[Vampyr]

I think that by the time you have got to the Rule of 22 you have a clear opener (unless you have a super ugly hand with you honours in the short suits) and don't need to break the tie.

[Zelandakh]

Liversidge, on 2017-February-07, 11:18, said:

Your second example raises a related query. Why do Kxx Kxx count as just 1 PT but 4 LTs?

This is to do with the different methodologies. PTs fell out from the days of HTs and use those as a base. Kx(x) is 1/2 a trick there, as you lead to it and either the ace is to your left or your right. The LTC uses a king as the basic unit for a trick (either A = K = Q = 1 for OLTC or A = 1.5, K = 1, Q = 0.5 for MLTC). Similar thoughts also apply to holdings like KQx (1 PT, 1 loser (2 winners) in LTC, 1.5 losers (1.5 winners) in MLTC, although here the 1 PT evaluation is probably too low in most cases). If you use "real losers" rather than any LTC variant for holdings like Kxx - that is 2.5 losers (0.5 PT) - while keeping KQx at 1.5 losers (1.5 PT), you are probably going to end up with a better overall evaluation, although I have never seen any sources suggesting this kind of modification.

[Liversidge]

Zelandakh, on 2017-February-07, 13:23, said:

This is to do with the different methodologies. PTs fell out from the days of HTs and use those as a base. Kx(x) is 1/2 a trick there, as you lead to it and either the ace is to your left or your right. The LTC uses a king as the basic unit for a trick (either A = K = Q = 1 for OLTC or A = 1.5, K = 1, Q = 0.5 for MLTC). Similar thoughts also apply to holdings like KQx (1 PT, 1 loser (2 winners) in LTC, 1.5 losers (1.5 winners) in MLTC, although here the 1 PT evaluation is probably too low in most cases). If you use "real losers" rather than any LTC variant for holdings like Kxx - that is 2.5 losers (0.5 PT) - while keeping KQx at 1.5 losers (1.5 PT), you are probably going to end up with a better overall evaluation, although I have never seen any sources suggesting this kind of modification.

Very helpful, thanks.

[Cyberyeti]

Vampyr, on 2017-February-06, 02:15, said:

You could use the EBU's rule and come up with five.

I held almost that hand yesterday; ♠AKQ1093 ♥K82 ♦AK74 ♣ -. I think that when it is marginal whether to open 2♣ (assuming that is your strong opening), counting losers can be helpful. This hand doesn't really look like a 2♣ opening, but then when you notice that it is a 3-loser hand, you will probably choose to open 2♣. Also there are hands which partner will pass (2 red queens?) with which game is pretty much a lock.

Another way of deciding whether to open (at the 1 level) marginal hands is the rule of 20. If the HCP plus the lengths of the two longest suits adds up to 20, then if you are in doubt, open. Some players use the rule of 19. This is not to say that a rule of 20 hand is always an opening bid, just that the rule can help when the decision is close.

That hand looks like a 2♣ opener to me, but via my other rule, Jx, xxx, QJ10xx, xxx looks like a possible pass on a hand that makes a pretty much cold slam, it's very decent without ♠J.

LTC helps on hands like this until you get enough feel and judgment to just look at it.

[nige1]

fromageGB, on 2017-February-05, 04:50, said:

From what I see, I think it is a common beginner* mistake to count losers when deciding to open or overcall a hand. This style of counting is ONLY suitable when you have prime support for partner, and only then when you cannot decide by other methods of determination.

* "beginner" includes people who have been playing for 30 years

65 years? I guess I'll never learn.

[rhm]

Playing tricks is a fuzzy concept.

If you hold AKQJTxx in a suit, it is not too difficult to conclude that you have 7 playing tricks there.
Unfortunately I rarely hold such strong suits.
Nobody has been able to tell me what the number of playing tricks are when I hold KJ9xxx in a suit or similar, and that is what I get dealt far more often.

So most of the time playing tricks is not a useful measure for most hands.

