Relay Systems???
#1
Posted 2013-March-05, 09:04
Would somebody please explain to me why the bids are allowed but there is a general rule that they are not allowed?
Thank you
Mike
#2
Posted 2013-March-05, 09:18
AFAIK with "relay systems" it's meant systems where one player can ask his partner's entire hand. This is not the same as a "relay bid", which asks something specific (Stayman asks for a 4 card M and stops asking the rest of the hand)
#3
Posted 2013-March-05, 09:36
As Free said, it doesn't apply to simple conventions that include relays, only systems that are built around relays. But there are some systems where as soon as a game force is established (e.g. after a positive response to a big club), the next player can start a series of relays to get detailed information about his partner's hand. And these systems are usually allowed.
#4
Posted 2013-March-05, 09:49
#5
Posted 2013-March-05, 09:55
mdietz39, on 2013-March-05, 09:04, said:
Would somebody please explain to me why the bids are allowed but there is a general rule that they are not allowed?
Thank you
Mike
Stayman, Blackwood, Gerber, etc., are not relay bids per se. They are artificial bids that ask a specific question and request a specific response. Bids following the initial response to Blackwood and Gerber act as relay type bids, but they don't comprise a system of bidding. Furthermore, they are so entrenched in the tradition of contract bridge after having been around for the better part of a century that players do not think of them as relays. And lastly, Blackwood and Gerber occur primarily in uncontested auctions at a high level and are easily explainable and well known.
Relay systems are systems which are based primarily on a captain/servant relationship between the players. One partner (the "captain") makes a bid which begins a series of asks and tells. The captain makes a minimum bid at each turn to call which asks the other partner (the "servant") to describe his hand according to a predetermined set of responses. After the response, the captain may either make another minimum bid to request additional information, he may place the contract or he may make a natural bid to commence natural bidding.
Typically, one of the bids that begins a relay sequence is an opening bid of 1♣, which, in these systems, is the big bid - strong artificial and forcing. When opener does not have a strong forcing opening hand, he opens something else - 1 of a suit (other than clubs) or 1NT - and responder can start a relay sequence, typically by responding 2♣, which is strong, artificial and forcing.
In well designed relay systems, the most common answers to the relay bids usually are the next bid or the bid above that. So, in many relay sequences, it will appear that the bidding side is using every available bid to ask and answer questions. It is not uncommon for a relay sequence to go something like this:
1♣ - 1♦
1♥ - 1♠
1NT - 2♣
etc.
Where 1♣ is strong, forcing and artificial, the 1♦ response shows some particular type of hand (typically a negative response, but not always), 1♥ by opener is a relay asking for more information, 1♠ shows some rather common distributional pattern, 1NT says tell me more, and 2♣ would either be additional high card information or additional distributional information.
At some point in the auction, the captain knows all that is needed to be known about servant's hand and the captain places the final contract. The strong hand may also "break the relay chain" by making some bid other than the next higher bid, which typically starts natural bidding. Opener's minimum strength is indicated by how many relay bids were made (the more relays made, the stronger the hand), and that would force the auction to a particular level.
The result of a relay sequence is that a great deal of information is known about one hand but virtually nothing is known about the other hand.
Why are relay systems not allowed? Probably due to their rarity, players' unfamiliarity in dealing with them, the amount of time that would have to be expended to properly explain them, and, in general, most players' dislike of such methods. Furthermore, given the time constrains on most events (typically pair events), the fact that one pair that was one of the earliest practitioners of these methods - Rubin-Granovetter - was known for taking exceedingly long periods of time to conduct their relay auctions, did not endear themselves to tournament officials. I believe that they were suspended for a period of time due to continued slow play in major events.
Even the best players who play relay methods have to acknowledge that there is a great deal of memory strain in using such methods, as the answers to relays are not intuitive. They have to be memorized, and many responses do not come up often enough to be easily remembered. This causes delays in auctions and slow play. It can also create problems when errors are made or incorrect explanations are given.
#6
Posted 2013-March-05, 10:20
Quote
such methods, as the answers to relays are not intuitive. They have to be memorized, and many responses do not come
up often enough to be easily remembered.
I used to play relay methods pretty regularly.
I think that the memory load is significantly less that playing normal systems.
I suspect that most players with experience with relay methods would agree.
Well designed relay systems are based on a small number of basic principles that are applied in a consistent manner.
Once you understand the basics of the system, the memory load isn't bad at all.
#7
Posted 2013-March-05, 10:41
Basically a relay sequence is a series of bids where one hand is not described at all, just making minimum steps as "tell me more" while the other hand describes. To qualify as a relay system, there must be a relay sequence of at least two bids (so a single relay is actually not a problem) and the relays must start prior to opener's rebid. That basically means that responder's initial response to the opening bid must be the first relay. Since a relay response is by definition artificial and wide ranging (remember it's not supposed to significantly describe relayers hand) most such calls are already forbidden on the general chart. Further, relay sequences starting with an opening of 1NT or above are allowed, so following up stayman with another asking bid (for example) is okay.
