now historically shows a good hand, a delayed game raise.
Ditto if pard rebids 2NT

Acol works best by making limit bids, if you don't raise hearts immediately then you deny four hearts or end up promising a big hand that you havent got


regards

Brian Keable
alias thebiker

There is, arguably, some point in bidding 1♠ first. If partner comes back at you with 2♣ you have a useful ♣J and 3 cards total for him in his 2nd suit. If partner comes back at you with 2♦ then you have probably a wasted ♣J and partner's diamonds not working as well as you might have hoped for opposite your xx in that suit. This might sway you one way or the other in making your decision. Of course, as you point out, opener will tend to take you for less than 4♥s on this sort of sequence whichever decision you make and this may not be helpful to him/her.

Also, if you're playing a weak NT as most MP Acol players are and partner comes back at you with 1NT showing strong NT values, then you have a much clearer shot at a higher bid than simply 2♥.

Alternatively it is just possible that opener will raise your suit - now you have double fit and calling 3♥ looks more attractive.

Of course, set against this is the fact that you have given away additional info about your hand and partner's to the opening leader by messing about. This is difficult to quantify - but Acol's 'bid what you think you can make' strategy certainly has its good points on a lot of hands. But just because it is a limit system doesn't preclude intelligent use of other possible calls sometimes.

Nick

[han]

NickRW, on Nov 5 2008, 10:52 PM, said:

Well, I did a sim. What I got from looking at cases where opener is not strong enough to go on over a 2♥ call, but *might* be strong enough to bid 4 after a 3♥ call was this:

About 65% of the time opener doesn't have a 4♥ call even if you bid 3. In these cases you go down in 3♥ about 17.5% of the time and will be a MP loser against a field that is simply calling 2♥ on the hand in the OP. The rest of the time 9 or more tricks are available and you draw with the field.

About 35% of the time opener does have enough to call 4 over 3. You win 58.5% of these cases as game does in fact make - but obviously you lose in the remaining 41.5% of the cases when the breaks don't happen or the hands don't fit as well as hoped for.

So (.35*.585) - (.65*.175) - (.35*.415) = -.05425 = you lose match points by bidding 3♥.

Obviously these results are extremely sensitive to exactly what you think is good enough for an opener in the first place and so on - so your mileage might vary a bit.

Also these results are close enough to make one think this is really a 50/50 guess non vul at IMPs and it is probably a winner to bid 3♥ vul at IMPs - though I didn't actually do the IMP calculation.

Further, my assertion earlier in this thread about the hand being better if the ♥J and ♣A swapped suits is without any significant grounds. I think I stand by my assertion about quacks in side suits versus quacks in partner's suit in the general case - but in this specific case the ♥J is worth no more than the ♣J - probably because in most cases opener has 5+ hearts and the minor honour in hearts is not as much help as it would have been if most fits were only 8 cards.

Nick

Please explain what restrictions on the hand you used and how you determined whether opener would bid. I'm particularly interested in what restrictions you used for RHO (who passed).

II'm very suspicious of that 65%, in my experience after the auction proceeds 1H-3H opener bids 4H much more often than she passes.

Also, was the play double dummy?

[NickRW]

Han, you ask complicated questions. [no suitable smiley icon scratching its head]

The results were double dummy.

As most posters who recommend the 3♥ call were doing so on the basis that aces are severely undervalued for the purposes of suit play, the principle criterion for opening, but not being a hand good enough for a trial bid over a 2♥ call was:

HCP + controls + "freakness" >= 18 and <= 22

HCP + controls is the same thing as the 6/4/2/1 count for high cards the same as the Zar point recommendation. Though not many people in general use the 6/4/2/1 count, given the emphasis placed on aces by posters in this thread in favour of 3♥, I felt this was a reasonable assumption in this case.

Freakness = 1 for length + shortages counted 1/3/5. This is not quite the same as 2a+b-d (Zar points) but is arguably better for use in conjunction with a 6/4/2/1 count. (But I'm not justifying that assertion here).

Other restrictions were:

0) At least 4 hearts (of course!)

1) Not 12-14(hcp) bal, but could be 5332 with 5 hearts (many Acol players don't include these in their weak NT opening unless the hearts are especially weak)

2) Not 4=4=1=4 shape (most Acol players open this 1♣, certainly in the 12-14 range anyway)

3) Spades not the same length (or longer) as hearts unless 4-4 (open the higher ranking when 5/5 or 6/6)

4) Hearts at least as long as either minor. (Some players tend to open 4 card suits up the line - so this would not be univerally agreed - but quite a lot open the major by preference).

