This blog provides supplementary thoughts and ideas to the site. If you haven't seen the main site, there is a lot there including the Martel and Rodwell interviews, photos, and articles. This blog is focused on advancing bridge theory by discussing the application of new ideas. All original content is copyright 2009 Glen Ashton.

Friday, November 09, 2007

I've switch to short RSS feeds - if you are using a feeder you only see the start of the post. To see the complete post click on an arrow provided, or if none, on the post date.
Playing with the house's money

In the last two rounds of the sectional pairs last night I took shots on the last 6 boards. These were not psyches but just anti-field positions, mostly playing notrump instead of investigating for suit contracts, or over-competing.

On a 8 top, our last 6 scores were: 2, 2, .5, 0, 6, 0.

3NT going down with 4M cold was awful on three boards, and going down 2 vulnerable in 3M instead of letting their 3m contract make was bad on two boards. The 6 was a 3M on a 5-2 that came in on a mis-defend. If I played right down-the-line, our scores would have likely been, based on the results of the other tables: 3.5, 3.5, 4.5, 4.5, 2.5 , 4

Thus 22.5 instead of 10.5. The final score then would have been 123 and 57% instead of 111 and 51%. 51% got 0 masterpoints, while 57% would have good 3rd in the section for .84 masterpoints, but nothing overall (bottom overall was 58%).

This was not exactly playing with the house's money, where heads you win, tails you break even, but it was close to it. I ended up throwing away .84 masterpoints. However let's examine if this is the day where 3NT makes, while 4M can't do better. The last 6 results could have been:
7, 2, 7, 0, 6, 7.

On two boards, the 2 and 0, the cows aren't coming home (the 3M over-compete doesn't work), but the other 4 boards are a charm.

The net result is 29, and a final score of 129.5, good for 3rd overall and 4.22 masterpoints. Thus I was risking .84 masterpoints (which I ended up losing for our partnership) to gain 4.22. Those are odds well worth taking.

In online tournaments, where one can see a running score, if you are just sitting around average heading in the final round, you can take your shots on improving the score. If they don't work, it costs nothing. If they work, you improve into the points. Tails you break even, heads you win.
In live bridge, like last night, you can take shots if you estimate well. The overall is where the glory is, so if one is running a game just above average, make some moves at getting in there. At worse, you cough up a fraction of a masterpoint. At best, you win the whole thing. Was that possible last night? Yes, 135.5 won the whole thing, so if we were really fortunately last night on 5 of the 6 boards (convert the 0 to a 7 to get 7, 2, 7, 7, 6, 7) we would have ended at 136. Thus making those coin flips could have paid off big time.

Take some risks at the end of an event if you need to do so to move into the overalls (or in an online tourney without overalls, to move into the points). It is a lose-little, gain-big gamble.

Wednesday, November 07, 2007

Here are some schemes that restructure the one level minor suit opening framework to incorporate a 1NT 10-12 opening. For a discussion on some of the related issues, see the entry just before this.

A) 1C Handles 15-17 Bal

1C: Cs or 15-17 balanced
1D: Ds or 13-14 balanced or 18-19 balanced

Analysis: Minor suit openings are both 2+. Straight-forward bidding sequences.

B) 1C Handles 16-18 Bal

1C: Cs or 16-18 balanced
1D: Ds or 13-15 balanced or 19-20 balanced

Analysis: Minor suit openings are both 2+. Straight-forward bidding sequences. 1D covers more range when minimum balanced. 1C is more bullet-proof.

C) 1C Handles 13-14 or 18-19 Bal

1C: Cs or 13-14 or 18-19 balanced
1D: Ds or 15-17 balanced

Analysis: Minor suit openings are both 2+. Straight-forward bidding sequences. 15-17 does not have the room that A has to unwind, while 13-14 gains room it will often not need.

D) 1C Handles 16-20 Bal

1C: Cs or 16-20 balanced
1D: Ds or 13-15 balanced

Analysis: Minor suit openings are both 2+. Straight-forward bidding sequences. 1D is only balanced when minimum.

E) Wide Range

Note: Some who play 10-12 still want the minor openings to be 3+. The consequence is wider range for 1NT rebids.

1C: Cs or 13-18 balanced with 3+Cs
1D: Ds or 13-18 balanced with 3+Cs
2C: Handles strong hands including 19-20 balanced (2NT is 21-22 balanced)

Complexity: 1m-1M--1NT is 13-16, 1m-1M--2NT is 17-18.