The only relay I can think of that's allowed by the general chart opposite an opening below 1NT is the 1NT forcing response to 1M. So this rule would forbid a method where the auction goes 1M-1NT(forcing)-some rebid-minimum step as "tell me more." There are a lot more examples of relays that you can play on the mid-chart of course.
It's important to recognize that while some strong club players do use sequences of relays after their 1♣ opening, the first two calls in that auction are actually quite informative (the 1♣ opening showing significant values, and the direct response) so the relay sequence really starts at opener's rebid. I have had confirmation from many directors that such methods are allowed on the general chart, as are methods where after a 1M opening the 2♣ response is an artificial game force, and opener makes some descriptive rebid and then relays start after that.
a.k.a. Appeal Without Merit
#8
Posted 2013-March-05, 10:59
hrothgar, on 2013-March-05, 10:20, said:
I think that the memory load is significantly less that playing normal systems.
I suspect that most players with experience with relay methods would agree.
+1 -- and IMO, the mnemonic ease is thanks in no small part to the invention of symmetric relays by this brilliant physicist who recently won the Albert Einstein prize for his work on rotating black holes:
Roy Kerr
#9
Posted 2013-March-05, 14:15
As for tv, screw it. You aren't missing anything. -- Ken Berg
I have come to realise it is futile to expect or hope a regular club game will be run in accordance with the laws. -- Jillybean
#10
Posted 2013-March-06, 06:40
ArtK78, on 2013-March-05, 09:55, said:
That depends a lot on the construction of the relay responses. If they are symmetric and use steps to show similar hand types, the memory burden is extremely light and answers are in fact intuitive! Similar situations occur often enough, so they can easily be remembered.
A simple example proving my point. Once a 2-suiter has been shown, you use the following steps:
step 1 = lowest suit is longer
step 2 = 5-5
step 3 = highest suit is longer, shortage in the highest remaining suit
step 4 = highest suit is longer, no shortage
step 5 = highest suit is longer, shortage in the lowest remaining suit, 5431 distribution
step 6 = highest suit is longer, shortage in the lowest remaining suit, 6421 distribution
step 7 = highest suit is longer, shortage in the lowest remaining suit, 6430 distribution
step 8 = highest suit is longer, shortage in the lowest remaining suit, 7420 distribution
As you can see, this is perhaps a long list, but it contains logic and symmetry. If you respond step 1, captain relays and the rest is similar to the original step 3 and up. Same for step 3: captain relays and you continue to show exact shape like the original step 5 and up.
So if we've shown ♦+♥ at 2♦, or we've shown ♠+♣, it's all the same:
2♥ = longer ♦ (resp. longer ♣)
2♠ = 5-5
2NT = longer ♥, short ♠ (resp. longer ♠, short ♥)
3♣ = 2=5=4=2 (resp. 5=2=2=4)
3♦ = 3=5=4=1 (resp. 5=3=1=4)
...
Do you even realise how many exact distributions are summarized in this list of 8 steps?
The only thing that requires memory is knowing how to show the basic hand type (single suited with the known suit, 2-suited with the 2 suits known, 3-suited with the shortness known, or balanced).
However, in the early days, I assume that symmetric relays didn't exist, that the systems didn't necessarily work in step, and when you have to remember each and every bid in the system, it's obvious that you'll lose a lot of time.
#11
Posted 2013-March-06, 09:35
Free, on 2013-March-06, 06:40, said:
I use a similar but different set of steps that achieves the same effect:
step 1 (typically 2♥) = 4 of higher suit, 5+ of lower suit
step 2 (2♠) = 5+ in both suits
step 3 (2NT) = 5 of higher suit, 4 of lower suit (5413/5422/5431)
step 4 (3♣) = 6 of higher suit, 4 of lower suit, 0-1 in higher fragment (6403/6412)
step 5 (3♦) = 6 of higher suit, 4 of lower suit, 2 in higher fragment (6421)
step 6 (3♥) = 6 of higher suit, 4 of lower suit, 3 in higher fragment (6430)
step 7 (3♠) = 7 of higher suit, 4 of lower suit, 1 in higher fragment (7411)
I find this division far more intuitive than symmetric relay. The point is there though; once you have been using a relay structure for a while it becomes absolutely intuitive. Even better, exactly the same structure can be carried over to some extent to the 5-5s as is used above for 5-5s. Hence
step 1 (3♣) = 5 of higher suit, 5 of lower suit, 0-1 in higher fragment (5503/5512)
step 2 (3♦) = 5 of higher suit, 5 of lower suit, 2 in higher fragment (5521)
step 3 (3♥) = 5 of higher suit, 5 of lower suit, 3 in higher fragment (5530)
It is this constant re-use of patterns that keeps memory strain down. It is considerably easier than learning gadget add-ons to a Standard system such as Gazilli, for example. In this respect, it makes little difference if you are playing a relay method or natural. What matters is how many exceptions to the normal rules there are and how easy they are to forget (3NT as an exception is more likely to be missed than 4♣!) in combination with familiarity. Yes, relays have a relatively high initial memory load because of that lack of familiarity, or rather there is a high load on the Relayer who has to process all of the information to make the correct decisions, but it goes very quickly. That lack of familiarity is also a problem for normal club players too. They do not understand them and some do not therefore want to play against them. And that is all the excuse the ACBL needs to ban them (in competitions where noone that matters will be playing anyway).
