DIP: Basic Hand Valuation

I’ve been distracted tangentially recently.

It started because I was thinking about the 1NT response structure:  in particular the invitational sequences.  This got me thinking again about hand valuation.  If it’s applicable in its basic form anywhere, it’s for auctions like 1NT 2NT and 1NT 3NT (or their DIP equivalents).  This led me (back) to the work of Thomas Andrews.  SD assured me his work was significant, and indeed it is.  You can find it, or get to it, from here

http://bridge.thomasoandrews.com/mozilla.html

It is his hand evaluation work that is of most interest.  As you know, DIP currently uses ppc (Practical Point Count) valuation, where A=4.5, K=3, Q=1.75 and J=0.75.  It would also be reasonable to name this valuation method “Fourths” (you will see why later, apart from the obvious point that it uses a quarter point granularity of sorts).

Though Andrews cautions with the risks of using his current conclusions (and some of his work is ongoing), I am sufficiently impressed to move DIP across to it now.  Specifically, DIP will adopt his “Fifths” evaluator, where

• A=4.0
• K=2.8
• Q=1.8
• J=1.0
• T=0.4

In DIP, we’ll refer to it as fpc (Fifths Point Count).  Like ppc, it shares a 40 point deck with the classic hcp count, which should make conversions and explanations relatively easy when needed.  Also, I have always found it annoyingly difficult to write 1/4, 1/2 or 3/4 in DIP, so am looking forward to being able to simply write “0.8″ or similar where needed.

In time, it may well be that I need to change some of the opening and response ranges to cater for the use of fpc.  We’ll keep them roughly as they currently are, for the present, and cross that bridge when we come to it (that is, after more experience with the method).

I think it is fair to compliment Woolsey as to how close his intuitive thoughts were to Andrews calculated ones, except for the apparent under-weighting of the J.  Scaling “Fifths” to where A=4.5, we see how close Woolsey was (ppc first, then fpc scaled in brackets) …

• A = 4.5 (4.5)
• K = 3 (3.15)
• Q = 1.75 (2.025)
• J = 0.75 (1.125)
• T = n/a (0.45)

Interestingly, Andrews also refers to Martelli’s valuation method (named BUM-RAP) which is geared to suit contracts.  It differs from ppc only in that Q=1.5 and T=0.25.  Nevetheless, I expect with many suit-contract auctions, users of DIP will be going down a different route than simplistic hand-evaluation, so this direction of thinking is not of great short term relevance.

I think I’ll use the change in valuation from ppc to fpc to change this blog’s structure slightly.  I’ll pin the current system, only as far as discussed in the blog, to the front page (and maybe begin a change log).  I’ll then let the current front page go to its “natural” place at the bottom.

Until then!

Regards, DipBridge

DIP: Responding to 1NT [II]

Since writing part [I] of this section, the signifcant other developer has been in touch about it.  I’m going to get sick of writing that term (significant other developer) after a while, so let’s refer to him here on as SD (which is preferable to the obvious alternative).

SD had some good ideas and thoughtful opinions as always.  I’m going to order them in this blog to suit my subsequent purposes, rather than how he sent them to me.  This is because they are somewhat intertwined (though not for reasons that are immediately apparent).  A summary of his ideas, opinions and thoughts is

• Why not switch the proposed treatment for 4H4S INV hands (currently 1NT 2D 2H 2S) with that for 5H INV hands (currently 1NT 2C 2D 2NT)?  SD’s main rationale was so that after 1NT 2D 2H 2S 2NT, responder could then bid 3m with 5m (or concentrated 4m maybe).
• If STAY(man) is to be used over 1NT (and more importantly, 2H to be available as R over 2D showing no M) why not allow 5m4m22 hands into 1NT?  There is sufficient space to do so (but not over 1NT 2S R as I had planned to use some time ago, but discarded).
• Given that if opener has a major he’s 4333, there is minimal concealment advantage when opener is declarer (which SD understands was behind the Scanian rationale for transferring with 4M’s).  So why not reverse the STAY responses, so 2H shows 4=3=3=3 and 2S shows 3=4=3=3?
• Should opener always super-accept a transfer with 4333 shape?  SD thinks maybe not.

SD also had some other comments (puppet-based alternatives, really) which are not directly relevant to what will appear below and won’t appear in this blog.

OK, let me re-state (and mildly re-order) the invitational hand types that we would like to find a home for below their “safety level”.

1. BAL, no M
2. 4S not 4H
3. 4H not 4S
4. 4S & 4H
5. 5S
6. 5H
7. 5S & 4H
8. 4S & 5H
9. 5S & 5H
10. 5H, 4+m
11. 5S, 4+m
12. 5+D (possibly with 4M) or 6+D
13. 5+C (possibly with 4M) or 6+D

This list is expanded from the original, with items 1-8 being the ones under primary consideration.

To repeat, SD’s first idea above was to play 1NT 2D 2H 2S as 5H, INV (swapping the 4S4H hands to 1NT 2C 2D 2NT).  I wasn’t initially a fan – it broke the “2D = 4 or 6 H if INV” rule, for one thing.  So I ruled it out, but then thought about it some more.  The fact that the 1NT can have one major, but not both, gives some interesting properties to STAY-like sequences in DIP.  Doing what SD suggested to achieve the primary desired effect (expression of INV 5H/5m hands) had a powerful side-effect:  all of cases (1) to (4) above could be brought safely into STAY, not just case (4).  On further thought, so could case (7), 5S & 4H.  The reason these 5 cases can be handled by STAY is because all potential 4-4 M fits are cleared up on the first response to STAY.  The only thing left to resolve from cases 1,2,3,4,5 & 7 is whether there is a 5-3+ S fit (and even that is already clear after a 2H 3=4=3=3 or 2S 4=3=3=3 response). I (very much) hope my logic above is sound, otherwise the whole house of cards below will fall apart.  Let’s cross fingers and assume it is.

So, by adopting SD’s suggested change, we have freed up at least four sequences that are safety-level relevant (1NT 2D 2H 2S, 1NT 2D 2H 2NT and 1NT 2H 2S 2NT and 1NT 2S) and only have two cases from the original 2C response that need alternative handling:  (6) 5H, for which we have already allocated a home, and (8) 5H & 4S, which we’ll give to the 1NT 2D 2H 2NT sequence.

So, the sequences 1NT 2H 2S 2NT and 1NT 2S are now freed up.  So what to do with them?  Let’s move to SD’s second suggestion (desire?):  allowing 5m4m22 hands into the 1NT opening.  Always a nice aspiration, my traditional fear has been the damage done to the safety-level of case (9) hands, those with 5S & 5H.  But now, we have an elegant way to deal with them:  1NT 2H 2S 2NT (NF).  I quite like the feel of this, as it is sort of analagous with 1NT 2D 2H 2NT showing H/S hand as described earlier.  So

• DIP can now officially allow 5m4m22 hands into the 1NT opening (with a decision still to be made as to whether to prohibit them from the 1S and 2C openings, with my inclination being “yes”), and
• 1NT 2S can be put to a more specific use.  This is likely to be both minors, of all strength ranges, a variant of what had been planned for it anyway.
• We’ve catered explicitly for cases (1) to (9) above, and cases (10) & (11) can be implicitly handled by responders 3rd round bidding of 3m after one of the three possible 1NT 2X 2Y 2S 2NT auctions

Assuming no logical flaw, we’re looking pretty good.  Case (12) and (13) can and should also be handled via STAY and simply rebidding 3m over the relevant response.  This gets a chance to find the 4-4 M fit on the way through, if relevant.  Finally, over a 2M response to STAY, 3M is INV, 3OM is R.  Note that in 1NT 2C 2S 3S, responder might have 4 or 5 S (if he chooses not to revalue the latter upwards).

So, the overall response structure to 1NT (though not complete) now looks like

• 2C:  STAY (ART FG R or BAL no M INV or 4S &/or 4H INV or 5S maybe with 4H INV or 5+m maybe with 4M INV)
• 2D:  TRF to H (WK H or 5H maybe with 4S INV or 5H with 4+m INV or some other strong hands as yet not spelled out)
• 2H:  TRF to S (WK S or 5S & 5H INV or some other strong hands as yet not spelled out)
• 2S:  both minors, any strength

SD’s third suggestion was to possibly reverse the M responses to STAY.  The short answer is that I don’t think you can and still play the above structure.  There are too many handling issues (basically, not enough bids over 2S=H and too many over 2H=S).  You COULD make it work by also switching the 5S cases with 5H ones wherever they occur, but playing 1NT 2D 2H 2S as showing 5S and not promising H is a bit too rich for me!

The final thought from SD was the merit of forcing a “super-accept” with a 4M333 when there is a 4 card fit to suit shown by TRF.  There is certainly now no compelling requirement to do so (with all the 4M invites moved into STAY) but it still feels right to me.

The above structure now feels right and sits comfortably within the system.  The downside would appear to be a fair bit of work on the 1NT 2C sequence which might be difficult to unscramble if 4th hand competes.  Nevertheless, the fact all WK sequences have been removed gives some latitude in this respect.  There is an argument for playing pass as forcing over some interference to 1NT 2C sequences (maybe at the two and four level)?  This blog is already long enough, so we’ll leave that for another time.

Regards, DipBridge

DIP: Responses to 1NT [I]

Back again.  This post would have been sooner except the responses structure to 1NT (BAL no M, or 4M333) have not been finalised.  Unfortunately, they still haven’t been finalised, but at least the thinking behind the requisite structure is becoming clearer.  As such, I am in a position to partially outline a potential structure.

The decision to include 4M333 hands as well (they now comprise 2 shapes of the 12 possible in the opening) was a fairly late one.  It is amazing how much the possible presence of a major suit complicates the response structure.  Nevertheless, here are some thoughts.

The “safety level” for invitational hands that are not 5+/5+ or 6+ in shape is usually 2NT.  I am going to assume for the present that inviting with invitational hands is regarded as desirable (there is a case for simply forcing to game with such hands, allowing extra space for strain investigation as needed).  If so, one needs to cater for the following invitational hand types at or below 2NT

1. BAL, no M
2. BAL, 4H
3. BAL, 4S
4. BAL, 4H & 4S
5. BAL, 5H
6. BAL, 5S
7. BAL, 5H & 4S
8. BAL, 4H & 5S

I’ve been through quite a few permutations as to how these might be handled.  Originally, I was proposing 2C as a puppet to 2D.  Or more accurately, I was thinking about a structure where each of these 8 types was somehow shown explicitly.  I just couldn’t get it to work* (maybe someone can supply a mathematical proof to show that I couldn’t)!

