You're correct in that my preference is very much for 2
♥ immediate negative. I have played it for many years.
You are also correct in that all conventions carry costs, and this is no exception.
Firstly: when opener holds hearts. My preference is to require a better hand for 2
♣ then hearts than for 2
♣ then spades. The reason is that it makes little sense to play the auction 2
♣ 2
♥ 3
♥ as non-forcing, since how else is opener to begin with a massive hand that can't commit to 4
♥, either because the suit isn't good enough or because opener has a massive 2 suiter? So for me the 2
♣ forces to 2N, 3
♠, 4m or 4
♥.
In addition, you point out that responder may have a problem if opener bids 3
♥ over 2
♥. This is minimized to a small degree, albeit with some other minor costs added, by having opener strain to rebid in NT, rather than hearts, with 5332 shape. Thus 3
♥ will usually be 6+ in length. Note that I generally play a modified form of puppet over 2N so we will usually get back to a 5-3 heart fit when appropriate.
You didn't mention the right-siding issue, which can (tho rarely) be another negative since the opening lead now goes through the strong hand if we play in hearts.
The huge NT is yet another issue, since we lose Kokish, as you observed. However, this is truly a rare problem. I suspect the mathematically inclined can tell me how much more often we hold 22-23 than 24+ but my subjective impression is that it is an order of magnitude. Of course, I am generous in upgrading 20 counts into 21-22, and not so generous upgrading 23's out of range so I really play 21+-23 for the 2N rebid and that greatly increases frequency. My experience also suggests that 24 count hands are so chunky that playing 3N rather than 4M on a 4-4 rarely means going down: tho it may well mean making one fewer trick. My bidding methods are optimized (as best as I can optimize anything) for imps, so I don't see this as much of an issue. The same seems true, again based on a small and not-remembered subjective impression, for 5-3 fits, and for 6 card suits we can transfer.
That is, admittedly, quite a collection of drawbacks and I suspect I have ommitted others. The key is, as you know, whether the gains justify incurring these drawbacks and, to me, there is no question about it.
2
♦ waiting has huge issues of its own, the main one being that it takes 2 rounds of bidding, and most of a level, for responder to show that he has either nothing or something. While he can pass 2N with a flat bust, he can't pass 2M or 3m. He has to make an artificial bid to show the negative.
Not only does this take away from our available bidding space directly, but say one has real clubs and a positive (but not enough for a direct 3
♣, perhaps because 3
♣ promises a certain minimum suit quality). Opener rebids 2M. Since cheaper minor 2nd negative is a common adjunct to 2
♦ waiting, responder can't bid 3
♣. This would be a rare occurence and might not be fatal anyway. The loss of bidding space is the greater problem. I know, 2
♥ immediate negative also costs bidding space when opener has hearts (see: I knew I'd overlooked at least one problem) but that happens less frequently.
And when responder has a positive hand for the wait, we have delayed giving that news for a round. Imagine opener rebidding 3
♦. What now?
Or opener rebidding 3
♣? When one has already shown a gf via 2
♦, it is trivial to play 3
♦ by responder as a stall, or 'noise', clearing the way for opener to show a 4 card major or to bid 3N. And if responder has shown a negative, it is trivial for responder to either bid the same 3
♦ or bid a major, without fear that partner will go nuts.
Contrast that to cheaper minor 2nd negative. Responder can't bid even a 6 card major over 3
♣ with a bad hand: he has to bid an artificial 3
♦ and the partnership is left groping.
I don't think you can solve these issues by, say, abandoning cheaper minor 2nd negative, since responder has to have some way of limiting his hand or of confirming values. You cannot safely bid a 2 count 5332 the same way you'd bid an 8 count 5332, at least not in my opinion.
In addition, telling opener immediately that responder either owns a control or doesn't can, on a few hands, be of immediate and useful benefit to opener. Btw, I prefer that 2
♥ says nothing about hcp: that it deny an Ace or a King. I'd respond 2
♥ with 4 Queens and 4 Jacks (tho I'd certainly get active thereafter
![:P](http://www.bridgebase.com/forums/public/style_emoticons/default/tongue.gif)
) but I know some who argue that 2
♥ denies as many as 6 hcp. I could go on at length as to why I see that as second-best, but that is a digression).
I suspect that there are more disadvantages to 2
♦ waiting than I have identified, but I wanted to give you at least some of my thoughts.
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari