[chalks]

So I've played 14 individual tournaments on BBO now. Adding all the IMPs together, my total is a whopping 3.38. What would be considered "good" for this format? What would be a good target to improve that score to? How 'bout for a pairs tournament?

As for my regular gameplay between tourneys, I am mainly playing with people I've played with only a few times before, or random strangers. I can see recent hands, but heaven help me if I'm going to go through all of them and add the scores together manually. What's a good metric for measuring those games on more than just a hand by hand basis?

I know these are all subjective questions, so I appreciate even vague answers. Also, to keep the topic a little more appropriate for this particular forum, what do you say to a beginner along the lines of "when this happens, you'll know you're getting better"

Thanks.

[Antrax]

a) When people tell you (with various degrees of complaint) that your self-rating as "beginner" is inaccurate.
b) You can get hand records from the BBO page for a period, and see average score over a period of time. I don't know how accurate a measurement this is, though, because personally as I improved in Bridge my average IMPs actually dropped.

[vuroth]

chalks, on 2011-December-08, 00:06, said:

what do you say to a beginner along the lines of "when this happens, you'll know you're getting better"

When you start noticing a lot more mistakes as you're making them, you're getting better.

Being a beginner is hard on the ego....

[mycroft]

Riffing off vuroth, I tell a lot of improving novices that there will be times when you feel you're going backwards - that you're actually making *more* mistakes than you did before. This is a *good* sign, and it's a sign of improvement, not backsliding - what's actually happening is that you're seeing more mistakes than you're fixing. You were making them before, you just didn't notice, and now you do.

But it is "hard on the ego".

[phil_20686]

There is a program called double dummy solver that will download your hands from BBO and perform basic analysis. It can, for example, give you your imps vs par. While this is an imperfect way to score, it is somewhat objective, and if you can beat par over a large number of boards that will tell you you are better than the people you play against.

When I was playing a lot in the MBC early in my bridge development (though I should say even earlier) my scores vs par went from about -1 imp a board to +1.5 imps a board when playing in the main bridge club. That was around the time I stopped playing with randoms. Now I mostly play with and against friends who are of a similar standard, and am generally about flat with par again. I only tell you this to make it clear that I really do think that imps vs par provide a somewhat valuable measure of improvement compared to the people you play against.

The other useful thing is that if you are not yet good at DD analysis, the solver will tell you where you lost large imp gains vs par and so you can look at those hands, where maybe you didnt realise you missed a good game or a making slam or that kind of thing. I really think that its a very useful program, though I havent used it for a while as I dont play on BBO much anymore.

[chalks]

phil_20686, on 2011-December-08, 17:55, said:

... double dummy solver ...

perfect. It does exactly what I'm looking for, thank you so much!

It helps to see just how poorly I bid all in one place.

[mycroft]

Please note that double-dummy solvers are perfect, and look at their opponents' hands. "Par" is the "least worst result both sides can get, if everyone bids and plays perfectly". In many cases, humans can get close to par. But when the par result requires the opening leader to underlead AKxx in an unbid suit, to catch partner with his two points being the Q, to get him to lead another suit through declarer's K, or play for a 3% squeeze when two or three 15% squeezes, never mind at least one-of-two 50% finesses are available, then the double-dummy solver will find it, and humans won't.

Similarly, on the rare cases where the only slam that makes is the 4-2 fit, because you need *all* the pitches the 6-3 will give you, and there's a 4-3 break that you can control - -100 is going to be the median score, even if par is +1370.

So, don't try to get to par on every hand - that way lies madness and resulting. Use the DD results as a guide to determine what you can feasibly do, and work on getting *there* - and, of course, playing it well enough to make it/beat it.

[wyman]

mycroft, on 2011-December-09, 11:27, said:

"Par" is the "least worst result both sides can get, if everyone bids and plays perfectly".

To be clear: "par" is the lowest level contract, possibly doubled, in which neither side can improve its score by bidding.

For example, if NS has 10 tricks available in spades and EW has 8 tricks available in hearts, par is 4S= at w/w (bidding 5H would result in 5Hx-3 for -500 for EW) and 5Hx-3 with EW favorable (-500 vs -620, and NS cannot bid over it; the best they can do is X and collect their penalty).