Dipbridge on Bridge Systems

November 3, 2008

DIP: 2D “Jammer” in 3rd Seat

Filed under: Bridge, DIP, Opening Bids — dipbridge @ 8:40 pm

Hmm.

I’ve been doing some simulations, and, discretion’n'valour and all that, I’ve provisionally decided to amend it so that this particular Jammer instance shows 4-5D rather than 3-5D.

Playing it as 3-5D had the power to be immensely destructive, but hard to handle (i.e. finding our best fit) if we’re in trouble.  This suggests to me that a 3-5D treatment might work OK as a “bunny-bashing” device, but maybe not so hot against decent and organised opposition.  I’ve always preferred to design my systems to face the latter category, hence the choice.

I am also conferring with SD as to whether there is any merit, in 4th seat in particular (and maybe even 3rd seat, though it would lose the above Jammer) as playing 2D as simply 11.0 to 16.8 6+D rather than 14.0 to 16.8 6+D good suit.  This is because the 1D opening itself is now fairly nebulous in 3rd/4th seat.  I am awaiting SD’s considered response on this one:  it is normally worth the wait.

In the meantime, I’ll adjust the system page to reflect the amendment to the 3rd seat 2D opening.

Regards, DipBridge

November 2, 2008

DIP: Opening Bid Frequency

Filed under: Bridge, DIP, Opening Bids — dipbridge @ 1:35 pm

A first for DipBridge:  two blogs in one day!

Fleshing out the weak range opening bids (excluding, for the present, 2NT and higher pre-emptive openings) allows me now to do a fairly accurate (subject to my programming errors and therefore subject to revision) analysis of the frequency of opening bids.  Thomas Andrews deal programme has been invaluable thus far for this purpose.

Here is the output, based on 1,000,000 deals, for DIP 1st/2nd seat as currently described.

Pass percentage = 42.16
1C percentage = 11.48
1D percentage = 12.98
1H percentage = 11.22
1S percentage = 2.75
1NT percentage = 4.44
2C percentage = 2.39
2D percentage = 4.83
2H percentage = 5.78
2S percentage = 1.96
Total percentage = 100.00

Having done that, I then looked at the effect of shading the opening requirements for all bids by 1 fpc (i.e. playing a 14+ “Strong” 1C and a 9.0 to 13.8 fpc intermediate range.  The figures change to those below

Pass = 34.03%
1C = 16.79%
1D = 14.52%
1H = 12.39%
1S = 3.05%
1NT = 5.33%
2C = 2.62%
2D = 4.33%
2H = 5.16%
2S = 1.78%
Total = 100.00%

I actually prefer this balance, so for those jurisdictions which allow it, for anyone fancying playing DIP, I recommend switching to to the above ranges.  Dropping a further point down (a 13+ “Strong” 1C) makes too many hands open 1C, in my opinion, so I wouldn’t recommend it.  If you do switch to a 14+ 1C base, I would also suggest upping your 3rd/4th seat openings by 1 fpc, i.e. an 18+ 1C base.  You can probably follow the logic of this if you think about it a bit.

I also checked the other way:  adding 1 fpc to all the openings and making a 16+ 1C base.  I am aware of jurisdictions who won’t allow a 15+ 1C. It produces the following figures

Pass = 50.82%
1C = 7.41%
1D = 11.06%
1H = 9.44%
1S = 2.27%
1NT = 3.65%
2C = 1.98%
2D = 5.16%
2H = 6.11%
2S = 2.09%
Total = 100.00%

It’s not aggressive enough for my liking:  too many hands passing, and even the ones that open are not very obstructive.  I’m sure the method would function fine, but I wouldn’t play this range if I didn’t have to.  Also, if you do go to a 16+ 1C, don’t change the strong and intermediate ranges in 3rd/4th seat.  Same logic as above.

