Duplicate bridge with 4 players
Tanel Teinemaa
two teams match where the closed room results are taken from the table
taking the high card points and distribution into account. Table was
calculated on the basis of over 60,000 results from world class
players.
Take a look: http://www.compensationtable.com
Tanel Teinemaa
Stig Holmquist
<tein...@gmail.com> wrote:
>Any kind of questions and feedback are welcome.
I'm unable to understand how to use the compensation table.
Please give a speciffic example from a home game.
Stig Holmquist
Tanel Teinemaa
> On Tue, 24 Nov 2009 11:12:49 -0800 (PST), Tanel Teinemaa
>
> >Any kind of questions and feedback are welcome.
>
> I'm unable to understand how to use the compensation table.
> Please give a speciffic example from a home game.
>
> Stig Holmquist
Added example to compensationtable.com. If it is still unclear, please
let me know, what you didn't understand.
Example:
Deal 1: You are playing 4H and just make it. You are in not vulnerable
so you get 420. Your line (you and your partner) has 25 HCPs and 9-
card heart fit. Compensation Table expects 390 to you (25 HCPs, 9-card
major, NV). You gain 420-390=+30, this means 1 IMP for you.
Deal 2: You are playing 3D and you are going down one. You are in not
vulnerable so you lose 50. Your line has 19 HCPs, so opponents line
must have 21 HCPs. Opponents also have 8-card major fit and they are
in vulnerable. Compensation Table expects 30 to them (21HCPs, 8-card
major, V). They gain 50-30=+20, this means 1 IMP for them.
Deal 3: Your opponents are playing 2S and they just make it. They are
in vulnerable and they get 110. Both lines have 20 HCPs. Now it is
important who started bidding. Lets say your partner was dealer and
started bidding. That means your line should score. Your line only has
8-card minor fit and you are in vulnerable also. Compensation Table
expects -40 to you (20HCPs, 8-card minor, V). You gain (-110)-(-40)
=-70, this means 2 IMPs for your opponents.
Deal 4: You are playing 7H and you just make it. You are in vulnerable
so you get 2210. Your line has 35 HCPs and 11-card heart fit.
Compensation table expects 1610+2*150=1910 to you (33+2 HCPs, 10+ card
major, V). You gain 2210-1910=+300 so 7 IMPs for you.
Congratiulations, you won 4-board match with +1-1-2+7 = +5 IMPs.
dranon
<stigf...@hotmail.com> wrote:
>On Tue, 24 Nov 2009 11:12:49 -0800 (PST), Tanel Teinemaa
><tein...@gmail.com> wrote:
>
>>Any kind of questions and feedback are welcome.
>
>I'm unable to understand how to use the compensation table.
Why does this not surprise me?
Rob Morris
Not many people have replied to this, so I just thought I'd say that it
looks interesting, well done for putting in the effort.
It looks like the sort of thing that you'd need to play a session of, to
get a feel for how it works. Have you managed to persuade three friends
to play with you for an evening using the compensation table, and did
they feel it was worthwhile?
--
Rob Morris
arr emm four four five (at) cam dot ac dot uk
Alan Malloy
I don't especially care for the "N-card fit in suit X" part of the
table; if you're going to play compensation, just using HCP is good
enough. But it's fine, I don't really object to it. What I think is
silly is how you decide on the scoring side when each pair has 20 HCP.
Take the following deal:
AKxxxx
Tx
Axx
xx
- Tx
AJxxx Kxxx
KQxxx JT9x
xxx KQJ
QJxxx
Qx
x
Axxxx
NS can make 4S, EW can make 4H. Let's say everyone is vulnerable, and
the table reaches the par spot of 5HXW-1, for +200 NS. Now, what's the
compensation table say for this deal? It depends on who dealt! If North
dealt, then he opens the routine 1S, and becomes the scoring side, so
the compensation table says he should get 120. 120 from 200 is +80: 2
IMPs to NS. But if East, South, or West deals, then West will be first
to bid, and he becomes the scoring side. Now the compensation table says
HE should get +120, and he actually gets -200. 120+200 is 320, 8 IMPs to
NS. A difference of 8 IMPs based on who deals, even though the same
final contract is reached, is a pretty big deal. Under these rules it's
a disadvantage to open the bidding when your side has 20 HCP, which
can't be right - usually in real competition a more aggressive style is
rewarded, because it puts the opponents on defense immediately.
Perhaps instead of breaking the tie by opener, you should break it by
highest-ranking fit. Frankly that could be better than the "major
suit/minor suit" distinction in the table in other cases, but I'm not
sure exactly how to apply it. At least for the 20/20 case it's clear how
to fix the problem.
--
Cheers,
Alan (San Jose, California, USA)
Frisbieinstein
> if you're going to play compensation, just using HCP is good
> enough.