LTC is different. The original one is not very useful, because it treated queens (almost) like aces and that is obviously nonsense.
More sophisticated versions are quite useful.

While it is true that LTC comes into play once a fit has been established, I still disagree with those, who claim you should not use it before that fit is known.
So you could say with LTC it is acknowledged that the value of a hand depends what you find opposite while playing tricks simply ducks this issue.
In bidding it is important to understand the potential of a distributional hand, even when you do not yet know you have a fit and even then you may not be able to tell wether your distributional assets face duplication or not.
Of course you do not open a 5 HCP hand with a one-level bid, because it contains only 7 losers, since you can not know at this time the degree of your fit.
But LTC can be used as a discriminator, when you are in doubt.
It simply can not be bad to note, when you hold a hand, whether weak or strong, its potential and get a rough idea how much this potential is.
As a side remark I note, that the rule of 20, which also takes distribution into account, treats 5422, 5431 and 54440 distribution the same, but these distributions have quite different potential.

Any initial action, whether opening the bidding or overcalling is to a large extent a bet, that you will buy something useful in dummy.
When your values are distributional and less in HCP your risk increases, because distributional assets are more dependent on a fit and degree of duplication.
But the rewards are also higher.
HCP are a zero sum measure, because the HCP assets you hold, can not be held by any other player. The total is 40 by standard measure.
The same does not hold true for distributional values.
When both sides can make many tricks, this almost always happens based on distribution.
You will be hard pressed to construct a deal where both sides can make game when everybody holds a 4333 distribution.

Rainer Herrmann

[1eyedjack]

LTC ("Losing Trick Count"), one common occasion when you are expected to count your "losers", purports to provide a statistical expectation of the average number of tricks won and lost that the hand can be expected to produce.

Where circumstances require valuation on the basis of playing tricks, you are generally considering a worst case scenario opposite (or close to worst case scenario, discounting the wholly unreasonable). Thus the "losers" in a playing trick evaluation (13 minus winners) is a differently defined concept. That is your maximum losers with no help from partner, not your average losers that LTC attempts to calculate.

Even in LTC, the "losers" is an artificial concept that only claims to have any relevance when combined with the losers counted by partner on the same basis.

[rhm]

1eyedjack, on 2017-February-19, 04:57, said:

Even in LTC, the "losers" is an artificial concept that only claims to have any relevance when combined with the losers counted by partner on the same basis.

No such claim.
LTC is a valuation method of your hand just like HCP and does not require that your partner uses the same evaluation method.
For example whether I make a mixed raise or a limit raise I may look at my losers to assess my strength.
I often play with partners who do not use it, either because they do not need it or because they have no clue about it.

That's it. There is nothing more than that

Rainer Herrmann

[1eyedjack]

rhm, on 2017-February-19, 05:52, said:

No such claim.
LTC is a valuation method of your hand just like HCP and does not require that your partner uses the same evaluation method.

*You* may make no such claim, but that the claim has been made is well established.

It must be so because I read it on the Internet.

http://tinyurl.com/zmtowj7

But I can see where you are coming from. A hand with low LTC is better than one with a high LTC. And a difference of 1 in LTC is broadly a difference of 1 trick. Even so, if partner is not on board the main consequence will be a drop in total partnership accuracy. And the main point is that the LTC in this method does not correlate to the number of tricks that you expect to win or lose in an absolute sense.

[helene_t]

rhm, on 2017-February-19, 05:52, said:

LTC is a valuation method of your hand just like HCP and does not require that your partner uses the same evaluation method.

Generally speaking, yes.

You need to speak the same language as partner, though. And if you use a highly flawed method like for example unmodifed LTC, it helps that partner also uses it because when you overrate your queens partner is likely to have a couple of aces which he will underrate so you will often access your combined assets adequately.

[rhm]

helene_t, on 2017-February-19, 08:27, said:

Generally speaking, yes.