Regards, DipBridge

DIP: Weak Range Opening Bids

Filed under: Bridge, DIP, Opening Bids — dipbridge @ 11:00 am

Continuing my diversion from fleshing out the 1NT opening responses, it seems a good time now to begin to describe the weak range openings.  One of the reasons I feel able to do this now is that I am able to conduct some useful simulations using Andrews’ deal software to see the potential effectiveness.  However, for the 1st and 2nd seat weak range openings, my ideas haven’t altered for some time.

The 1st/2nd seat 2S opening is the simplest.  It is a sort of weak two, 6.0 to 9.8 fpc, except

  • It guarantees exactly 5 spades
  • It guarantees at least 2 cards in the other three suits
  • It cannot contain 4 hearts

The above conditions should ensure the obvious

  • Spade fit can be accurately gauged
  • 4-4 heart fits shouldn’t be missed
  • Escaping from when in trouble is relatively easy (except, of course, with 1444)

The 1st/2nd seat 2D and 2H openings are inspired by a (May 2005 I think) Bridge World article by Doug Bennion.  Some 20 years ago, in pairs, I played a somewhat similar method (2D = any 3-suiter, including 5431’s) but this was before assumed-fit theory had even been thought of.  Here is a link to where you can find a copy of that original article now

http://www.bridgebuff.com/jammer.html

Once you have read this article, you will know what I mean when I say that

  • 2D::  Long pointed, short round, Jammer, 6.0 to 9.8 fpc
  • 2H::  Long round, short pointed, Jammer, 6.0 to 9.8 fpc

An additional constraint for each of the above is that they show at least four cards in the guaranteed major.  This (with the fairly tight range) ensures the openings serve a constructive as well as pre-emptive purpose.  There are some other subtle differences (for example 5530’s are allowed in DIP’s version).  So, to clarify the shape possibilities above

  • 2D:: 4-5S, 3-5D, 3-5H & 0-2C OR 4-5S, 3-5D, 3-5C & 0-2H
  • 2H:: 4-5H, 3-5C, 3-5S & 0-2D OR 4-5H, 3-5C, 3-5D & 0-2S

In a recent blog, I mused that the 3rd seat openings may remain the same as the 1st/2nd seat ones (we have discussed the 4th seat ones previously).  I have changed my mind on this – possibly aided by the ability to run some simulations.  In 3rd seat, then

  • 2D::  Long diamond Jammer, 0.0 to 10.8 fpc:  Note:  3+, not 4+ diamonds, guaranteed
  • 2H::  “Weak Two” H, 0.0 to 10.8 fpc:  shapes similar to 1st/2nd seat 2S except 6H322’s also allowed
  • 2S::  “Weak Two” S, 0.0 to 10.8 fpc:  shapes as per 1st/2nd seat 2S except 6S322’s also allowed and the prohibition against exactly 5422 removed

As you can see, we are playing somewhat faster and looser in 3rd seat now, but some key structural integrity points remain

  • Escaping from the Weak Two’s, if in trouble, needs to be supported, so the 2+ cards in side-suits rule needs to remain, and
  • The 2D opening guarantees 3+ diamonds.  This is important IMO – I don’t like giving the opponents two shots when I am showing a weak hand type.  Having 2D showing D “length” prevents this.

I hope to update the system front page shortly.

Regards, DipBridge

October 30, 2008

DIP: 3rd & 4th Seat Opening Bids

Filed under: Bridge, DIP, Opening Bids — dipbridge @ 7:19 pm

While I prevaricate for a time on valuation methods and stronger responses to the 1NT opening, it seems an ideal opportunity to discuss and outline 3rd and 4th seat strong and intermediate range openings.

In DIP, 4 card major suits openings are emphasised in 1st and 2nd seat.  The as yet undiscussed weak range 2D and 2H openings also emphasise 4 card majors as well.  So, once you get around to opening in 3rd and 4th seat, it is assumed that many possible 4-4 fits with moderate strength or better between the two hands have already been discounted.  Remember, there are some balanced hands in the 10.0 to 11.8 fpc range which will have been passed, but if they contain a major, they can only be 4M333.

