[steve2005]

Why do relay systems use the order of bidding high shortage, equal/middle shortage, low shortage?
Is there an advantage over, low shortage, equal/middle shortage, high shortage?

[helene_t]

steve2005, on 2023-June-03, 16:12, said:

Why do relay systems use the order of bidding high shortage, equal/middle shortage, low shortage?
Is there an advantage over, low shortage, equal/middle shortage, high shortage?

With shortness in spades you probably need to play in a non-spade suit so you have less bidding space below the contract you will be signing off in.

When the suit is already agreed I don't think it matters.

Edit: it's probably more about that with shortness in a major you are more likely to belong in 3nt.

[DavidKok]

I really don't know enough to answer. I guess in theory we might decide on each auction whether it is more likely that we belong in 3NT or 4M, and then assign the low or high shortage to the bid that is least likely to end up being our final denomination (if that suit has not been bid on previous rounds) - i.e. bidding 2♠ to show spade shortage is usually fine, but bidding 2NT to show heart shortage less so (since if we have minor suit shortage and heart length there's a greater chance we belong in 4♥, so that hand type can marginally improve by bidding 2NT - unless we previously bid 1NT to show our two suits). The equal needs to be in the middle if you want to keep it symmetric. I think in practice it doesn't matter, and it's easier to have one rule cover all cases.

[Gilithin]

When using the later shape relays, relayer is typically either looking for a fit in a major fragment or has slam interest. If there is slam interest then it probably does not matter which order the suits are ranked in but for the more common case, having low shortage last allows for the safe use of zooming. This ends up being a small but useful efficiency gain for the overall system.

[foobar]

IMO, the H/M/L is just arbitrary, and I have seen a couple of systems that use L/M/H. It likely just comes down to what the partnership can remember.

[paulg]

foobar, on 2023-June-05, 11:30, said:

It likely just comes down to what the partnership can remember uses when they first start playing.

[Phil]

We always used HML. No particular reason, just ripped this fron Welland/Fallenius and what they did.

[mikeh]

steve2005, on 2023-June-03, 16:12, said:

Why do relay systems use the order of bidding high shortage, equal/middle shortage, low shortage?
Is there an advantage over, low shortage, equal/middle shortage, high shortage?

As far as I know, it’s all partnership agreement. I currently use NLMH, where N is none, in relays where replier hasn’t promised shortness and LMH when he has, and sometimes LH, when replier has shown two suits but not yet defined shortness

It’s not just in ‘relay systems’. Over 1S 2N (gf raise), 3C is any non horrible minimum and 3D asks, bringing to NLMH into play, as one example

[Gilithin]

mikeh, on 2023-June-06, 20:30, said:

As far as I know, it’s all partnership agreement. I currently use NLMH, where N is none, in relays where replier hasn’t promised shortness and LMH when he has, and sometimes LH, when replier has shown two suits but not yet defined shortness

It’s not just in ‘relay systems’. Over 1S 2N (gf raise), 3C is any non horrible minimum and 3D asks, bringing to NLMH into play, as one example

It is very different when you have a known fit though. Now zooming is not an issue. In symmetric relay, zooming past 3NT with high shortage can lead to difficulties.

[Kungsgeten]

When I started learning relays the mantra was "length bottom up, shortness top down". So when showing a suit, the first step is clubs etc and when showing shortness the first step is spades etc.

From a memory standpoint this is nice, because the rules mesh well together. Let's say opener has shown 5+H and 4D, and now we're in the short-legged two-suiter scheme:

Step 1: High shortness (short spades)
Step 2: Equal shortness (5422, or maybe 7411)
Step 3+: Low shortness (short clubs)

But instead of thinking "shortness top down" you could just as well think "length bottom up":

Step 1: Low length (clubs longer than spades)
Step 2: Equal length
Step 3+: High length (spades longer than clubs)

The rule becomes more useful when showing single-suiters, since it can be a bit hard to remember what to bid with various 6322 hands (length bottom up, so the first step is bid with three cards in the lowest ranking suit etc).

[nullve]

A major suit fit is less (more) likely after Teller has shown high (low) shortage. So it seems that 2N as high (low) shortage will wrong-side otherwise attractive notrump contracts more (less) often.

[foobar]

Gilithin, on 2023-June-06, 23:57, said:

It is very different when you have a known fit though. Now zooming is not an issue. In symmetric relay, zooming past 3NT with high shortage can lead to difficulties.

Isn't this random, in that the issue could have very easily been the other way around? For example, assuming a zoom with ♠ trump and 6♠430 with H/M/L:

• 4♣ (high ♥ shortness) -> Allows easy sign off in 4♠
  
• 4♥ (low ♣ shortness) -> Can't sign off in 4♠

If 1 and 2 are flipped, then L/M/H would have worked better/worse.

