Just about every Tuesday evening I play tennis with a group of guys for a couple of hours before we head to a local bar for a couple of beers. After reading this item on Freakonomics I'm thinking we should reverse that order:
The key finding of the new research is that the intoxicated participants solved more items on the Remote Associates Test compared with the control participants (they solved 58 per cent of 15 items on average vs. 42 per cent average success achieved by controls), and they tended to solve the items more quickly (11.54 seconds per item vs. 15.24 seconds). Moreover, the intoxicated participants tended to rate their experience of problem solving as more insightful, like an Aha! moment, and less analytic. They also performed worse on a working memory test, as you might expect.
In the past I've found that if I happened to have a drink before playing – say on a Saturday after mowing the lawn I had a cold beer and then headed out to play a couple of sets of tennis – I'm generally much more relaxed and play much better. As a good friend once told me, my brain is my worst enemy on the tennis court, so maybe there's something to this concept. I'll run a little experiment and let you know how it goes.
Befuddled 
For years, I played in a little band designed to do pig pickings. There was never a stage and as things got rowdy, the dancers often encroached on our space. Generally, the night would end with people joining in to sing their fave song. As usual, we all eventually got married and retired. However, it was customary for me to have a drink or two before we began, as it loosened me up and I played a lot better. Of course, the challenge was not to drink so much that I got too loose.