The number of hands which would naturally need to relay after an intermediate range opening has also effectively been reduced to nil.  As such, the benefits of transfer style openings in the majors has also largely been lost.  Further, because partner would have opened the majority of 10.0+ fpc count hands, there is little reason to stretch to open.  So somewhat antithetically to normal methods, DIP practitioners should go more conservative in 3rd/4th seat intermediate range openings than “normal” bidders.

With respect to the strong range, there are not too many hands which could easily FG relay opposite a 15.0+ fpc opening.  Whilst there are quite a few semi-POS possibles, there are fewer competitive advantages to be gained from a semi-POS structure, with at least one opponent and possible both already having been silent.  This suggests a possible switch to a more classic 1D ART & NEG, 1H+ FG POS relay structure, opposite a slightly stronger 1C opening (to bring more of responders potential hands into the FG POS picture).

Where does all the above lead to?  A 17.0+ fpc strong 1C opening and mostly natural 11.0 to 16.8 fpc intermediate strength openings.  Also, it is not incumbent to open 11.0 to 12.8 fpc hands if you don’t like the feel of them (which would likely mean in practice not opening if forced to show a bad suit, or to open 1D if you don’t have many of them, 4405 being the worst case scenario).

With 13.0+ fpc, you should always open.  Remember, the “best” hand that partner can come up with after having passed is an 11.8 fpc 5m4m22, so you shouldn’t miss a game if you pass a “bad” 12.8 fpc or worse in 3rd or 4th seat.

Speaking of “bad”, Thomas Andrews describes the ranges between points as follows

  • 12.2 fpc::  “a good 12 count”
  • 12.4 fpc::  “a very good 12 count”
  • 12.6 fpc::  “a very bad 13 count”
  • 12.8 fpc:: “a bad 13 count”

This seems a not unreasonable explanation method to adopt, when needed.

There are a few other residual points to consider for the intermediate range openings.  The first is the 1NT opening.  11.0 to 16.8 is clearly too wide a range for this.  As such, BAL hands are split into two ranges:

  • 11.0 to 13.8 fpc
  • 14.0 to 16.8 fpc

For 3rd and 4th seat openings, one of these BAL hand ranges is opened 1NT and one is slotted into the now slightly catch-all 1D opening.  There are two conflicting theories as to how one might allocate the above.

  1. If the opponents are known to be going to compete, it would be better to have the stronger range in 1D:  then partner with diamonds can act, realistically hoping for one of (i) genuine diamonds, or (ii) a strong(ish) hand. However
  2. If the opponents might double you for penalties, it would be desirable to have 1NT containing the stronger range:  this reduces the chance of getting in serious trouble.

As is often the case, the solution is a compromise, though a fairly optimal one I believe.  In 3rd seat (when the 4th hand opponent might still be very strong) 1NT gets the 14.0 to 16.8 range and 1D defaults to the 11.0 to 13.8 range.  In 4th seat, when both opponents are known not to be very strong, we try to capture the auction, with 1NT instead getting the 11.0 to 13.8 fpc range.  We put the stronger range into the 1D opening, to get the potential advantages if the opponents compete as discussed above.

Another residual point to consider is what to do with 4D5C hands.  In 1st and 2nd seat, these are put into the 2C opening.  Many other MOSCITO-like relayers would put these in with 1S(=D) but I think that cripples the 1S opening in competition:  4+ diamonds with a median of 6 diamonds is too great a range to handle.  By doing what DIP does, the minor suit showing openings average about 5.8 cards in the shown suit, which I think is quite workable.  However, with the ability to open 1D “sort of” showing diamonds in 3rd/4th seat, I think the balance changes considerably.  So, in 3rd and 4th seat, DIP opens 5C4D hands that are unsuitable for 1NT with 1D.  Consequently, 2C guarantees 6+ clubs, which is of course another good thing in competition.