Analysis: Minor suit openings are both 3+. 1NT rebid is a bit too wide. Hard for opener to show particular balanced range on some sequences when the opps are into the bidding.

F) Narrow

Note: the concept here is to use narrow ranges for 1NT rebids thus less invites-and-pass contracts – either go to game or stay low

1C: Cs or 15-18 balanced
1D: Ds or 13-14 balanced
2C: Handles strong hands including 19-20 balanced (2NT is 21-22 balanced)

Complexity: If it goes 1C-1D, opener can use Kokish 1H rebid (showing Hs or balanced) to show 17-18 balanced – thus 1C-1D--1NT is 15-16.

Analysis: 2+ minors. NT rebids are two-point ranges, so less invites necessary.

G) Hybrid

Note: Somewhat based on ETM Victory system, the idea was that 1C would be either Cs, or big balanced, or some hybrid of hands that were both Cs+balanced and just less than big balanced

1C: Cs or 15-17 balanced or, 13-14 balanced with either 5Cs and/or just 2Ds, or 18-19 balanced not suitable for 1D opening
1D: Ds or 13-14 balanced with 3+Ds, or 18-19 balanced with 3+Ds

Complexity: uses 1C-1M--2D as two-way: D reverse or extra value raise to 2M. This allows 1C-1M--2M to be bid with the less than 15 hands only. 1C-1M--1NT is 15-17 or 5Cs and 13-14 (so decent playing value).

Analysis: 1D is 3+ instead of 2+. Some complexity.

H) System using the Fantoni-Nunes model

For whole system ideas, one can take the Fantoni-Nunes system and modify it slightly to incorporate the 10-12 NV 1NT and make it a little lighter.

It could be:

1C: 15+ Balanced or 13+ Cs
1D: 13+ Ds or 13-14 Balanced
1M: 5 or longer major, 13/14+ or 11-13 with both majors
1NT: 10-12
2C: 5+Cs, 9-12
2D: 5+Ds, 9-12
2M: 5 or longer major, 9-12, must have a singleton if just a five card suit
2NT: 21-22 Balanced
Weak/Mini Notrump Issues


This is a discussion on point range issues that comes up when playing 10-12 mini-notrumps and 12-14 weak notrumps. It will be worth reading for strong partnerships using one or both of these notrump ranges, and for those interested in the bidding theory involved. As the discussion quickly gets intricate it is not suitable for casual reading.


This discussion will use four special terms for point ranges. For the first two, it assumes if a game invite is made opposite a 15-17 "strong notrump", that there are some 16s that will accept the game invite, and some 16s that will not accept. This will be termed:

16n: 16 point hand that would not accept a game invite of a 15-17 notrump
16y: 16 point hand that would accept a game invite of a 15-17 notrump

The other two special terms will be for 13 point hands. The first of these, 13b, will be for a 13 point hand that is quite bad - either any 4-3-3-3 without lots of Aces, or any 4-4-3-2 (any 4 card suits) with lots of queens and jacks. The last term will be the converse of that, and we will define the term 13d, for decent 13, to cover all 13s not considered bad. Thus we have:

13b: Bad 13 point hand.
13d: Decent 13 point hand, 13s not including 13b.

The 1D-1NT Response

Playing the weak notrump, one runs into the situation where 1D is opened, and responder has a standard 1NT response, so: 1D-1NT showing no four card or longer major.

For opener to bid again over 1NT with a balanced hand, there are two thresholds to be considered: 2NT should have at least 23 points between the two players, and 3NT should have at least 25 points. These are just guidelines of course, and hands with good playing value should upgrade their "points".

If 1D-1NT shows 6-10, opener will invite with 15 in case of 10+15=25, but then might reach 2NT on just 21 (6+15). So the standard 6-10 range for the 1NT response is not playable when using weak notrump openings.

There are two solutions employed by weak notrump partnerships, and let's look at the more complex one first:

1) Make the 1D-1NT response on a narrow range, and bid 1 of a major (1D-1M) on some hands with just three cards in the major, a strategy endorsed by weak notrump expert Eric Kokish.

Some partnerships using this strategy make the 1D-1NT response 8-10. Thus with no major and less than 8, but a hand wanting to respond to 1D, responder improvises by bidding a 3 card major. Sometimes this will get to a dicey 4-3 major contract, but other times it will succeed in landing in 1NT.