In a related, but different conversation, I had suggested to the significant other developer that maybe Stayman and Transfers had “had their day” as they were unlikely to have naturally evolved in an optimal form by accident.  Well, that may be true, but I find myself somewhat eating my words with what I’m about to suggest.  I think 2C as STAY(man) is probably going to work best here, but with one slightly unusual caveat:  that if invitational, at least one 5M is promised.  The reason will become apparent.  The responses to STAY are fairly normal (though by definition, more narrow)

• 2D:  no M
• 2H:  3=4=3=3
• 2S:  4=3=3=3 (it is possible that a subsequent version will have 2S:  4=3=3=3 MIN and 2NT:  4=3=3=3 MAX, but until the rest of the structure is complete, it is not yet clear)

Over these responses (and 2D in particular) 2S will show 5S INV and 2NT will show 5H INV.  It should be clear from these continuations and the fact that STAY has already accounted for any 4-4 M fits, that cases 5-8 above can be satisfactorily handled by this method.

The STAY path will also be used by strong hands that wish to R.  I had originally toyed with a direct 2S over 1NT fulfilling this functionality (better when 4th hand intervenes) but

• I need the direct 2S now for something else, and
• The extra step gained by using an indirect 2H rather than a direct 2S as R is quite significant when it gets to control showing.

So, over the 2H 3=4=4=4 and 2S 4=3=3=3, at least one three level bid (yet to be determined) will need to be reserved as control asking R.  Over 2D (no M) then 2H R, the structure will continue

• 2S:  1/S clubs (resolves as for 3C+ below after 2NT R)
• 2NT:  4C & 4D
• 3C:  3=3=4=3
• 3D:  2=3=5=3
• 3H:  3=2=5=3
• 3S:  3=3=5=2, B+2 or greater QP’s (9+ QP’s, i.e. A=3, K=2, Q=1)
• 3NT:  3=3=5=2, Base or B+1 QP’s (i.e. 7 or 8 QP’s, rarely 6, which is counted as “Base”)

You are getting a bit of a taste of the underlying relay structure above

• Low length first
• High shortage first
• Base = MIN(ppc) * 6 / 10, rounded down to the nearest integer (12 * 6 / 10 = 7.2 = 7 in the above case).  Note also that the use of ppc rather than hcp protects a little against B-1 QP hands being in range for the various openings.
• When two steps are available to run on and show QP controls (3S & 3NT), 3NT shows B or B+1, 3S shows the rest (although in some circumstances, 4C+ or 4D+ can be used to show B+4 or better)
• When three or more steps are so available, all steps but the first show B, then B+1 etc.  The first step then shows the rest (see the 4C & 4D example immediately below)

The both minors hands above show controls slightly differently (as they have an extra step).  After 3C R

• 3D:  2=3=4=4
• 3H:  3=2=4=4, B+2 or greater QP’s
• 3S:  3=2=4=4, Base QP’s
• 3NT:  3=2=4=4, B+1 QP’s

I’ll expand more on the resolution and control showing at a later time.  Suffice to say, it is not absolutely optimal and not intended to be.  It is intended to be practical, easy to remember, and nearly optimal.

OK, cases 5-8 from the original list are handled by STAY.  What about cases 1-4?

We handle 3 of the 4 cases by using the transfer mechanism provided by 2D(=H) and 2H(=S).  It is worth noting that if INV strength, these will have a 4M and/or a 6M, but not a 5M.  The relevant sequences are

• 1NT 2D 2H 2S (4H & 4S, INV)
• 1NT 2D 2H 2NT (4H, INV)
• 1NT 2H 2S 2NT (4S, INV)

That deals with cases 2-4.  It is almost certain that opener will be required to super-accept with 4 trumps opposite a transfer.  After this, there is not likely to be re-transfers thereafter (except, maybe, in strong auctions) as there is little advantage to opener being declarer when known to be 4333 shape.

The remaining case, is handled by what is initially a range-asking 2S.  It is almost certain that if showing a MAX, opener will be required to give a minor suit preference, so after 1NT 2S

• 2NT:  MIN
• 3C:  MAX, C >= D
• 3D:  MAX, D> C

I hope you can see where all the above is heading.  I hope I can (sensibly) complete it.

