NCAA Tourney - greatest sporting event in the world

[Moonliner]

Originally Posted by Ghoulish Delight

I heard an interesting discussion on NPR the other day. An econimonist feels that he has demonstrated statistically that points-shaving is rather rampant in NCAA basketball. He bases this on 3 observations he made by studying a large sampling of games in which one team was heavily favored over the other from the past few years:

1) The win-loss distribution (not factoring in the spread, just raw wins and losses) was exactly what would be predicted by the odds.

2) There were significantly more wins by the favored team that were decided by just under the point spread than statistically predicted.

3) Conversely, there were significantly fewer wins by the favored team that were decided by just over the point spread.

It's certainly a compelling argument. He was quick to point out that he didn't have enough data from any single team to point fingers and say, "These guys were point shaving!" He can only look in aggregate and conclude that something is definitely skewing the results from their statistical distribution.

So how can I use that to make a killing with my sports bookie?

[Alex]

Based on the title I was assuming this thread would contain talk of both the NCAA basketball tournament and the greatest sporting event in the world.

Since the NCAA basketball tournament involves college basketball it seems clear to me they can't be the same thing. So when do we begin discussing World Cup? What do you think of the efforts Germany is taking to prevent (predominantly British) hooliganism from marring the games?

[Alex]

Originally Posted by Moonliner

So how can I use that to make a killing with my sports bookie?

It won't make you a killing but it suggests you shouldn't take the favorite when the favored has a huge spread (what defines "heavily favored" I don't know).

Rather than point shaving, though, this statistical analysis, it seems to me, might instead show a psychological skewing in the voting pool (the bettors) as to just how easily a heavily favored team should dominate a lesser team. All the bookies are trying to do is keep an equal number of people on both sides of the decision.

So, if people tend to think "School A is so much better than School B that they should win by 20 points" when in actuality it is very difficult to consistently win by more than 15 points no matter how much better you are, then this might explain bookies setting the spread at 18 and the games more often being on the low side than the high side (rather than equally on both sides as a purely rational system would suggest).