[han]

cherdano, on 2011-January-11, 16:44, said:

I agree with the rest of your post, but I would bet against this.

I don't know if cherdano was betting against gnasher running out of rant-juice or wereagles getting it, but I think he won the bet.

[whereagles]

gnasher, on 2011-January-12, 09:42, said:

1. My main point is that carrying out a good simulation is difficult.

2. In my opinion, simulators should:
- Post their criteria
- Post their code, unless they're very sure it's free of errors
- Inspect some individual hands and revisit the criteria if necessary
- Include details of any subjective decisions they made (such as excluding hands manually)
- Be sceptical about any surprising results

1. It's not difficult to get an approximation. However, if you wanna be very precise and cater for most possibilities, you definitely need a bit of spare time (which I don't have ).

2. But I did just that

[nige1]

Fluffy, on 2011-January-11, 19:31, said:

well since this has became dealer's thread now, does dealer have double dummy analysis?

The good news is that it has a tricks function to work out the trick expectation by a specificed declarer in a specified strain The bad news is that this seems to depend on GIB being available


I'm grateful to Whereagles for introducing me to dealer and I've had lot's of fun with it. I think his simulation is adequate. He published his scripts which enable us to criricise his assumptions (in my case unfairly). The law of diminishing returns applies. In order to do an accurate simulation, you need to know the methods and habits of the players. Every refinement you add is more work and increases the chance of coding error. Usually, it also increases the number of deals that you need to generate to produce a meaningful sample Although I suppose that is less of a big deal nowadays