Regards, DipBridge

* remembering that I also needed to have a sensible sequence to allow simple FG relaying

DIP: Responses to 1S

Hi All,

Because the nature of the 1S(=D) opening is quite different from that of the 1R(ed) openings, the placement of the emphasis on the response structure varies as well.  The “Pivot”, however, will be kept:  it’s effectiveness will be reviewed over time.  So the response structure …

• 1NT:  R, much as for the major suit openings (i.e. strong, or moderately strong without clear direction, a 1NT’ish type of hand and maybe some raises)
• 2C:  Puppet to 2D
• 2D:  2D (sometimes 3), little if any game interest
• 2H:  5+H, NNF, <4S
• 2S:  5+S, NNF, <4H
• 2NT:  4+D INV or FG ART SPL
• 3C:  Pass (with 4+C) or Correct
• 3D:  3D (sometime 4), little if any game interest
• 3H:  Fit-showing, 6+H, usually only 3D
• 3S:  Fit-showing, 6+S, usually only 3D
• 3NT:  To play
• 4C:  Fit-showing, 6+C, 3+D
• 4D: PRE
• Game:  To play

It is probably worth noting what is planned for the 5+/4+ both majors hands, Constructive or better.  These will be handled via the 2C pivot.  The full continuation spectrum is still to planned, but

• 1S 2C 2D 2H will show 5+H & 4+S (NF)
• 1S 2C 2D 2S will show 5+S & 4H (NF)

MOSCITO2007, for those interested, handles these directly over 1S and in transfer fashion (2C = 5+S/4+H, 2D=5+H [and presumably possibly 4S], 2H = 5+S).  I prefer the approach chosen for DIP as I prefer NNF rather than transfers where the auction is not definitiely constructive in nature, and, the both major versions are relatively less frequent.  Also, going through the pivot allows the variants with either major longer to be shown explicitly.

There is some case for the pivot initially to be treated as a form of diamond raise (maybe a BAL INV).  In which case, opener would have the option of not bidding 2D if “accepting”, and then taking into account the other possible hand types responder might have and selecting the 2H+ rebid accordingly (and, as is typically the case with auctions of this nature, making the first bid which would have been passed were it responders rebid).

Regards, DipBridge

DIP: Responses to 1H

This blog should be mostly a formality:  the responses to 1H(=S) are mostly analagous to those for 1D(=H).  Without further ado …

• 1S:  R(elay), ART(ificial).  The kinds of hands included are likely to be from one of the following catgories:  (i) Slam Interest, or (ii) FG, but strain not clear, or (iii) INV, but strain not clear, or (iv) INV, with 3S, or (v) a hand that might bid 1NT if were available in it’s natural sense.
• 1NT:  “Pivot”:  A puppet to 2C.  Continuations will include (i) INV with 4-5S, and (ii) INV with a primary suit that isn’t S, and (iii) INV, 5+5+ not S, and probably (iv) a slightly stronger way to bid 2S than directly
• 2C:  5+C, NNF, < INV
• 2D:  5+D, NNF, < INV
• 2H:  5+H, NNF, <INV
• 2S:  3S (sometimes 4S BAL), little if any game interest
• 2NT:  4+S, BAL INV or FG SPL
• 3C:  Fit-showing, NF, 6+C, usually only 3S
• 3D:  Fit-showing, NF, 6+D, usually only 3S
• 3H:  Fit-showing, NF, 6+H, usually only 3S
• 3S:  4S (sometimes 5S BAL), little if any game interest
• 3NT:  4S, BAL, FG
• 4C:  4+S, ART VOID (D or H) SPL
• 4D:  4+S, REV TRF VOID (C) SPL
• Game:  to play (note this includes 4H)

It’s hopefully much the structure expected after seeing the 1D response structure.  It was decided to make 1H 4H natural, which means some fancy footwork on the 4C void splinter (4D R then 4H C void, 4S H void).

Until next time, DipBridge

DIP: Responses to 1D

To my audience (small as it might well be), I’m sorry for the delay in getting this blog done.  It came from reviewing as to how best to use the “Pivot” I alluded to earlier, and in particular, how the overall raise structure may work.

What seems clear is that the first step, the R step, should be for powerful hands, where slam might be possible opposite the right opener, or for more moderate hands, without clear direction and/or something to show.  The 1NT “Pivot” should contain hands unsuitable for the above, which will usually mean they have a clear direction, and are probably not FG.

You had a flavour of what this might be in the previous blog, but here are the basic response to 1D(=H).

• 1H:  R(elay), ART(ificial).  The kinds of hands included are likely to be from one of the following catgories:  (i) Slam Interest, or (ii) FG, but strain not clear, or (iii) INV, but strain not clear, or (iv) INV, with 3H, or (v) a hand that might bid 1NT if available in it’s natural sense.
• 1S:  4+S, F1 (Forcing 1 Round), < INV (about 6 to 11 3/4 ppc)
• 1NT:  “Pivot”:  A puppet to 2C.  Continuations to be fleshed out more fully, but will include (i) INV with 4-5H, and (ii) INV with a primary suit that isn’t H, and (iii) INV, 5+5+ not H, and probably (iv) a slightly stronger way to bid 2H than directly
• 2C:  5+C, NNF, < INV
• 2D:  5+D, NNF, < INV
• 2H:  3H (sometimes 4H BAL), little if any game interest
• 2S:  Fit-showing, NF, 6+S, usually only 3H
• 2NT:  4+H, BAL INV or FG SPL
• 3C:  Fit-showing, NF, 6+C, usually only 3H
• 3D:  it-showing, NF, 6+D, usually only 3H
• 3H:  4H (sometimes 5H BAL), little if any game interest
• 3S: 4+H, REV TRF VOID (D) SPL
• 3NT: 4H, BAL, FG
• 4C: 4+H, REV TRF VOID (S) SPL
• 4D:4+H, REV TRF VOID (C) SPL
• Game:  to play

I expect the above to both (i) flesh out, and (ii) change – once it’s used a bit more in anger.  But, it gives a fair to good indication as to the direction DIP is heading in.