Unfortunately the dealer program I use doesn't make allowances for singleton honours - so that aspect wasn't quite as I would have liked it.

Specifically, for those hands that I designated as good enough to to raise 3♥ to 4, the hcp + controls + freakness was 21 or 22. I got the 65/35 ratio from running a 1000 deals - er it might have been 10,000 (dont recall just now) with this restriction set to 18 to 22, and then rerunning it with it set to 21 or 22 and comparing the pecentages of deals that passed the tests.

The 21 to 22 range was picked on the basis that I personally would raise to 3♥ using this method of valuation with 15-17 and would want to be in game with 36 total.

The other percentages I got were from 2 batches of 200 deals analysed double dummy - one with hcp+controls+freakness = 18/19/20 and one with it = 21/22. A larger sample would be better - but 200 is time consuming enough anyway.

I didn't restrict the defending hands - oops - though I am not sure this makes a huge difference to the relative percentages involved of tricks expectations for our side - I'm certainly not going to rerun the analysis - it was quite time consuming.

Hope this answers your questions.

Nick

[han]

So what you showed was that 18-20 Nickpoints is more likely than 21-22 Nickpoints by a factor 65-35. That sounds about right, but isn't it possible that partner has 23 or more Nickpoints? And wouldn't partner also bid game with those hands?

Therefore, shouldn't you determine the number of hands with 18+ Nickpoints and compare it to the number of hands with 21+ Nickpoints?

Apologies if I misunderstood your post. If I did then I would appreciate it if you could rerun the simulation.

[NickRW]

han, on Nov 6 2008, 07:15 PM, said:

So what you showed was that 18-20 Nickpoints is more likely than 21-22 Nickpoints by a factor 65-35. That sounds about right, but isn't it possible that partner has 23 or more Nickpoints? And wouldn't partner also bid game with those hands?

Well - re "Nick" points - this is not just my idea. The "freakness" parameter available in Richard Pavlicek's dealer program is almost exactly the same as recommended (for the distributional count in conjunction with 6/4/2/1) in these forums by a poster who used to post here a lot around the same time as Zar Petkov was pushing his Zar points. Look up Tysen's posts from several years ago (his login might have been tysen2k - not sure - or maybe that was his handle on rgb).

But, whatever, you're right, a hand could hold 23+ "Nickpoints". However these are getting into the sort of zone that folks would be considering another bid over a 2♥ response - therefore responder might well get another chance to speak anyhow if opener has a hand as strong as that. I therefore concluded that those cases were of no real interest to the analysis and promptly excluded them.

Quote

Therefore, shouldn't you determine the number of hands with 18+ Nickpoints and compare it to the number of hands with 21+ Nickpoints?

Apologies if I misunderstood your post. If I did then I would appreciate it if you could rerun the simulation.

Well, given that I excluded the 23+ cases for sensible reasons, that is what I did, didn't I?

Nick

(As justification for my selection of 23 as the cut off point, 5 points in this system is worth about a trick (just as 3 [or so] is roughly a trick on the 4321 system) - therefore a 23 opener is about a trick better than minimum - therefore opener should be safe venturing a trial bid over 2♥ with 23+).

[han]

I see, I didn't read carefully enough, you explained it in your first post.

I'm not sure if the 18, 21 and 23 numbers are reasonable or not.

[Gerben42]

Since Acol is a system where you can open quite light, probably 2♥ is best.

[NickRW]

han, on Nov 6 2008, 09:03 PM, said:

I'm not sure if the 18, 21 and 23 numbers are reasonable or not.

The numbers are pretty reasonable. Petkov (with his 6/4/2/1 and 2a+b-d) recommends opening on 26 and being in game with 52 total. But 2a+b-d counts all hands as a mimumun of 8 (a is never less than 4) - so this is equivalent to opening on 18 and being in game with 36 the way I did it.

I play this way in real life - it does work OK.

Nick

[han]

I was around when Petkov was posting here and I didn't think that all of his recommendations were reasonable. I also thought that there was something wrong with his whole approach to the game.

The way I play limit raises, I would bid game over a limit raise with more than half the hands that would pass after 1H-2H. Therefore the numbers you chose are not reasonable for me.