Other partnerships using this strategy make the 1D-1NT response 5-7. Thus with 8-9, responder improvises by bidding a three card major, and then makes another bid. For example 1D-1M--2M-2NT in this style will be 8-9 and only 3 in the major. This style is complicated since opener's 2M rebid can be based on either 15-17 balanced with 4 in M, or 10-15 unbalanced with 4 in M, and if the latter, the partnership needs to figure out landing sequences such as 1D-1M--2M-2NT--3C/D.

2) Make the 1D-1NT response on 6-9.

Some partnerships, like longtime weak notrumpers Martel-Stansby, play 1D-1NT as 6-9. Now opposite the 1NT response, opener does not need to invite with 15. In addition opener can pass with 16n since it would pass the sequence 1NT(15-17)-2NT(8-9 invite) - that is if the partnership was playing strong notrumps a 16 count that would not accept an invite.
Playing 1D-1NT as 6-9, then some 16-17s (16y-17) will sometimes reach 2NT on 22 (16+6), but these will be good 16s.

However if 1NT becomes 5-9, the partnership will overreach to 2NT on 21 (16+5), so for this reason responder, with 5, must pass 1D instead of responding 1NT. Thus, when responder has 5, the partnership may languish in 1D as the contract, when the strong notrump folks are playing a better 1NT contract.

Why opening 1C works better for big balanced

Opening 1C works better than opening 1D with big balanced, since responder has a 1D bid available which can be bid as sort as a "waiting" bid with nothing else to bid. So some partnerships (such as Gordon-Kraft) play:

1C-1NT: 8-10
1C-1D: can be waiting with 5-7

Others (such as Martel-Stansby) play:

1C-1NT: 5-8
1C-1D: can be waiting with 9+

So the 1C opening together with selected use of the 1D response allows for responder to define point range to arrow levels, compared to the 1NT response to the 1D opening.

Mini-notrump Overview

When a partnership uses a mini-notrump (10-12 or so), there are additional issues to sort out for the notrump ranges. First there is the question of what range to play opener's rebid on the sequence 1D-1M--1NT.

Some sources, such as British expert David Stevenson (writing on newsgroup, say that the 1NT rebid can be played as 13-17. After opener bids 1NT to show 13-17, it is then suggested to play a range check for responder to ask for opener's strength within 13-17. The problem here is that responder will need to invite with flat 8-9 counts to reach games opposite the 16y-17 hands. However if 8-9 invites (via a range check), it will reach 2NT on many 21s and 22s (13+8, 13+9, 14+8), and 13-14 balanced opposite 8-9 will be a frequent occurrence.

Therefore a better range to play the 1D-1M--1NT rebid is 13-16n. In this case the flat 8-9s can safely pass, since at most opener will have a 16 count that would not accept a game invite (of a 15-17 notrump). In addition, by playing the 1NT opening as 10-13b, instead of 10-12, the rebid range can be tighten up a bit to be 13d-16n.

What to do with 16y-17

If the 1D-1M--1NT rebid is 13d-16n, then we must decide what to do with 16y-17. One option is to play the 2NT rebid as 16y-18, moving 19s into the 2NT opener or using some specialized techniques to open on the two level (e.g. the Mexican 2D opening). The first problem with this is not too serious: some 22s will be reached, but they will be only 6+16y, so a good 22.

The second problem is that responder cannot make much lighter responses to 1D than 6 points. For example if one responds 1M with a 5 count and a 5 card suit, the bidding may continue: 1D-1M--2NT showing 16y-18. Now responder can decide to pass, reaching 2NT on some 21s and 22s, or signoff in 3M at the three level, often with just a 5-2 fit and 21 or 22. In comparison, the strong notrump partnerships are just bidding these out as 1NT-transfer--2M, so the style of rebidding 2NT with 16y-17 will cost over time if the partnership responds lighter than 6.

The Short Club

The other option for handling 16y-17 is to move them all in the 1C opening. This means:

a) 1D-1M--1NT will be 13d-16n.
b) 18-19s bid as they are now (open 1m, rebid 2NT)
c) 16y-17 opens 1C, even if just 2Cs.

In this option, 1C-1M--1NT will be 13d-17, and if responder has a five card major a checkback will be used to find the range and support. To avoid getting too high with flat 8-9s, responder without a five card major and 8-9 will rebid 1D first, to find out more about opener's hand and range. Also note one could play 1C-3C as not a purely preemptive raise, but with some values, enough for opener to bid 3NT if 16y or better, and then this would allow 3C to be the response on just 5Cs.

