[keylime]

During the holiday and birthday break, I was browsing through the notes of Ultra Club preparing for the regional in Williamsburg, VA coming up in about 10 days time.

In Ultra when patterning out we show our shortages in low-high-mid-none order, but I am starting to consider using high-low-mid-none order instead.

Does order really matter?

[awm]

There are a few instances where it matters. In particular:

(1) The various game contracts are at different levels. Sometimes you want to decide between 3NT if partner is short in your major versus four of your major if partner's shortness is elsewhere (and thus he has 2-3 cards in your major). This is more frequent when you're considering playing in a major suit (a lot of times 3NT is better than 5m anyway), so if one of high shortage and low shortage will take you past 3NT, it's better that it be low shortage (i.e. the hand with length in the side major bypasses 3NT).

(2) Hands with no shortage at all typically have less playing strength. For this reason it might be nice for "no shortage" to resolve lower than the other options as you're less likely to have slam.

[BillHiggin]

IMHO (which may be considered worthless!) it may be useful to pin 3N as always none or equal and then work either hi, lo, mid or lo, hi, mid (no real opinion on that) around 3N.

[kenrexford]

I like to work from high, then to mid, and end at low, but I usually get none.

[the hog]

In Moscito and while using Odworodtka in PC we have used H/M/L and L/M/H. Marston has also switched between both. There has been no advantage/dsiadvantage either way.

[dake50]

As long as Ultra. Does it matter to relay for short before 2nd suit?
I think knowing short leaves high hopes -- so keep relaying (or kills hope right now). Those 'wrong' 2nd suit hands won't even try whereas 'right' short eagers on.

The short is more promotable/demotable than 2nd suit so that info needed 1st.

[Free]

I usually played shortages high-mid-low. Point is that when you have a low shortage, you usually have a Major suit contract to play (so no problem of getting a bit high), while with a high shortness you might get to 3NT more often.

[keylime]

Here's an example to illustrate why the current order may not be optimal:

1C* - 1H*
1S* - 2C*
2D* - 2H*
2S*

Here's a typical sequence for us: strong club, 4+ spades G/F, waiting, any canape, which one, club canape of 5 clubs, and now asking for the remainder of the hand.

Responses: 2NT - showing 4-3-1-5, 3C - showing 4-1-3-5, and 3D being the 4225 hand. If I used a H-M-L I lose the ability to play in the third seat by implication, but if H-L-M, then I am still ahead of the curve so to speak.

With 5M and 4m handtypes over a forcing club, this is where the problem is starting to occur

1C - 1H
1S - 2D*
2H* - 2S* (5M, 4D)
2NT*

Over this, 3C becomes 5-3-4-1, which nominates the shortage directly. With H-M-L I think I have more flow in the structures hence the possible change.

[hrothgar]

keylime, on Jan 3 2008, 05:36 PM, said:

Here's an example to illustrate why the current order may not be optimal:

1C* - 1H*
1S* - 2C*
2D* - 2H*
2S*

Here's a typical sequence for us: strong club, 4+ spades G/F, waiting, any canape, which one, club canape of 5 clubs, and now asking for the remainder of the hand.

Responses: 2NT - showing 4-3-1-5, 3C - showing 4-1-3-5, and 3D being the 4225 hand. If I used a H-M-L I lose the ability to play in the third seat by implication, but if H-L-M, then I am still ahead of the curve so to speak.

With 5M and 4m handtypes over a forcing club, this is where the problem is starting to occur

1C - 1H
1S - 2D*
2H* - 2S* (5M, 4D)
2NT*

Over this, 3C becomes 5-3-4-1, which nominates the shortage directly. With H-M-L I think I have more flow in the structures hence the possible change.

Hi Dwayne:

I'm skeptical (to say the least) whether its a good idea to design homebrew relay systems. You're issue isn't the order in which you're resolving suits, but rather that the relay structure itself is incoherrent.

I think that you'd do much better to adopt a standard symmetric relay type module over your strong club opening. A lot of folks have spent a lot of effort working on Symmetric. I don't think that you'll do a better job reinventing the wheel.

Case, in point - Heres a standard symmtric scheme to describe two suiters with Spades and Diamonds

1♣ - 1♥ 1C = strong, 1♥ = 4+ Spades
1♠ - 2♣ 1S = relay, 2♣ = two suited with Spades and Diamonds
2♦ = 2♦ = relay

At this point in time

2♥ = 4 Hearts and 5+ Diamonds (2♠ is relay, after which you mirror 2N+)
2♠ = 5+ Spades, and 5+ Diamonds
2N = 4 Diamonds and 5+ Hearts, high shortage
3♣ = 2=5=4=2 shape
3♦ = 1=5=4=3 shape
3♥ = 1=6=4=2 shape

The entire scheme is completely symmetric. Any 5431 is always resolved at 3♦. Any 6421 is always resolved at 3♥.

This significantly improves efficiently and decreases memory load.

[akhare]

hrothgar, on Jan 3 2008, 10:33 AM, said:

keylime, on Jan 3 2008, 05:36 PM, said:

Here's an example to illustrate why the current order may not be optimal:

Hi Dwayne:

I'm skeptical (to say the least) whether its a good idea to design homebrew relay systems. You're issue isn't the order in which you're resolving suits, but rather that the relay structure itself is incoherrent.

I think that you'd do much better to adopt a standard symmetric relay type module over your strong club opening. A lot of folks have spent a lot of effort working on Symmetric. I don't think that you'll do a better job reinventing the wheel.

Second that -- using symmetric relay makes much more sense that trying to do it anew.

BTW, all the relay systems I have used High-Mid-Low -- maybe it's a symmetric thing?