But then, I wouldn't make a limit raise with Axxx Axxx Jxx xx.

[NickRW]

han, on Nov 7 2008, 04:10 PM, said:

I was around when Petkov was posting here and I didn't think that all of his recommendations were reasonable. I also thought that there was something wrong with his whole approach to the game.

The way I play limit raises, I would bid game over a limit raise with more than half the hands that would pass after 1H-2H. Therefore the numbers you chose are not reasonable for me.

But then, I wouldn't make a limit raise with Axxx Axxx Jxx xx.

Well Han, while I probably find quite a bit more agreement with Petkov than you do, I don't buy all of his recommendations either. Also, utilising his stuff would require some to consider their system choices - a price too high in some cases probably. (As using it requires that you can open [or preempt possibly, depending on exact system choices] with some hands that many would consider to be too "light" - this doesn't suit the 2/1 [or even SA] opening philosophy - it does suit Acol and some other systems)

Note that the hand in the OP, counting it the Petkov way exactly, is a 3♥ raise, but counting it the way I used in the sim I did (similar to Petkov but not quite the same) isn't - and the sim proves that it isn't - at least not for MP scoring anyway.

So you're right

Nick

[el tombo]

I bid 3H. In acol the opener will either have an unbalanced hand with 5 hearts or 15+ points. With 15 or 16 points balanced he would pass 2H (although game might not be great if he had 15 points and a balanced hand).

I play Acol quite a lot and in the Acol club, Andrew Robson once conducted a lesson. He said "you must be agressive with limit raises in Acol because it is too easy to miss game if you have borderline hands and choose 2H". If you post your partner's hand we might be able to see if he was right in going to slam (It might be already posted and I missed it)

[mikestar]

For me, this hand is worth its nominal 10 points including distribution. While the aces are a major plus, the isolated jack and the total lack of body in the long suits are sufficiently negative to cancel this.

Whether or not this is a limit raise depends on how light your partnership opens and how aggressively you accept limit raises.

Another point--conservatism has an edge playing 2/1 because partner might go slamming. The limit raise is safer in a big club system, where a hand strong enough to go slamming after a limit raise will start with 1♣.

[1eyedjack]

mikestar, on Nov 13 2008, 08:21 AM, said:

Another point--conservatism has an edge playing 2/1 because partner might go slamming.

Could you expand on that explanation, please?

[mr1303]

el tombo, on Nov 8 2008, 11:52 AM, said:

I bid 3H. In acol the opener will either have an unbalanced hand with 5 hearts or 15+ points. With 15 or 16 points balanced he would pass 2H (although game might not be great if he had 15 points and a balanced hand).

I play Acol quite a lot and in the Acol club, Andrew Robson once conducted a lesson. He said "you must be agressive with limit raises in Acol because it is too easy to miss game if you have borderline hands and choose 2H". If you post your partner's hand we might be able to see if he was right in going to slam (It might be already posted and I missed it)

No, I'd already said that I can't remember the actual hand (and really can't remember it now, several weeks later).

It was something like a reasonable 17 count or so, but she went 2 off in 6 and blamed me for raising [to the 3 level] on a 9 loser hand.

[mikestar]

1eyedjack, on Nov 13 2008, 08:39 AM, said:

mikestar, on Nov 13 2008, 08:21 AM, said:

Another point--conservatism has an edge playing 2/1 because partner might go slamming.

Could you expand on that explanation, please?

Glad to. The #1 thing that can go wrong by making the limit raise is that partner bids 4♥ off one. The #2 thing is that partner passes and 3♥ is down one. Neither of these contingencies are affected by the system choice.

However, two other possible sources of loss are #3 partner Blackwoods and bids slam finding out you have two aces and goes off one or two, and #4 partner makes a slam try and doesn't respect your 4♥ signoff, getting to 5♥ down one.

Now neither of these are at all likely to occur in a big club system--there are no hands strong enough to go past game opposite a limit raise which don't have the 16/17 HCP or extreme distribution needed to open 1♣.

This being the case, in 2/1 you have 4 ways to lose, but in Precision or other big club systems you have only 2, as the slam related cases don't happen. The one way to win is the same in any system--getting to a making 4♥ that you won't reach after a single raise.

The single raise is probably the better choice by a slight margin even in the big club case, but your odds are better than in the 2/1 case.