[Gilithin]

foobar, on 2023-June-07, 10:07, said:

Isn't this random, in that the issue could have very easily been the other way around? For example, assuming a zoom with ♠ trump and 6♠430 with H/M/L:

• 4♣ (high ♥ shortness) -> Allows easy sign off in 4♠
  
• 4♥ (low ♣ shortness) -> Can't sign off in 4♠

If 1 and 2 are flipped, then L/M/H would have worked better/worse.

You are missing the point. Say spades are shown and the other hand has a minimum GF with 5 hearts and no alternative fit. Typically relays will continue until the heart fragment has been resolved. In LMH this fragment will resolve more or less immediately, meaning that when a 3-5 fit is available there will be minimal information leakage. Unfortunately when there is no fit and the highest shape is held, zooming might take the response above 3NT, which might have been the least making contract. In HML, it often means that more shape is disclosed along with the 3-5 heart fit. Against that, the shape that zooms past 3NT will contain a heart fit, reducing the chances of going overboard. If you know where your fit is (in your case ♠) and are therefore continuing relays to decide on the right level, then yes, it probably makes no difference in the grand scheme of things. It does potentially matter when we are minimum and asking to find out about a fit in a major fragment.

[foobar]

Gilithin, on 2023-June-08, 00:27, said:

You are missing the point. Say spades are shown and the other hand has a minimum GF with 5 hearts and no alternative fit. Typically relays will continue until the heart fragment has been resolved. In LMH this fragment will resolve more or less immediately, meaning that when a 3-5 fit is available there will be minimal information leakage. Unfortunately when there is no fit and the highest shape is held, zooming might take the response above 3NT, which might have been the least making contract. In HML, it often means that more shape is disclosed along with the 3-5 heart fit. Against that, the shape that zooms past 3NT will contain a heart fit, reducing the chances of going overboard. If you know where your fit is (in your case ♠) and are therefore continuing relays to decide on the right level, then yes, it probably makes no difference in the grand scheme of things. It does potentially matter when we are minimum and asking to find out about a fit in a major fragment.

Not to belabor the point, but I am hard pressed to think of a single hand in the several hundred relay auctions that I have had over the years, where playing H/M/L felt like a distinct advantage. Perhaps, that underscores its superiority by implication .

[blackshoe]

Heh. I noticed the other day that Standard Modern Precision goes low to high and Santa Fe Precision goes high to low. Or maybe it's the other way 'round. Other than this, in many ways the two systems are identical. No idea which one is "better".

[pilun]

We have always shown high shortage first. There are two extra theoretical advantages.

When steps are lost, encroaching on 3NT can be a problem. Showing shortages top down can ease the anxiety. Take this:

1♣ - 1♦
1♥ - 2♠ = 5-7 balanced, no 4cM for us. (1♥ was 19+ any)

2N - 3♦ = 5♣s (3♣ = 5♦, 3♥ = 2-3-4-4, 3♠ = 3-2-4-4, 3N = 3-3-(43))

3♥ - 3♠ = 2-3-3-5, 3N = 3-2-3-5, 4♣ = 3-3-2-5

Here we are catering for opener having 5-cards in a major. High shortage first makes it safe to continue relay.

More common are cases when the quest for an 8-card major fit has failed. Then it's good to be able to switch to natural to check out stoppers, maybe a 5-2 or 4-3 major fit. In the early stages, we used to relay out everything. These days we are more than happy to break.
Basically, high shortage needs to be revealed early to allow asker to switch horses.

1♣ - 1♠ = hearts
1N - 2♣ = diamonds
2♦ - 2♥ = ♦>♥
2♠ - 2NT = HS

Opener has ♠AKxxx ♥Kxx ♦AQx ♣xx

With a spade fit ruled out, opener can break. We play 3♦/♥/♠ as natural, texture.


So 3♦ over 2NT. Partner will realise 3NT needs a club stopper. 3♥ over your 3♦ would show a chunky 4-carder, etc.

If your method uses 2NT to show LS, responder will bid 3♦ directly over 2♠ with 1-4-5-3 and opener will be none the wiser.

Incidentally, the same principle applies to using 2♥ as "reverser" (or canape). Some early relay pairs used 2♥ as "enhancer", meaning the first suit was longer. Again, hands with extra major length generally do not need to offer space for chain breaks.

[olien]

Phil, on 2023-June-05, 17:43, said:

We always used HML. No particular reason, just ripped this fron Welland/Fallenius and what they did.

Well, subconsciously, systems also tend to be designed with relay sequences tending to show suits "up the line," which also goes with what you're saying. Well, showing length low --> high meshes very nicely with showing shortage high --> low.

[cheapcheap]

I am not sure if the original relay designer have thought of this, but this is one disadvantage using LMH to show shortness:

Assume you have single suited spades. If you show shortness in a LMH manner, and the relay happens to start from 3♣, you will always be bidding at your short suit (3♣=♣ short, 3♦=♦ short, 3♥=♥ short). The opponents could then show length or direct the lead by doubling. While using HML scheme, you only bid 3♦ when ♦ is short.

HML reduces the chance you bid exactly at your shortness (compared to LMH).