I’ll try to get the 1H response structure out a little more quickly, but it should be quite analagous to the above.

Regards, DipBridge

DIP: Some thoughts on relaying in DIP

Over the next few days, I hope to blog about the responses to the Intermediate range openings.  Before doing so, however, it seems prudent to discuss the DIP philosophy on relaying.

I’m sure most if not all long time proponents of strong club and strong pass systems will tell you that it’s all about the Intermediate range openings, not the Strong range (and the typically associated constructive auctions, including Symmetric Relay based ones).  The only time I suspect this may not have been true was the 60’s and 70’s (Neapolitan, Roman, Blue and Precision Clubs) when people were fairly respectful of the Strong 1C openings.  Certainly, if the above is the consensus view, I share it.

When I first came across FPR (Forcing Pass Relay) people used to relay up to S+2 or S+3 over interference, that is, they still continue relaying even when they had lost 2, or even 3, steps.  I imagine many still do so.  I come from the almost the opposite school of thought.  When the opposition interfere, and show a suit, or even strongly imply a suit, I don’t want to relay at all.  Further, even if they don’t show a suit, if they cost us any steps at all, I don’t want to relay (remember, the D structure opposite 1C is already up one step from S).  What do I mean by not showing a suit?  Well, there’s CRASH (Colour/Rank/Shape) but there’s also the kind of stuff I used to play against a forcing pass in my youth:  like a 1S fert over a strong pass, for example!

The significant other developer of DIP doesn’t completely share my views on relays in competition, but, for the present at least, is deferring to my preferences in this regard.

The reason I started the above from the end (i.e. relays in competition) rather than the beginning is that my thinking on relay use and optimisation pervades purely constructive auctions as well.  You have already seen an example of this with the decision to include direct semi-POS and transfer captaincy POS bids opposite 1C:  not optimal from a relay perspective.  The former are good, however, in competitive and fast-arrival auctions, and the latter more effective at determining the correct strain for game (if occasionally less effective at getting to a good slam).

Moving on to what is coming up in this blog, the same sort of thinking is applicable to the relay responses to the Intermediate strength openings (1D, 1H, 1S, 1NT & 2C).  DIP has largley adopted the ideas I first saw in MOSCITO2007, namely

• Allowing responders step R response to contain the kind of stuff you would expect (some INV hands without clear direction and almost all FG hands) but also what would have been a normal 1NT response
• As a consequence, the first round rebids by the intermediate strength opener are largely NAT, to cater for the “1NT” type hand and facilitate a pass by responder

This doesn’t hurt the subsequent relay structure too much, but it would be disingenuous to say it had no cost at all.

No longer needing to bid 1NT with the 1NT type hand obviously frees up an extra low-level bid.  To take 1D(=H) as an example, with 1H now R and including the 1NT type hands, you now have 1S, 1NT, 2C, 2D, 2H, that is, 5 bids at or below 2 of the shown suit.  It would be entirely normal to pre-allocate four of these:  one to a raise, and three to show the other suits (whether naturally, or via some sort of transfer).  The questions are, how to best use the valuable extra bid that has been freed up and whether to play transfers or natural to show the other suits?  Some obvious candidate answers for the former question

• As an extra raise, that is a stronger raise to 2 of the suit.  This is what MOSCITO2007 does IIRC (with 2D being a stronger way to get to 2H over 1D)
• As an explicit game force relay
• Some sort of puppet/Lebensohl’ish type structure

In a previous life, I’ve played the latter method in a slightly different system, and found it useful.  I plan also to use it in DIP in the first instance.  It also fits in quite well with NNF type responses in the other three suits, which I have a slight preference for over transfer responses (the latter obviously being more effective when a highly constructive auction would be useful, but less frequent and hence effective in this scenario IMO).

So, to give you a partial flavour of what is coming up, over 1D(=H)

• 1H:  R, either (i) ART, FG, or (ii) INV no primary support, or (iii) 1NT type hand
• 1S:  4+S, NNF
• 1NT:  “Puppet” to 2C, typically some form of discrete INV hand-type
• 2C:  5+C, NNF
• 2D:  5+C, NNF
• 2H: 3H, NNF

You’ll see the above principles in broader action shortly.

I suppose the final thing to say is that including the 1NT hand types in th R response is also not without cost.  It would not have been unreasonable, in their absence, to have the R response set up some sort of Forcing Pass type auction over interference, at some levels at least (e.g 2 and 4).  The inclusion of the 1NT hand type makes this impractical, so, it’s take-out doubles in competition as high as you want to play them (up to and including 4H is my preference) and transferable value doubles thereafter.

Regards, DipBridge

DIP: Other responses to 1C

Just a shortish blog today (I hope).

To round off the responses to 1C, we now deal with 3C+.

There are a set of hands where the nature of the hand (solid and semi-solid suits) has a higher probability of being more important to get across (first at least) than exact shape.  We deal with these using bids from 3C+.  The responses to 1C are as follows

• 3C:  POS, ART, any solid suit
• 3D:  semi-POS, semi-SOL hearts
• 3H:  semi-POS, semi-SOL spades
• 3S:  semi-POS, semi-SOL clubs
• 3NT:  semi-POS, semi-SOL diamonds

There are probably some sensible uses for 4C+ (I’m open to suggestions) but I’m not going to worry too much about them for the present.

