Quote:
Originally Posted by cakewalk
Hello again! What are your research question(s) and experimental design?
Hi!
Here are a couple research questions I'll be working on this semester in the project with my dad's hands (and some I'm just interested in) and these are questions I would want to apply to a sample:
1) How much do pocket aces (or just one ace in the hand) increase your chance of winning, and by how much
2) How much does table position influence chance of winning, and by how much
3) What are the payoffs of using really basic hands chart that tells you when to play hands depending on position
4) I would love to look at if the number of tables being played at one time influences performance (but don't think that's in my data anywhere)
5) Does a big win or loss influence future behavior
6) I tried to do a risk volatility study using coefficients of variation based on stakes but maybe the different levels of stakes weren't a wide enough range
Experimental design:
Some of the questions I've done empirical work on (aces, position, hand chart, volatility) but it was all very simple linear regression and most of them probability models with all dummy variables which is about as basic as you can get. There are more advanced probability models I can use, like Probit and Logit, to reevaluate the questions I've already looked at and now I can compare between two models. Other questions, like the behavior change after win/loss, that I will treat as an event and compare before and after results. I also think some interaction terms would be interesting: pocket aces in the BB vs. pocket aces with the BTN.