Hand Evaluation Tool

[frank0]

Do you regularly use some advanced hand evaluation tool beyond HCP(Zar point, Kaplan point, LTC and some modified version, SST/WP, control count, etc)? How do you like those methods? Do you rely on those tools or your judgment and feeling based on your experience when you evaluate the hands?

[antonylee]

I sometimes use LTC but mostly rely on "how I feel about the hand". On the other hand my regular partner seems to justify all his bidding on Zar points...

[Foxx]

If I could somehow learn and memorize the K&R formula, that would be my first choice. http://www.bridgebase.com/forums/public/style_emoticons/default/smile.gif But that still wouldn't be a solve-all, because it doesn't account for fits in partner's suits, etc.

[nigel_k]

The most important thing is to revise your value of the hand as you get further information from the auction. The problem with Zar, Kaplan and similar methods is that people use them for opening bids and commit themselves to getting too high when the hands don't fit well. A better approach with shapely hands is to start out with something in reserve, then bid more later if a fit is found and be able to stay at a makeable level if not.

[Quantumcat]

I use Marston's 26-point thing if I'm not sure whether slam is on (count 2 points for ace in pard's shortage, and nothing for any other card. If your total points come to 26, you can probably make slam). I sometimes use LTC. I use control points in my relay auctions (A=3, K=2, Q=1). That's it.

[helene_t]

For notrump I use HCP and then add or subtract up to one (rarely one-and-a-half) point for length, fitting honours etc.

 

For suit contracts I imagine some possible layouts and try to figure out if we have more than three losers.

[Siegmund]

I am fond of LTC. The textbook version is imperfect, of course -- but it makes you think about counting tricks rather than counting points. When the outcome of the deal isnt what you expected, you ask yourself "which of my cards did I think would be a loser but wasnt" or vice versa, and refine your judgment accordingly in future.

[SteveMoe]

I like Wirgren & Lawrence's Short Suit Total approach outlined in their book "I Fought the Law of Total Tricks". Andrew Gumperz offers a distillation here:

 

Gumps Tips Short Suit Total-1

 

and continued here:

 

gumps Tips Short Suit Total-2

 

or see details on Wirgren's Website

 

Separately, before trying MLTC, be sure to internalize Jeff Ruben's In and Out Valuation - A, K, and Q differ in the likelihood they will win tricks based on whether they are working together or separately, in long or short suits, and on which is in partner's suit or in a side suit. Opposite partner's 1♠ opening bid, Qxxx AKxx xxx xx > Axxx KQxx xxx xx > Axxx Kxxx Qxx xx > xxxx Axxx Kxx Qx... the Q is more likely to win a trick when in partner's suit, and less likely in an outside suit. The A is less valuable in partner's suit as a lesser honor is in on outside suit. Honors working together are more valuable. Qx, QJ, KQ, AK combinations are highly overrated.

[Zelandakh]

To be honest I usually just use hcp + judgement these days. Before the judgement kicked in (and sometimes still if I am in doubt) I used adjusted Milton where an ace is 4.5 and an unsupported queen is 1.5. Another modification I have used is to count distributional points based on length of partner's suit minus length of my shortest suit. This is an attempt to take account that a partial fit for partner is better than a misfit. The good part about using tools like this is that you see when they work and when not. Such patterns are the basis of switching to a more judgement-oriented system.

 

I strongly believe that "judgement" is actually just another evaluation method, it is just more difficult to define. I also believe this is an area where work should be done to help advance bridge computers, in exactly the same way as intensive work on the expert (GM) evaluation of positional aspects of chess positions radically improved chess computers. The results of such work might trickle down in some form to help out non-expert bridge players too.

[CarlRitner]

Point count is a conscious evaluation system.

Judgment is a sub-conscious evaluation system.

 

Both are a series of calculations and yes/no decision points based on those calculations;

the sub-conscious performs many additional steps much faster and then passes along an answer to the conscious.

[Phil]

Point count is a conscious evaluation system.

Judgment is a sub-conscious evaluation system.

 

I dream about K and R points.

[semeai]

Point count is a conscious evaluation system.

Judgment is a sub-conscious evaluation system.

 

Both are a series of calculations and yes/no decision points based on those calculations;

the sub-conscious performs many additional steps much faster and then passes along an answer to the conscious.

 

I count my HCP and then pass the pair (hand, HCP) to my judgement function (which, as you say, passes some judgement back to my conscious self).

 

This does mean I'm sort of stuck with HCP as it's one of the variables my judgement is tuned to, but I think it's actually fairly good as something to pass to the judgement function. The judgement function then gets some holistic sense of shape, A/K-vs-Q/J-iness, honor placement and combination, lesser high cards [all K&R stuff so far], and then all the additional how-is-it-going-to-play concerns. In some contexts I'll pass a few extra variables to the judgement function. Unadjusted LTC is actually something I pass to the judgement function moderately often --- not because it's a great evaluation method, but because it's quick to compute and because I have some extra judgement tuning in place with it for whatever reason.

[rsteele]

I sort the cards and w/o much thqought use the Pitch count with Goren's short suit method. If it is close I apply Culbertson's quick trick evaluation and then make a decision. If a fit is found I us LTC and if none is found I haul in the reins, (judgement?) describe my hand and let pd make a decision.