MrAce, on 2011-July-28, 15:09, said:
IF you think after giving a long lecture about the percentage of ♣J, frequencies of a flat 5-6 hcp hand, asking for simulations which has to apply all the hands generated to all the different openings, overcalls and then plays and defends for you, as if you have a laptop available at the table, AND THEN using words like "At the table..Real Bridge" in same post is common sense
And this is pretty close to how a decent bridge player's brain works. I looked at my hand and went through this thought process:
1: It's a 20 count
2: It has pluses and minuses, I don't know how big it really is
3: If partner has a flat hand, how many points does he need for me to make game in NT ? I'll treat it as 25 minus that.
4: Are we likely to miss game if partner has 4 spades and 4 hearts ?
For point 3 I guessed 6-7 hence my valuation of 18.5 and did this in a time frame I could take at the table.
Thought process (snap assessments, very rough)
Most of the missing points are in the majors so start there then try diamonds.
A♥/Q♠ not good
Q♥/A♠ better, but probably no better than 50:50
3 queens chances, but again not great
♥AQ nope
♠AQ nope
Edge cases J♣ makes a big difference but only hold that 1/3 of the time, ♥J109x or similar and useful cards outside might do it.
For point 4 I answered no given the system I play
I called for simulations to see if the evaluation of 6-7 points needed for 3N was correct.
The sources of hand evaluation you used are only as good as the people who programmed the algorithms. Without knowing the code I wouldn't begin to trust them, and in a case like this I'd trust the human brain more.
I'd be very interested to see how other bridge players approach the "at the table" evaluation process (and it might be quite informative for B/Is)