Join Bridge Winners
All comments by Doug Bennion
You are ignoring the author of this comment. Click to temporarily show the comment.
Jxx doesn't carry its weight, neither does KJ doubleton. At matchpoints I just go for the plus. Close though. The KnR is 9.20.
Aug. 25, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I've played them for years in local clubs, with zero issues. Just FYI the methods I would use if we played ourselves is this: over the 1 (= hearts) response, X = takeout of hearts, and suit bids are all transfers, beginning with 1 = spades, etc.
Aug. 20, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Mark: Pairs who play transfers there (I do) would consider that a perfect 2 holding. I alert it as ‘constructive’ rather than ‘goldilocks’. A direct 2 raise can be very light (that hand subtract the ace), and should be alerted as such.
Aug. 17, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I'm not sure exactly what you mean with your ‘take out doubles’ question. Can you give me an example?

I'm sure Sabine/Wellands methods are far sexier and wayyyy more successful. Yes knowledge that partner is unbalanced, reverberates through everything, including your bidding judgement and decisions, your defence, your system, and occasionally even your leads. We probably don't take advantage as much as we should.

As I mentioned previously we've been able to incorporate transfer rebids by opener into our forcing 1NT routine, which neatly allows opener to easily distinguish between minimum/good/great holdings. I could elaborate on that if there is any interest, although I'm very sure it isn't optimal. We've also significantly modified our 2NT raise, since opener will always have a nice playing hand and usually has shortness (I think of opener as having a superb dummy for a 4-card major opener). By putting all out-of-range balanced hands into 2+ 1, you pretty much lose the club suit, at least until the rebid, but you really sharpen up your other openers, 1 being 4+ unbalanced and usually 5+, 1M containing no 5332 hands. We play 11-14 1NT, so play transfer responses to 1 to rightside M and notrump contracts when opener is 15+ balanced. 5422 hands are sometimes ‘balanced’ sometimes not (4=5=2=2 we use Flannery, 5=4=2=2 we open 1 having an easy rebid, 5m4M22 are ‘balanced’, 5M4m22 usually opened 1M).
Aug. 6, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I never open 1M with 5332 unless 18-19 (open them 1NT or 1 out of range). Knowing 1M is always ‘unbalanced’ is a big help in some auctions, and permits transfer rebids by opener after 1M - 1NT.
Aug. 5, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Alan: Yes good idea. For the 3 invitational version you can't pull up in 2, but otherwise nice.

And if the overcall is (2), your ‘takeout’ bid is 3 (3 used for heart transfer), so you can't land in a 3 contract.
July 27, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Frances:

It's trivial to construct hands that would benefit from other methods. In the cases you cite, a negative double would obviously be nice to have. Other times, you'd be wishing your negative double was a penalty double, or just maybe could be used as a transfer. The trick is to optimize what you play, maybe you think you have.

In my experience playing negative doubles, the ideal hands rarely arose. And when they did arise, partner rarely had the nice ideal fit and we ended up in 2M-Moysians and worse, 3m-Moysians when the opp's contract wasn't making. And when responder in some frustration at not being dealt an ideal hand, decides to double with a less-than-ideal hand, the results are worse. They aren't for me, but hey if you love them ..

In the first example, yes you would pull up in 2NT and possibly miss a spade fit.

In the second yes you would pass. However opener is encouraged to liberally balance in that position.
July 27, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
For what it is worth, the south hand has a K&R value of 16.70.
July 22, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
To the best of my knowledge, all the written defences in the ACBL database apply to opening bids. The immense ‘difficulty’ with transfer responses to 1 is responder's bid, so any requirement for a written defence would be precedent setting. We do provide one if asked, as a courtesy.
July 18, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I like the transfer idea, although you will lose some 4=4 spade fits I guess (fewer here if you play Flannery). Any reason you don't use a ‘normal’ transfer-suit sequence?

X = normal heart raise
3 = spades
3 = clubs
July 18, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
One way to reduce the range of that 3 bid is to play transfers in those positions, at the cost of whatever you now use XX for. So after (1) 2 (X), then XX = diamonds, 2 = ‘cue’ raise, 2 = spades, 2 = quality club raise, 3 = junk.
July 6, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I too like 1M 2NT; 3 to show any minimum. You'll find half the time responder will simply bid game, and you'll have disclosed nothing to the opps.