The final residual point to consider is what to do with the two level pre-empts in 4th seat.  The answer is somewhat like most systems:  don’t have them.  The question is, what to use them for, or more accurately, how to use them to better define other bids.  The current proposition is to use 2D, 2H and 2S to show 14.0 to 16.8 fpc hands with good suits (probably KQxxxx or better).  This means that a same suit rebid of 2D, 2H or 2S either is minimum or doesn’t contain a good suit.  This should stop responder pushing too hard.  As always, there is a conflicting alternative:  play them as minimums, somewhat like Strong WK 2’s (for competitive reasons).  However, in 4th seat, the constructive use is preferred, so strong with good suit it is!

There is a case for making 2D, 2H and 2S similarly constructive in 3rd seat.  However, there is a competing case for maintaining pre-emptive weaponry in 3rd seat as well and currently, that case is deemed more compelling.

Please read the front page which contains the system for the net effect of all the above.  I’ll updating it fairly soon.

Regards, DipBridge.

October 28, 2008

DIP: 0.2.2

Filed under: Bridge, DIP, Opening Bids — dipbridge @ 12:39 pm

Cogitare, Constituere, Conficere!

Below is the new front page for DIP.  It will contain the system and ideas contained that have been discussed in the blog.  The other posts in the blog will be in reverse chronological order, so the most recent post is the next one down.

Openings Bids (1st/2nd Seat)

  • 1::  ART, STR, 15.0+ fpc
  • 1::  4+, 10.0 to 14.8 fpc:  inc 45m but not 46+m, inc 44, not < unless 454m, not 3433
  • 1::  4+, 10.0 to 14.8 fpc:  inc 45m but not 46+m, inc 5+5+, >=, not 4333
  • 1::  5+, 10.0 to 14.8 fpc:  usually 6+ (unless 5+4+), inc 6+4M, not 2254
  • 1NT::  BAL, 12.0 to 14.8 fpc:  inc 5m4m22, no M unless 4M333
  • 2::  5+, 10.0 to 14.8 fpc:  usually 6+ (unless 5+4), inc 6+4M, not 2245
  • 2::  4+, 6.0 to 9.8 fpc:  long pointed, short round, “Jammer”:  4-5, 3-5, 3-5 RND, 0-2 O-RND
  • 2::  4+, 6.0 to 9.8 fpc:  long round, short pointed, “Jammer”, 4-5, 3-5, 3-5 PTD, 0-2 O-PTD
  • 2::  5, 6.0 to 9.8 fpc:  5332 or 54m22

Openings Bids (3rd Seat)

  • 1::  ART, STR, 17.0+ fpc
  • 1::  0+, 11.0 to 16.8 fpc:  inc 45, if BAL then 11.0 to 13.8 fpc
  • 1::  5+, 11.0 to 16.8 fpc: 
  • 1::  5+, 11.0 to 16.8 fpc: 
  • 1NT::  BAL, 14.0 to 16.8 fpc:  not 5M332, inc 5m422
  • 2::  6+, 11.0 to 16.8 fpc:  inc 6+4M
  • 2::  4+, 0.0 to 10.8 fpc:  long diamond “Jammer”, 4-5, 3-5X, 3-5Y, 0-2Z
  • 2::  5-6, 0.0 to 10.8 fpc:  6322, 5332 or 5422
  • 2::  5-6, 0.0 to 10.8 fpc:  6322, 5332 or 5422

Openings Bids (4th Seat)

  • 1::  ART, STR, 17.0+ fpc
  • 1::  0+, 11.0 to 16.8 fpc:  inc 45, if BAL then 14.0 to 16.8 fpc
  • 1::  5+, 11.0 to 16.8 fpc: 
  • 1::  5+, 11.0 to 16.8 fpc: 
  • 1NT::  BAL, 11.0 to 13.8 fpc:  not 5M332, inc 5m422
  • 2::  6+, 11.0 to 16.8 fpc:  inc 6+4X
  • 2::  6+, 14.0 to 16.8 fpc:  good suit, usually KQxxxx or better
  • 2::  6+, 14.0 to 16.8 fpc:  good suit, usually KQxxxx or better
  • 2::  6+, 14.0 to 16.8 fpc:  good suit, usually KQxxxx or better

