Friday night game is Columbia vs Yale. Yale is 14-4 overall and 5-0 at home, while Columbia is dreadful with only 6 wins and 0-10 on the road. The public will certainly be lining up to bet Yale no matter what number the books put up. Ask yourself if you were a sportsbook, would you give a fair number to the public on Yale? Of course not.
The total is 131.5 and Yale is -16. If we do some quick math, the final score should be 74-58 to land directly on side and total. So the free 16 points on Columbia is 21.5% of the anticipated 74 scored by Yale. In college basketball, anytime we can get over 20% of the anticipated points scored by the favorite, the underdog has covered 53% of the time since 1985. This percentage goes up when the favorite is a highly recognizable name by the public, and goes even higher when the favorite has no extra motivation to kill the underdog (heated rivals, perceived payback for something done or said, etc) .
Yale is going to win the game, but they play again in 2 days so no need to push starters with big minutes. I suspect they won't score 70, which makes the free points even more valuable.
My model has this game at Yale -12. So the books have added some points knowing the public will bet Yale anyway. So we get to take that extra value on a team in a good situational spot against a team with no desire to run up the score.
Consider a bet on Columbia +16 or +15.5.