[Cyberyeti]

helene_t, on 2011-September-22, 11:39, said:

Jack: that's called Patton. I agree it's a great form of scoring. There is a yearly Patton event at 't Onstein bridgeboerderij in Voorden, NL.

I like it too, is what I was describing in my earlier post.

If you want to bring Max Bavin out in a cold sweat, ask him about the time they tried to use it in the easter Guardian about 30 years ago with no computers on the tables and a load of LoLs trying to calculate aggregate/swing.

[gnasher]

Fluffy, on 2011-September-22, 03:17, said:

the big difference between bam and MPs is that 430 equals 420 at BAM, or at least that's what I though. (correct me if wrong)

That's common in parts of Europe, but not in North America. There's a thread about it here:

http://www.bridgebas...-board-a-match/

[han]

Vorden.

[JLOGIC]

Just pretend like it's MP. If you never try to predict the other room, you will probably be ahead of most other people who vary their bids/plays by trying to predict the room. It's like how most card counters in blackjack make errors so casinos don't care.

More on how BAM and MP are very similar: Say you have a normal/good average plus board that you would expect 75 % on in MP. In BAM, you might have a 50 % chance to win the board, and 50 % to tie it. It is the same thing.

[barmar]

semeai, on 2011-September-22, 11:24, said:

I'm happy if you're saying all this because of the difference between your teammates and the generic member of a matchpoint field.

If not, I disagree. If we suppose x% of the field will get 150, y% will get 180, z% will get 200, and w% will get 210 (and ignore the rest, as they're irrelevant; for simplicity let's suppose no 160/170 etc), and we know nothing about your teammates, we should also suppose, as Helene said, that your teammates have a probability (in percentage form for clarity) of x% of getting 150, y% of getting 180, z% of getting 200, w% of getting 210.

If you're considering trading 180 for probability p of 210 and (1-p) of 150, the difference in your EV is the same at either matchpoints or BAM. In both cases it's p*(w/2 + z + y/2) - (1-p)*(y/2 + x/2).

Added: If you're worried the matchpoint field isn't large enough for certain results to appear, more of the same sort of analysis will again show the EV is the same (show BAM and your matchpoint event are both the same in terms of EV as some idealized really big matchpoint event).

What you're basically saying is that a strategy that averages 60% on all the boards in MP is the same as one that wins 60% of the boards in BAM, but this assumes that your teammates are equivalent to a matchpoint field. If you think your teammates are better than the field, you can assume that they're not giving away the gifts you're receiving. Then when you get a gift, you now have the opportunity to take a risk that's likely to at worst give it back (only risking half a board) but which might turn a loss into a win if something weird happened at the other table.

On the other hand, if you think your teammates are worse than the field, you should play carefully -- you can't afford to give gifts at both tables.

[helene_t]

barmar, on 2011-September-23, 10:33, said:

What you're basically saying is that a strategy that averages 60% on all the boards in MP is the same as one that wins 60% of the boards in BAM, but this assumes that your teammates are equivalent to a matchpoint field.

Exactly, that is semeai's point. Semeai said "If not[i.e. if there is no difference between your teammates and the generic ...], I disagree.".

[Stephen Tu]

barmar, on 2011-September-23, 10:33, said:

On the other hand, if you think your teammates are worse than the field, you should play carefully -- you can't afford to give gifts at both tables.

Eh? Giving gifts at both tables doesn't matter, one gift already lost the board. Just play not to lose the board as previously stated. Winning the board at both tables, or losing the board at both tables just is overkill and doesn't count. That's what Richard Pavlicek emphasized on r.g.b., he won 4x Reisinger so I'll take his word for it.

If you get a gift, try really hard not to give it back. The situation talked about was an extremely rare example, where you got a presumed gift of 2 tricks, where 1 trick was presumed enough to win the board under normal circumstances. Only then can you risk 1 of the tricks back in order to get the third trick to save against your teammates screwing up by going for 200. Even here you have to estimate the chances of teammates getting in the auction to be greater than them giving one trick rather than two on the defense. This situation almost never comes up, in my experience (admittedly limited).

If your teammates are worse than the field, at BAM you simply get clobbered. Just enjoy your day of competition against generally better opps (than at a typical regional), and plan your next day doing something else.

[barmar]

Stephen Tu, on 2011-September-23, 14:53, said:

Eh? Giving gifts at both tables doesn't matter, one gift already lost the board. Just play not to lose the board as previously stated.

That's not what I meant. You don't know whether they've given a gift on any particular board. If they've given a gift on board 1, there's not much you can do about it, but don't make things worse by giving a gift on board 2.

[Stephen Tu]

But you also said

Quote

Then when you get a gift, you now have the opportunity to take a risk that's likely to at worst give it back (only risking half a board) but which might turn a loss into a win if something weird happened at the other table.

which I object to. You can't afford to risk half boards at BAM. You *assume it's a win*, that your teammates didn't give away your victory at the other table.

The extremely rare situation presented, you got two gifts, not just one. There you can risk one gift (since still win board if teammates were normal), in order to get a third gift to maybe save the board if teammates weren't normal. I am saying this almost never happens. If you only get one gift, you try like hell not to give half the board back, you don't risk it to cater to something weird at the other table.