Mr. Walsh

The use of 1C-1D on a variety of 8+ hands, as a waiting bid, combines elements of the Walsh style and the Montreal Relay (MR), and thus it will named "Mr. Walsh" (Mister Walsh or Montreal Relay Walsh). In this style:

a) hands that are not worth a game invite opposite 15-17 always show the major first, even with a longer minor.
b) hands that are worth a game invite, or just a bit better, usually about 8-10, bid 1D first
c) hands with a major and diamonds in the 11-12 or stronger range, can bid either 1M or 1D first.
d) 1C-1NT shows 5-7 (or poor 8), no four card major, a hand that wants to play in 1NT opposite 13d-17 - this 5-7 flat hand does not bid 1D even with 4 or 5 Ds, since will not have a rebid playing this style.

After 1C-1D (Mr. Walsh), opener rebids:

1H: Natural Hs, or balanced 16y-17 with or without Hs (this Hs or balanced is like Kokish over 2C).
1S: Natural 4Ss, either unbalanced, or balanced 13d-16n.
1NT: 13d-16n without a four card major.
Rest: As over 1C-1D in standard.

After 1C-1D(Mr. Walsh)--1H, responder rebids:

1S: 8+, 4Ss, forcing. Opener's 1NT rebid is now 13d-16n with no S fit. 2NT rebid is 16y-17.
1NT: 8-10, no four card major, not forcing. 2NT is now invitational.
2H: 8-10, 4Hs, not-forcing. Opener's 2NT rebid now shows 16y-17 without H support.
2S: "Fourth suit forcing" - artificial game force (often standard for expert partnerships).
Rest: as over 1C-1D--1H in standard, except 1C-1D--3H-3NT is 16y-17 without H support.

After 1C-1D(Mr. Walsh)—1S, responder rebids:

1NT: 8-10, not 4Ss, not forcing. 2NT is now invitational.2S: 8-10, 4Ss.
Rest: as over 1C-1D—1S in standard.

1m-1M--2M Recommendation

Playing this style, 1D-1M--2M is 13d-16n if balanced, since 1C is opened if 16y-17. However 1C-1M--2M, covers 13d-17 balanced with 4 support, but this wide range is not much of a problem as responder will not have the 8-9 hand if just 4 in M.

Minor Opening Load Balancing

If 1C handles 16y-17, the partnership has a number of options to harmonize, or "load balance" the hand types across the two minor suit openings. Here are four options:

a) For 13+ balanced, open 1C if 3-3 in the minors. Open 1D only if 4 or longer Ds, or 4-4-3-2 exactly. This option puts a lot of the load on 1C, but makes 1D very natural.
b) For 13+ balanced, open 1C if 3-3 in the minors or if 4-4-3-2 exactly (just 2Cs), and only open 1D if 4+Ds. This option puts the most load on 1C, but responder now knows 1D is always 4 or longer Ds.
c) For 13-16n balanced, open 1D if 4+Ds, or if 3-3 minors, or if exactly 4-4-3-2. This means responder knows the 1C opener is 4+Cs anytime they have less than 16y, but responder can only depend on the 1D on having 3Ds.
d) For 13-16n balanced, open the longer minor, or if both equal, the better minor. This keeps the best lead directional character of the openings, but might help the opponents lead the other minor when it is right for them.

Mr. Walsh using weak notrumps

The Mr. Walsh approach can also be used when playing weak notrumps, in order to provide flexibility in opening 1NT or not. The key points are:

a) 1NT shows (11)12-14, but 12s are optional.
b) With 13d-14, one can open either 1NT, 1C or 1D. With 4-4/4-3/3-4 in the majors, opener could decide to pick a minor suit opening to find a major fit.
c) The minor suit opening style and the rebid ranges stay the same as given above - for example 1D-1M--1NT is 13d-16n.

This approach permits the weak notrump to be used when a hand is notrump-oriented, but allows a suit opening to be made when the hand is about the majors, or if the hand is flawed for notrump play in some way.

One can also employ a variation style:
a) If holding 13-14 balanced with a four card major, one can open 1NT or 1 of a minor at opener's discretion (if no four card major, opener if 13-14 must open 1NT unlike the approach just above).
b) 1C-1D(Mr. Walsh)--1M is natural (1H is not two-way), and if balanced will be 13-14.
c) 1C-1D(Mr. Walsh)—1NT is 15-17, may have a four card major. Responder is to use Checkback Stayman to find a major fit if 8+ and a major.

This second approach eliminates the 1C-1D--1H two-way bid, but can only be used when playing the weak notrump, and not the mini-notrump, since for the mini-notrump the 1C and 1D openings have to cover the 13-14 balanced hands without a four card major.