You need to speak the same language as partner, though. And if you use a highly flawed method like for example unmodifed LTC, it helps that partner also uses it because when you overrate your queens partner is likely to have a couple of aces which he will underrate so you will often access your combined assets adequately.

Oh I agree that it is always best when you agree with your partner and use the same methods
Unfortunately I have seen cases where this does not happen.

Rainer Herrmann

[RedSpawn]

Kaitlyn S, on 2017-February-06, 22:04, said:

Many are teaching the Rule of 20+2 requiring 20 HCP + length in the longest two suits and two quick tricks. The intent was to avoid the debacles like opening:
S-K H-Qx D-Qxxxx C-Kxxxx using the Rule of 20; but the Rule of 20 plus 2 says to pass your hand also.

There is a whole string in Intermediate and Advanced forum dedicated to opening on recycle bin/junk material. And the spirit of rule of 20 and 22 is designed to get you to stick to opens where the working values of the suits of the hand in question are respectable and somewhat quick. The queen and jack hands with absolutely no quick tricks are basically support hands with very low/slow working values. In my opinion, the queen and jack hands are not hands worthy of putting the partnership assets at risk UNLESS you are in 3rd seat and no one has opened. Thus all bets are off.

So, under Standard, you can open them at great risk, but why are you essentially writing a check for a contract with insufficient funds and less HCP strength than average 10 hcp hands? An opening hand doesn't need to be textbook classic, but depending on partnership agreement, it should be decent, respectable, and usable to the declarer with say a 6 hcp response hand.

Also, QJ honors in a doubleton are dubious and present their own valuation and trick taking potential concerns.

[mycroft]

• I wonder who started that thread. Wait, wait...Don't tell me...
• That's the whole point of the +2.

I will note that the Rule of [20, whatever] was designed, like Pearson Points, LTC, and basically everything else, to be a decider for hands one is unsure about opening. "When in doubt, open weaker with shape than balanced; it's more likely you have a fit and your hand will improve when you find it." "When in doubt in fourth seat, value the boss suit, because it will be a competitive fight, and you're guaranteed not to go negative if you pass." "When in doubt, honours in long suits and empty short suits will likely play better than honours in short suits and empty, or single-honour, long suits, so open the first and pass the last."

There are those, like with Pearson points, LTC, and basically everything else, that turn this decider-of-last-resort into the primary requirement. Viz the partner I played with who didn't open a 10-12 1NT "because I don't open 10-loser hands" or the 8 QT754 AK J8752 1♥ bidders because it's "20+2". Fine, whatever. Feel free to play against me. But throwing the method out because some of its users use screwdrivers as chisels and oddly enough, have trouble both turning screws and cutting wood - is equally bad.

[steve2005]

fromageGB, on 2017-February-05, 04:50, said:

This style of counting is ONLY suitable when you have prime support for partner, and only then when you cannot decide by other methods of determination.

I agree loser count is only suitable with a fit. I have one partner who opens all 4 loser hands 2♣ often with disastrous results.
However, if a good fit is found loser count often is superior or equal to other methods, particularly point count.

[nige1]

We count winners.
A = 1.5, K = 1, Q = 0.5, J = 0.2, T = 0.1,
void = 2.5, singleton = 1.5, doubleton = 0.5.
Trump control = 1.
Adjusting for honours in short suits, reinforcing honours, texture, and duplication.
This produces a similar result to the LTC.

Whatever evaluation method you use, if partner uses the same method, then being on the same wavelength has obvious advantages. e.g. in agreeing the meaning of calls, and in assigning blame after a disputed lapse of judgement.

[Zelandakh]

steve2005, on 2017-April-28, 19:28, said:

I agree loser count is only suitable with a fit. I have one partner who opens all 4 loser hands 2♣ often with disastrous results.
However, if a good fit is found loser count often is superior or equal to other methods, particularly point count.

You do understand that LTC and MLTC are just point count methods dressed up in different clothes, right?