[thepossum]

Hi all

Can anyone please advise on a few matchpoints strategies (and maybe also how the decisions vary in IMPs)

I know there is always a large component in MPs and IMPs of trying to work out what everyone else is likely to bid with a particular hand. I've also read that the 50% chance of games is a good rule of thumb for bidding in MPs

However here are a few possible scenarios

You are non vulnerable versus vulnerable in a non-competitve auction. You have a (almost certainly) sure game versus small slam that you estimate is close to dependent on the position of a missing King

Ignoring what everyone else does you have 50% chance of beating those in game and 50% chance of doing worse than those in game

If everyone thought the same way maybe half (some arbitrary percentage P) of the people choose game and half (1-P) choose the slam

You think the slam looks like a good 50% slam and go ahead and bid, go down 1 and score around 25%

So clearly the assessment was incorrect. Or was it? What has been missed in the consideration?

So is my 50% figure for slams incorrect. Should it be 75% and 50% for games. Ive heaard others say that the grand over small slam decision is a 75% one

I would be very interested in discussion around these points, estimation of P and how these things vary between MPs and IMPs.



regards P

PS I know the difference is that some people may have had a way to check for the missing kings but it wasnt very easy in my auction

PPS Also assume that you know your average MP percentage in this cohort - eg 55%

[HardVector]

Mike Lawrence wrote a book on this. I think it was called, "Play Bridge With Me".

[steve2005]

If everyone will bid game (which isn't always the case) bidding a 50% slam is reasonable.
But you wont get rich bidding 50% slams I think you need more but 75% is too high a bar.
Also, how good is the bidders at estimating your slam odds before you have seen dummy?

[helene_t]

Hi Possum,

Bidding small slam under simple assumptions:

If you assume that either everybody makes 11 tricks or everybody makes 12 tricks, and everybody will bid either game or slam, then you should bid the slam if it has 51% chance and not if it has 49%.

To see this, imagine a fraction p bids the slam, and the slam has a probability q of making.

If you bid the slam you get 1 MP against the slam bidders, that is a score of 1*p=p, while against the game bidders you get 0 if game doesn't make and 2 if it does, that is an expected score of 2q(1-p).

So your expected score by bidding slam is p + 2q(1-p).

For bidding game, the expected score is 1 against the game bidders (expected score 1-p) while against the slam bidders you get 0 if slam makes and 2 if it doesn't, that is an expected score of 2p(1-q).

So for bidding game you expect 1-p+2p(1-q).

So your net gain from bidding slam is p+2q(1-p)-1+p-2p(1-q) which simplifies to 2q-1 so it is positive if q>0.5.

A simpler way of getting to this result is just to reason that you gain 1 MP against everyone by making the right decision compared to what you would get if you made the wrong decision, so you simply need to maximize your probability of making the right decision!

Heterogenous fields:

This can be a bit different if all kind of weird things happen in the field. If not everybody makes the same number of tricks, you may reason that if you make 4M+2 you will get a good score since some will make only 11 tricks, but on the other hand, you could also reason that 4M+1 is a poor score so you should be more aggressive.

On the other hand, maybe you think that some people may not even be in game, while some will bid a no-chance grand slam. In that case you generally should be a bit more conservative.

It is usually difficult to judge what the field is doing so my advice would be to be a bit more conservative with bidding slams, maybe require some 55-60% for small slam but still about 50% for game. Vulnerability doesn't matter unless you think the field may be defending against a sacrifice, but again I wouldn't worry about that.

Grand slam:
Grand slam decisions depend a lot on the strength of the field. In a weak field you should almost never bid a grand slam, but in a very strong field you don't need much more than 50% chance.

IMPs:
At IMPs its a bit different because 4M+2 is not significantly better than 4M+1, so therefore at IMPs you bid small slam if it has >50% chance (vulnerability doesn't matter), while nonvulnerable games require 50% and vulnerable games somewhat less, and grand slam requires somewhat more.

The big difference (other that safety play and penalty doubles) is not so much in bidding decisions but more in preparations. At MP, it is important to work on your fitness as it is more exhausting (no boring boards where you can relax), and it is more important to make good agreements about invite versus competitive bidding. At IMPs, it is important to have good slam bidding agreements. At MP you can afford not to worry too much about slams. But slams often decide IMP matches.

It is also sometimes said that bidding is relatively important at IMPs while play/defence is much more important at MPs. This sounds somewhat plausible but I am not sure how good the evidence is.

[thepossum]

Thanks everyone

Great advice and much appreciated

regards P

[pescetom]

 helene_t, on 2019-February-27, 20:11, said:

At IMPs its a bit different because 4M+2 is not significantly better than 4M+1, so therefore at IMPs you bid small slam if it has >50% chance (vulnerability doesn't matter), while nonvulnerable games require 50% and vulnerable games somewhat less, and grand slam requires somewhat more.

My limited experience of IMPs suggests that those percentages for game are a bit conservative. FWIW I was taught to bid game 40% when vulnerable at duplicate teams, and I see that the paper "Percentages for Bidding Games, Small Slams, and Grand Slams at Duplicate Teams" by Glen Meeden and G. M. Prabhu calculates that nonvulnerable games require 45% and vulnerable games 38%.

[hrothgar]

If you find yourself in five of a minor (and this isn't a sac), you might as well just bid six

[Cyberyeti]

 hrothgar, on 2019-February-28, 16:06, said:

If you find yourself in five of a minor (and this isn't a sac), you might as well just bid six

Only if you think 4M or 3N are also making.