As this is a fairly short blog so far, I’ll touch on my current ideas on continuations after the above.  After 1C 3C(=any SOL POS), then

• 3D:  “I know your suit, tell me your shortage?” – resolve as normal, up 3 steps, specific control showing ignores the known suit
• 3H:  “I can’t determine your suit, tell me what it is?” – 3S=C, 3NT=D, 4C=H, 4D=S, 4H+ D but stronger than 3NT and not wanting to risk a pass
• 3S:  “I know your suit, do you have extra length?” – 3NT = 6, 4C = 7 etc
• 3NT:  To play, suggests a minimum and that the suit is known
• 4C:  ?
• 4D:  ?
• 4H+:  strong suggestion to play – “I either know your suit or don’t care”

As always, open to suggestions for 4C and 4D above.

For the semi-SOL, semi-POS’s, my thoughts were game (any strain) or bid of the shown suit – to play, otherwise, analagous to the SOL stuff, that is first step normal relay, second step asking extra length (remember, the suit is known, so that step is not needed).

As an aside, I first saw the “I know your suit, tell me …” when some Scandanavians were in Challenge the Champs in The Bridge World, sometime in 1997 (maybe it was Fallenius and Nilsland).  They also used 3C over their 1C opening to show a SOL suit, however, they played 3H as that option.  I’m unfamiliar with the rest of their structure, so I can’t tell you if their use for 3D was better, but it seems to me that opener will more often than not know what suit responder has, so the cheapest bid should cater for that possibility if it doesn’t otherwise obstruct the method.  I don’t believe that what I am proposing for DIP does so obstruct:  if you have to ask, only the major suits (or a very strong D hand) force past 3NT.  That’s no problem for the majors, you can play 4M, only losing out on the reasonably rare occasions when you didn’t merely like, but needed to play 3NT (this may be a little different at MPs, but DIP is geared to IMPs).

That’s it for today.

Regards, DipBridge

DIP: Semi-Positive Responses to 1C

Back again.

Today, I’m going to blog about semi-POS responses to 1C. This is one of the most controversial areas of this type of system. Firstly, not everyone reckons they’re a good idea in the first place. If you’re in that school, too bad: as described yesterday DIP has them and they are not under consideration for removal. Secondly and thirdly, having decided to have such a semi-POS oriented structure, how do you allocate responders semi-POS bids, and how do you ensure opener has sufficient rebid tools to diagnose appropriate fits when they’re not obvious.

Before I proceed, I should mention that the latest version of “official” MOSCITO that I have seen (circa 2007) if I can use the term “official”, differs from DIP. When I talk about “MOSCITO2007″ from here on, it is this version that I will be alluding to. The main reason for the difference is that it uses 1H rather than 1S as the DBL-NEG. This is uglier for continuations after the DBL-NEG: you need to use 1C-1S-2C, if anything, rather than 1C-1H-1S for very strong hands, as you want to keep 1C 1H(or 1S) 1NT as (wide ranging) natural. However, having 1H (as well as 1NT+) available for the semi-POS structures gives advantages there to MOSCITO2007.

The way MOSCITO2007 chooses to use the above advantage is to differentiate those hands with a 5+ major (bid 1NT+) and those without (start with 1H). 1H will more often than not be BAL in this scenario, so I imagine 1NT is a fairly common rebid by opener. However, this gives responder a second chance, allowing responder a simple correction into a long minor. I think the 1C 1H auctions are a strongish part of MOSCITO2007. However, there is some memory strain (and the already mentioned DBL-NEG relative difficulties) to deal with as a consequence.

Before I proceed, it is probably wise to further discuss nomenclature. I have previously said what “S” and “D” are. During the course of the blogs, you will probably hear the following types of terms and phrases: “D+1″, “up a step”, “down a step”, “loses a step”, “gains a step”. To the initiated in such things, these are probably already meaningful in context. However, they could probably do with some explanation for novices in the area. Let’s consider them by example.

“D” is generally “S+1″. The plus in this context is a bad thing: it means DIP takes one higher bid that Symmetric to descibe a given shape. For example, POS 5431’s in “D” resolve at 3H rather than 3D as they typically do in “S”. I have heard “up a step” and “down a step” used in both directions. Conventionally, I’ll equate “up” to “plus”: that is, “D” is up a step from “S”. However, “gains”, perhaps unintuitively, is used in the opposite sense. As you have seen me use it previously, so “gains” a step is like “down a step” and “loses a step” is like “up a step”. The reason I use “gains” this way is because of the sense of the word (that is, being plus or up a step is a bad thing in this context, hence it loses). I hope that isn’t too confusing and that you can figure it out when you need to.

Back to DIP, it takes a different approach than MOSCITO2007. Only time and use at the table will tell if it is better or worse. I’ll present the semi-POS schema in its probable final form, and then tell you how to get there using first principles from “D” if that helps with the memory*.