Also there are times when opener will want information from responder, yet captaincy (at least) initially rests with responder. Our responses are these:

3 = any minimum
3 = shortage somewhere, 3M then asks
3!M = no shortage, the likes of 5422 or 6322
3NT = 5332 18-19 (weaker 5332 in 1 and 1NT)

That leaves 3OM, 4 and 4 for control-asking duties.
July 4, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Flying by the seat of my pants here, don't actually play these transfers (but might talk myself into it). The transfers would help distinguish when opener is some 64xx or 65xx, compared to some 54xx. To show say AKQx/AKQxxx/xx/x, he can self-transfer with 3, then rebid 3 to show this hand. With AKQx/AKQxx/Kxx/x he would rebid 3 (my hurried suggestion), or your suggested 3.

You're asking which method is better: 3 = h + c and 3 = h + s, or 3 = h + s and 3 = h + c. Maybe the former making it easier for responder to show spades when opener has h + c?

The transfers might also make it easier for opener to show some 3-suiters. With AKQx/AKJxx/AQxx/- he rebids 3 = diamonds, then if that accepted, bids 3 to complete his 3-suiter.

But I'm making this us up as I write.


June 14, 2015
Doug Bennion edited this comment June 14, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Could also play transfers there; 3 = diamonds, 3 = more hearts, 3 = hearts and clubs.
June 14, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Oh I was speaking from memory, and my memory was faulty, when I said KR subtracts 1.0 points for 4333 hands. In fact it subtracts 0.5 points. Sorry about that I must have been thinking about JEC's evaluation trick. The actual exercises used the right value.

10000 deals, 15 HCP, 4333 hands (KR subtracting 0.5):


KRP < 13 = 356
> 13.00 = 340
> 13.25 = 500
> 13.50 = 791
> 13.75 = 978
> 14.00 = 1124
> 14.25 = 1210
> 14.50 = 1408
> 14.75 = 1101
> 15.00 = 697
> 15.25 = 618
> 15.50 = 454
> 15.75 = 237
> 16.00 = 186


So one probably should consider downgrading at least 30% (978+791+etc) or so.

[edited for consistency}

June 7, 2015
Doug Bennion edited this comment June 7, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Well that conclusion from Mr Andrews' analysis seems unlikely to me in a practical sense, when the whole world agrees that aces and kings are undervalued, and quacks overvalued, as in one ace really equals six jacks, not four, approximately. KR adjusts for that among other things, HCP does not.

I'm no statistician, and I stand to be corrected, but since the author's analysis is for all hands, not just balanced hands, perhaps that confounds things. Maybe somebody smart could revisit his exercise just for balanced hands.

Also HCP is coarser than KR. The 4432 exercise I did (for 14 HCP balanced hands) shows that average KRP over 10000 hands was exactly 14.00, hence exactly equal to average HCP. However KRP values ranged from 11ish to 17ish, so I don't know how the fact that one measure materially spans another, plays out in Mr Andrews' analysis. In this example it makes little sense to me to claim from ‘average HCP = average KRP’, the two measures are equally good at evaluating for notrump contracts when you can fine tune one, but not the other.
June 7, 2015
Doug Bennion edited this comment June 7, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Yuan:

Yes your method looks like a reasonably easy and effective scientific eyeballing :-)

Georgiana:

For sure 6:4:2:1 captures the relative strengths better. Twenty some odd years ago, with the help of a deal generator and double dummy analyser, I developed from first principles (being trick-taking ability) the various honor values, starting with J = 1. My exercise produced these values: A = 6.5, K = 4.5, K = 2.5, J = 1. I think you added a point for doubleton honors, maybe another one for tripleton. I called them Little Jack Points.

Phil:

Thanks. Good point about the doubleton honors. Positional considerations are critical in valuing honor combinations as well. With strength in RHO, AQ(x) is very nearly as good as AK(x), AQJ nearly AKQ, ditto AKJ, AJT can be as good as AQx, heck KJx can be nearly AKx value. But if strength in LHO, all of those nice combinations lose a lot of value. For sure good expert judgement can trump KR.

Also is partner bidding? If partner is showing majors, Qxx/QJxx/xx/xxxx is nice hand, but if he’s showing minors, it’s crap.

Also if partner is bidding, maybe you shouldn’t be downgrading. Your isolated quacks may now be supporting partner’s aces and kings. Partner might have those filler T’s and 9’s that you lack.
June 6, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
North has announced a weak notrump, opposite which south has a pass. Sure it's a max pass, but a pass regardless. Sometimes weak notrumpers miss 14-10 games, like sometimes strong notrumpers miss 17-7/8 games.
June 1, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
North is not remotely close to having an upgrade-worthy hand. 13.85 on the KR site.
June 1, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Nice try but dammit the hand doesn't always have four card or even three card support for a 2nd suit. I'd want to open 1C with the likes of xx/xx/Kx/AKJxxxx, so you would want to include long-suit points as well. Maybe carry around a Kaplan Rubens Evaluation calculator.
May 30, 2015
.

Bottom Home Top