1. So the Olympics are about to start--yawn. More interesting to me than the competition or the John Tesh commentary is Andrew Bernard's work predicting national medal counts.

2. What were the folks at OLN thinking when they didn't have a camera ready to cover Armstrong's crossing the finish line in 6th win in the Tour de France? Unbelievably bad programming. (By the way, I sure hope Armstrong will ride the Tour de Georgia again in 2005--he won the 2004 stage that ended in downtown Rome.)

3. It is often argued that one of the benefits of sports programming to television networks is its usefulness as an advertising platform or lead in for other shows. (A good example might be the lead in that NFL on CBS used to give to "60 Minutes.") I'm skeptical of the validity of the argument in general, and I'm particularly puzzled by TBS's plugging of "Sex and the City" reruns during Braves games. I just can't see much of the Braves audience being potential viewers for Sarah Jess and friends. Maybe I'm wrong ...

4. A week or two ago, the Weekend section of the WSJ carried an article purporting to evaluate baseball managers by comparing their teams' actual wins to their teams' projected wins based on the Pythagorean method. I have no qualms with the results per se--if I recall correctly, Bobby Cox received the highest rating with an average of about 2 wins per year above what would have been projected.

I do, however, quibble with the methodology. The Pythagorean wins formula is based on a team's runs scored and runs allowed. (The projected win percentage is run scored^2/(runs scored^2 + runs allowed^2)--though some statheads have tinkered with other exponents.) This formula is not entirely based on things outside of a manager's control; stat geeks like to point out that poor managerial strategy such as overuse of bunting and base stealing can cost teams runs. Thus Pythagorean wins cannot be considered entirely exogenous to a manager's performance and used for comparison with teams' actual wins.