On Friday, June 7, 2013 4:02:00 PM UTC-4, Bertil wrote:
> Some textbooks say that a TO shows an opening hand without a good 5-card suit.
That seems to be an oversimplification to me. I can think of quite a few examples that don't fit that definition.
The takeout double's message is fairly simple. It is "I want to compete but would like your input on where and how high. I am prepared for whatever you bid." That means the doubler is ready to either pass or raise advancer's response, or has enough to make an independent suggestion later.
> Another description says it shows a hand with 13 valuation points, so that
> the TO hand can count short suit points, 5-3-1, to obtain the 13 pts.since the
> TO hand expects to become dummy with an ordinary TO hand. Is this realistic?
>
It's been fairly common practice for several decades now, although I'm sure experts don't stop to add the distributional points, they just know instinctively when to double. If it was unrealistic, don't you think it would have gone the way of the many "standard" bidding practices from that era that are now obsolete? Or that there might be a strong and visible minority doing other things? (I believe there are people who do other things, but I'm not sure they're that visible to the casual observer.)
> The TO hand has no assurance of an 8-fit, and thus is not justified in
> counting short suit points.
Like all bids, the takeout double is a bet that you can improve your score. When you have three suits, betting that partner can fit one of them is not unreasonable. A takeout double where two suits have been mentioned by the opponents already (especially if the auction is at the two level) is acknowledged to be more risky.
>
> As an alternative I would suggest counting the number of losers in the opened
> suit and add 11. Then if the total equals at least 13 I would examine the hand
> for potential trump support of at least 3 cards. Since the advancer cannot
> guarantee at least 4 card in the bid suit, there is no justification for up-
> grading the TO hand.
It should be fairly feasible to test that theory. There should be an identifiable subset of hands where your suggested method and "standard" methods diverge. Testing a sample of those might give an answer as to whether better scores are achieved. As they say in math, "show your work."
>
> Keep in mind I'm referring to Std.Am. type bidding.
>
Which is a nebulous concept at best, and does not dictate hand evaluation methods. Standard methods simply divide the spectrum of hands into "weak", "minimum", "medium", and "strong." Evaluation methods are tools to assist in deciding how to classify particular hands. Differences are usually only found in borderline cases, which have always been debatable.
--
- Jon Campbell
Ottawa Canada