That's called Russian scoring and I've played it a lot. It has less
luck than rubber bridge and is a good idea. The trouble is that a
routine major game with 23 HCP makes 10 IMPs, while that tough 3NT
contract never gets you anything.
On the other hand I'm not interested in a lot of complicated tables
when playing for fun. I'd make a simple adjustment for suit
contracts, add two HCP or something, and leave it at that.
Stig Holmquist
<patmp...@gmail.com> wrote:
>On Nov 26, 10:34�am, Alan Malloy <alan.NO.S...@malloys.org> wrote:
>> if you're going to play compensation, just using HCP is good
>> enough.
>
>That's called Russian scoring and I've played it a lot. It has less
>luck than rubber bridge and is a good idea. The trouble is that a
>routine major game with 23 HCP makes 10 IMPs, while that tough 3NT
>contract never gets you anything.
Can Russsian scoring be used in rubber bridge, where you get points
for honors when playing Chicago 4-deals?
Stig
Frisbieinstein
> On Wed, 25 Nov 2009 18:58:33 -0800 (PST), Frisbieinstein
>
Yes, some play rubber bridge with Russian scoring as well. Two
separate scores are kept. The rubber winner is the lucky one, the
Russian winner is whomever played better (sort of). Then both sides
may win. Why not? It is possible to win the Russian scoring even
though ops are dealt an endless series of easy slams and games.
This sort of scoring is especially good online, where the computer
takes care of the details of the scoring.
Frisbieinstein
Hmmm, on second thought, what do they do when you already have part
score? If it were me I would say, "for rubber I pass here, for
Russian I bid on." Then rubber bridge is freed of these boring hands
where the only stake is the number of overtricks.
Richard Pavlicek
1. I tried Russian scoring (simpler than Tanel's version) ages ago
and found it was biased to the weaker side. For example, one would
much rather have a combined 14 HCP than 26, because the gain from
a set is huge, while the loss to a routine game is small. If you
consider that only about 65 percent of contracts are made (experts
assumed) the disparity is more apparent. Tanel's method is surely
an improvement, but the compensations are too high.
2. When we played Chicago bridge at home, we always used duplicate
scoring (honors don't count and partscores don't carry over) so the
practice would benefit tournament play. I also modified the scoring
to award 150 points for all partscores (e.g., bidding and making 2S
scored 210). This made lesser hands more meaningful and provided a
competitive arena similar to IMPs.
--
Richard Pavlicek
Web site: http://www.rpbridge.net
henry...@yahoo.com
If there are 4 people who want to play duplicated bridge, Doop is
still the best alternative, in my opinion.
Does anyone know if they are still producing Doop refills?
Henrysun909
Stig Holmquist
Baron & Barclay seems to have them
>Henrysun909
Tanel Teinemaa
> to play with you for an evening using the compensation table, and did
> they feel it was worthwhile?
I have to admit I have never used my Compensation Table yet. But I
have used older version of this table which is little bit less
accurate and it seemed fair.
> What I think is
> silly is how you decide on the scoring side when each pair has 20 HCP.
I agree it would be more accurate when highest ranking fit would
score. I thought about it too before making this table. But I wanted
it to make fast and easy to use. What if you have this rule that
higher ranking suit should score and then, if no-one have fit? You
still need additional rule for that situation. I also thought that
side who have more cards in spade suit should score, but it is still
more confusing and not as fast as my silly solution. After all, this
type of tables can't be perfectly accurate anyway, so boosting one
certain situation with many additional rules seemed silly for me.
> On the other hand I'm not interested in a lot of complicated tables
> when playing for fun. I'd make a simple adjustment for suit
> contracts, add two HCP or something, and leave it at that.
Your idea to add some HCPs to suit contract is not bad, but it makes
NT contract more attractive, because you get much more imps when you
make NT contract. In theory, expected result shouldn't come form
bidding and final contract. This table isn't actually too complicated,
because declarer should know how long fit they have anyway.
> This sort of scoring is especially good online, where the computer
> takes care of the details of the scoring.
True, it is good for online bridge, but not especially good, because
computer can get better results from other real rooms (like in Bridge
Base Online) or computer can play the deal (with GIBs for example) to
get more accurate result for that deal. And I really tried to design
it to be easy to use for humans.:P
> 1. I tried Russian scoring (simpler than Tanel's version) ages ago
> and found it was biased to the weaker side. For example, one would
> much rather have a combined 14 HCP than 26, because the gain from
> a set is huge, while the loss to a routine game is small. If you
> consider that only about 65 percent of contracts are made (experts
> assumed) the disparity is more apparent. Tanel's method is surely
> an improvement, but the compensations are too high.
Compensations are indeed a little bit high for non-expert players
because they are calculated from world class results. But with 26
HCPs, my scoring expects about 40 points less than Russian scoring
when no fit and about 10 points less if you have 8-card major fit.
Great thanks for feedback,
Tanel