• 1S: H (1, 2 or 3 suited, including all 3-suiters with H)
• 1NT: BAL (including all 5332’s)
• 2C: S, not H (2 or 3 suited)
• 2D: D, no M(1 or 2 suited)
• 2H: C (1 suited only, hence 6+)
• 2S: S, not short H (1 suited only, hence 6+)
• 2NT: S, short H (1 suited only, hence 6+)

When I was first devising this structure, I was thinking along normal symmetric lines, in the relay system “order” that I grew up using (S H BAL D C) with the odd tweak. This meant, as it still does, 2H showing 6+C. Then a wise person asked me: what do you do after 1C 2H holding a (relatively common) 5M332 (say 5S332) minimum as opener? With 2S being relay, do you bid 2NT, 3C, 3S … ? As it happens, I am generally a 3C bidder (but maybe, going through 2NT Bad/Good first: that is a story for another blog).

My initial thought to compensate for this was to reverse suit order, something like 1S=C, 1NT = D, 2C = BAL etc. As a different wise person pointed out, this gave a different, and arguably worse set of problems: difficulty in finding major suit fits opposite major/minor hands (amongst other things).

The problem with a semi-POS structure is that there are multiple objectives, and they are usually in conflict. You want to show a major if you have one, explicitly and early. You also want (as per the difficulties in the above example) to have the higher the initial bid meaning it to be more likely you’ll know exactly how to continue as opener (because, if you don’t, you haven’t got much space to figure it out). Finally (for the present at least) you want to be able to diagnose all major suit (and especially, the often difficult 5-opener, 3 responder) fits relatively easily when they exist.

The solution that you now see I think is a fairly optimal compromise between the various objectives, though this is of course somewhat a matter of opinion. Whenever you have a major, you show it, and you show it quickly. The highest two semi-POS’s contain 6+ spades: opener will usually know what to do when he sees that. The 2NT bid as 6+S, short H is needed to allow the semi-POS structure to be no worse than D+1. The good thing about 2NT is that with long H and very short S, opener will be well informed. Similarly, 2S guarantees at least 2H, once again, helping opener to know what is safe with a long H hand.

The relative weaknesses? 1NT risks mild wrong-siding from time to time. There is some pressure on 1S(=H) by including the additional 3-suiter (HDC) but we have devised a fairly efficient way of dealing with this. 2D(=D or D&C) and 2H(=C) are the largest latent weaknesses – there can be some difficulties in constructive auctions diagnosing the best action hence best fit opposite these bids. However, they are not high frequency, they have their pre-emptive effect, and they seem quite good in non-game level competitive auctions.

Most of the semi-POS structure is D or D+1 levels. For example, 1S(=H) is mostly at D levels, but there is a small portion at D+1 to cater for the additional 3-suiter. 1NT is notionally at D-1 level, but in practice, it is at D level, as we intend to use 2C as STAY and 2D as R.

We are going to use a lot of an ASPTRO-like method in opener’s rebids opposite semi-POS’s, so you may want to familiarise yourself with it before subsequent blogs. Here is a useful link for when this comes up

http://homepage.mac.com/bridgeguys/Conventions/Asptro.html

I’ll finish by summing up what hasn’t already been mentioned above re DIP versus MOSCITO2007 in the semi-POS area. MOSCITO2007 is better at quick diagnosis of 5 responder – 3 opener major suit fits. DIP is better at quick diagnosis of 4-4 major suit fits and 6 responder – 2 opener spade fits. Also, my development colleague informs me that quite a few of DIP’s semi-POS’s, especially 1NT, have quite good pre-emptive effect, shutting the opponents out of the auction. This is because many of MOSCITO2007’s semi-POS auctions go through 1H, allowing a double of it, or a bid of 1S, to get a major suit in.

Regards, Dipbridge

* roughly speaking, start with D, then

• Swap the meanings of 1NT (now BAL) and 2C (now S). This is so 1NT can be natural, and hence easily passed.
• Take the HDC 3-suiter from 2D and put it in 1S(=H). This is consistent with the DIP idea of showing all majors asap.
• Take all the 2S+ bids from D, which showed both minors, and put them in 2D(+D) using the space vacated by the HDC 3-suiter. Bidding 2S+ with these hands would have been a bit aggressive with only semi-POS strength and you would also often be poorly placed in constructive auctions.
• Unfortunately, the original 1NT – 2C swap meant that 2C(+S) was at D+1 levels. We fix that by taking the spade 1-suiters out of it, and using 2S and 2NT for them. This has a couple of nice pluses: quick spade fit diagnosis and some pre-emptive effect. Also, it is nice to know after 1C 2C that responder has a guaranteed second suit (in the case of a spade mis-fit).
  

DIP: Positive Responses to 1C

Hi again.

Shortly, the POS(itive) responses to 1C in DIP, but first, a very short history lesson.

In the original Symmetric Relay (“S” from hereon), 1D was a negative response to 1C, with 1H+ being NAT(ural), POS and IIRC, FG (forcing to game).  With the advent of FPR (Forcing Pass Relay) the extra step freed up by the use of PASS as strong(ish) enabled the introduction of a semi-POS structure, so 1C was NEG(ative), 1D was ART and POS and 1H+ were mostly NAT semi-POS’s.  Arguably, the semi-POS structure was even more important here, because opposite a 13+ hcp “PASS”, the semi-POS range was even more common.

