We could also control for Thursday night, Sunday night and Monday night games by including an indicator variable for each situation. Run the regression Bump(i) = A0 + A1*WinPct(i) + ThursNightGame(Y or N)(i) + SunNightGame(Y or N)(i) + MondayNightGame(Y or N)(i) + e(i) where the (Y or N) variables are 1 or 0…

The yellow line has a positive slope which means that as the winning percentage increases, the bump in ticket price will increase. If you want to interpret this on a team-by-team basis, then you want to look at the vertical distance between the helmet and the yellow line. The teams with helmets above the yellow line…

Nice regression results ! Perhaps we can add an additional variable (such as market population) that controls for the # of viewers in the market. The resulting regression coefficient on value would then be above and beyond what the market for the team dictates.

What is the future of college hockey on television? I saw some games on NBC sports network earlier in the year, nothing since

Sorrycuse