Definitions & Abbreviations

Where possible, definitions will conform to WBF Abbreviations, a copy of which can be found here

http://www16.plala.or.jp/KOKO33/system/abbrevi.html

In addition, there are the following

  • fpc:  “Fifths Point Count”:  (A=4.0, K=2.8, Q=1.8, J=1, T=0.4), Note that fcp ≈ hcp = both 40 point decks
  • inc:  “Includes”:  normally meaning it includes the subsequent hand shape

Version Control Log

  • 0.2.2::  Added in the motto “cogitare, constituere, conficere”
  • 0.2.1:: Changed the 2 Jammer opening in 3rd seat to guarantee 4-5 (rather than 3-5)
  • 0.2.0::  Added in weak range openings up to 2
  • 0.1.0::  Added in 3rd and 4th seat strong and intermediate range openings as described in 30/10/2008 blog.
  • 0.0.0::  First version of system using “fpc”:  Earlier information from the blog also included.

October 1, 2008

DIP: A bidding system

Filed under: Bridge, DIP, Opening Bids — dipbridge @ 1:02 pm

[This blog used to be stuck to the front page.  It has now been allowed to drop to its natural position at the bottom of the blog, with the latest version of the system as described during the blog now stuck on the front page]

Hi to those interested.

This is about an as yet untried (though somewhat tested) bidding system. It’s working name is DIP. It is primarily being developed by me, with the help of one other significant person, but like all such things, stands on the shoulders of others. As this method is somewhat MOSCITO-like, those shoulders are predominantly Roy Kerr from the theoretical standpoint and Paul Marston from the practical.

I thought it would be useful writing the thinking down over a period of time. In part, it will document the thinking so I don’t lose it. In another part, I hope it will invite external critique (and perhaps even discussion if I’m lucky) which will hopefully be of benefit to me and others. But, to be clear, the side discussion of ideas is an added-value advantage. My end goal is the completion and refinement of DIP.

OK, to start with nomenclature. For strength ranges, I will be using Kit Woolsey’s PPC (Practical Point Count). This is A=4.5, K=3, Q=1.75, J = 0.75. This still gives a 40 point deck, so you can think HCP instead if you like. Whether you use ppc or hcp is not relevant overall.

Like most MOSCITO’ish systems, there are three opening strength ranges, which you might refer to as Strong, Intermediate and Weak. Today, I’ll just deal with Strong and Intermediate. In part, this is because in theory, you could simply take these, and graft on your own Weak structure if you don’t like DIP’s.

1st/2nd Seat Openings

1C: ART(ificial), 15+ ppc

1D: 4+H, 10 to 14 3/4 ppc: not H<S unless 5S/4H/4m, can be 4H/5m but not 4H/6+m, not 3=4=3=3

1H: 4+S, 10 to 14 3/4 ppc: S>=H, can be 4S/5m but not 4S/6+m, not 4=3=3=3

1S: 5+D, 10 to 14 3/4 ppc: 6+D unless 5D/4+C, can be 6+D/4M

1NT: BAL, no M unless 4M333, 12 to 14 3/4 ppc

2C: 5+C, 10 to 14 3/4 ppc: 6+C unless 5C/4D, can be 6+C/4M

In the Intermediate range, all 3/S’s (three-suiters) with H’s go through 1D. Primarily, this is because it makes the relay structure work a little better. 4H/4S hands are opened 1D(=H). 5S/5H hands are opened 1H(=S) mainly to facilitate a H showing rebid in competition.

I hope this first blog piques the interest of some.

Regards, Dipbridge

Blog at WordPress.com.