When FPR became unpopular, largely due to playing restrictions, MOSCITO came into vogue.  Having lost the step gained playing FPR, the semi-POS oriented structure was also lost (for a long time, at least).  Recently, however, it has been re-adopted by some.  On a side note, as well, some people who still play variants of FPR play Pass-1C as the POS and Pass-1D as the NEG, gaining either a further extra step OR the ability to reverse the captaincy of the relay auction.

DIP’s design criteria reflect

1. The desire for a semi-POS (rather than POS) oriented stucture, and
2. The ability to transfer captaincy in POS auctions

The cost of (1) & (2) above is the loss of a step versus “S”.  As will be shown below, DIP retains an admirable symmetry (read:  “relative ease of memory”) between its positive auctions (whether captaincy transferred or not) and its semi-POS auctions.

There are three paths, one with two sub-paths, and one exceptional path, that responding to 1C in DIP can take.

• 1C 1H shows a DBL-NEG (something like 0 to 5 3/4 ppc).  This is analagous to auctions like 1C 1D(=NEG) 1H(=R[elay]*) 1S(=DBL-NEG) in many strong club systems.  I prefer the nomenclature DBL-NEG to distinguish it from the simple 1C 1D(=NEG) which can contain both DBL-NEGs and semi-POS’ in many methods.
• 1C 1S through 1C 2NT show the semi-POS hands, and will be discussed in more detail in the next blog.
• 1C 3C+ show a series of special hand concentrations (the exceptional path):  mainly SOL(id) and semi-SOL suits, which will be discussed in a later blog.
• All POS auctions (10+ ppc or so) begin 1C 1D.  Opener can then transfer captaincy by bidding 1H, or can show shape himself by bidding 1S+.  1C 1D 1S+ and 1C 1D 1H 1S+ show the same shapes, just with a different captain.  This 1S+ structure is the archetypal relay structure for DIP and will be referred to as “D” from hereon.

There are three relevant orders for shape/hand-type showing in DIP.

1. H S NT C D to show general hand-type
2. C D H S when resolving length
3. S H D C when resolving shortage

The “D” structure for DIP is simply an elaboration of these principles in the “S” style, so after 1C 1D or 1C 1D 1H …

• 1S = Hearts (1, 2 or 3 suited)
• 1NT:  Spades (1 or 2 suited)
• 2C:  Balanced (including 5332’s)
• 2D:  Clubs (1 or 3 suited, if the latter, then short in a major)
• 2H:  Diamonds
• 2S:  5+ Clubs & 4 Diamonds
• 2NT:  5+ Clubs & 5+ Diamonds
• 3C:  4 Clubs & 5+ Diamonds, short in Spades
• 3D:  2=2=5=4
• 3H:  3=1=5=4
• 3S:  2=1=6=4
• 3NT:  3=0=6=4, NF (weaker than 4D+)
• 4C:  2=0=7=4 or 1=0=8=4
• 4D+:  3=0=6=4 (same as 3NT, but stronger, not wishing to risk a pass, will be elaborated on in a later) post)

For those who don’t know and wish to understand more about “S” like structures, some notes which discuss it (amongst other things) can be found here, in pdf form.

http://homepage.mac.com/bridgeguys/pdf/RelayPrecisionKerr2000.pdf

I suppose it remains to explain why one would want to transfer captaincy after 1C 1D, so that system users know when to do so.

Relay systems generally reach more accurate contracts, if a balanced hand is opposite an unbalanced hand, when the balanced hand is R and the unbalanced hand is RR.  This is because R can then evaluate the worth of any honours he holds opposite the shortage.  When able to transfer captaincy, if opener is unbalanced, he can transfer captaincy and begin to show shape immediately, usually gaining if responder turns out to have been balanced.  Similarly, if balanced, opener can instead retain captaincy and bid 1H, usually gaining if responder turns out to be unbalanced.

A second reason for retaining captaincy (and sometimes overriding the desire to show shape) is if the specific honours in your hand would be difficult (lengthy) to show using our (your?) chosen control asking methods.  This will be method specific (and DIP’s will be discussed in another blog), but a common theme here is a hand containing lots of Jacks:  these are typically diagnosed late in most control asking methods.

A final consideration is:  if opener would normally retain captaincy with a balanced hand, then why include steps in the relay structure that show a balanced hand (or even, semi-balanced) hand?  The answer is, you don’t need to, but DIP chooses to.  You have two options

• Keep such steps in the structure, but assign them a similar, but more defined, meaning, or
• Remove them from the structure

I don’t like the latter, because of the memory strain mainly.  But if you wanted to go down that route, you could get rid of the BAL response after 1C 1D and move everything from 2C+ up a step.  Equally, you could, for example, remove 6322’s and 5422’s from other parts of the structure.

DIP, instead, takes a different approach.  If I can explain it by analogy in the first instance.  In some 2/1 methods 1M 2m 3NT shows something like 15-17 BAL.  One of the reasons for this is that it is sometimes difficult to bid a tight slam on power until one partner shows some extras.  DIP adopts a similar approach:  1C 1D 1NT shows a specific range of non-minimum BAL hand.  The specifics of this range will be discussed in a later blog.

Regards, Dipbridge

* Throughout this and subsequent blogs, the person making the enquiry will be known as “R” (Relayer) and the person responding to the enquiry by describing something about their hand “RR” (Relay Responder).