#12
Posted 2013-March-06, 10:07
Zelandakh, on 2013-March-06, 09:35, said:
Zelandakh has brought up an interesting point:
Its difficult to adopt a relay method piecemeal.
We can't really specify a relay scheme for two suit hands and then go out and play, praying that we don't get dealt a three suiter, a balanced hand, what not.
In contrast, the memory load for adopting any one convention is probably less than for a complete relay system and these can often be incorporated piecemeal.
#13
Posted 2013-March-06, 10:46
It is probably worth clarifying what a relay bid is, as many get mixed up with the similar terms "transfer", "relay" and "puppet". A transfer is a request to bid a suit, typically the next suit up, and implies a holding in that suit. A relay is a cheap bid to say "tell me more about your hand", where the other hand has a choice of bids he can make. A puppet is a request to bid a suit, usually the next step, and implies nothing at all about that suit, but keeps the bidding low so that the puppet bidder can describe his hand.
What makes it confusing is that even common convention names get it wrong. "Puppet stayman" over 2NT for example, uses 3♣ as a relay, not a puppet. Many people describe a Lebensohl 2NT bid as " a transfer to clubs", when in fact it is a puppet to clubs, and the 2NT bidder could well have a different suit. (If the opposition describe one of their bids as a transfer, best ask them what it means, if it affects your bid.)
#14
Posted 2013-March-06, 11:55
fromageGB, on 2013-March-06, 10:46, said:
It is probably worth clarifying what a relay bid is, as many get mixed up with the similar terms "transfer", "relay" and "puppet". A transfer is a request to bid a suit, typically the next suit up, and implies a holding in that suit. A relay is a cheap bid to say "tell me more about your hand", where the other hand has a choice of bids he can make. A puppet is a request to bid a suit, usually the next step, and implies nothing at all about that suit, but keeps the bidding low so that the puppet bidder can describe his hand.
I had always assumed that the difference between a puppet and a transfer is that
1. A puppet forces partner to bid the next higher suit
2. A transfer suggests bidding a (specified) higher suit, but responder can bid something else
#15
Posted 2013-March-06, 12:03
hrothgar, on 2013-March-06, 11:55, said:
Yes, of course partner can make a transfer break, but this is rare and normally the transfer is completed. Some people never transfer break. The point is that the transfer implies a holding in a specific suit. A puppet doesn't.
#16
Posted 2013-March-06, 18:10
Again thank you all for you time and attention to this to the benefit of a complete stranger and beginner.
Mike
#17
Posted 2013-March-06, 19:01
#18
Posted 2013-March-06, 19:45
glen, on 2013-March-06, 19:01, said:
As usual, Glen recommends lying about agreements to get an edge...
Still kinda disappointing, no matter how often he does so.
#19
Posted 2013-March-06, 19:56
hrothgar, on 2013-March-06, 19:45, said:
Still kinda disappointing, no matter how often he does so.
Typical shot from this person:
"as usual" - implies all the time
"lying" - as if describing it as an asking bid is lying
"an edge" - to avoid unknowledgeable TDs
"no matter how often he does so" - repeats shot
this is the type of post that needs moderation, and a ban. It is a clear personal attack, and not only harms me, but harms the reputation of BBO forums for permitting it.
#20
Posted 2013-March-07, 01:40
Free, on 2013-March-05, 09:18, said:
AFAIK with "relay systems" it's meant systems where one player can ask his partner's entire hand. This is not the same as a "relay bid", which asks something specific (Stayman asks for a 4 card M and stops asking the rest of the hand)
hrothgar, on 2013-March-05, 10:20, said:
I suspect that most players with experience with relay methods would agree. Well designed relay systems are based on a small number of basic principles that are applied in a consistent manner. Once you understand the basics of the system, the memory load isn't bad at all.
glen, on 2013-March-06, 19:56, said: