Monday, November 8, 2010
Rejecting the null hypothesis
In my statistics classes right now, I'm teaching hypothesis testing. The null hypothesis can always be translated into the sentence "nothing special is happening", which usually means we assume two things we are measuring are equal until we get strong evidence they are not.
The null hypothesis would be that the Raiders suck. They sucked last year, and the year before that, and for several years before that. The worst they got was 2-14, really truly awful. The best they got was 5-11, not good enough to be mediocre, just consistently not very good.
As of yesterday afternoon, the Raiders are 5-4, over .500 after more than half the season. More than that, they are the second highest scoring team in the league. Even better, the rest of their division has no team that is a world beater, so they are only one half game out of the division lead after beating the division leading Chiefs yesterday in overtime in Oakland. So far, the Raiders are 3-0 in the division, which could be an important tiebreaker.
With a record of 5-4, it's not fair to say they are awesome. One of their losses was against the struggling 49ers, and they also lost to the Cardinals, another team under .500. Still, the evidence is strong enough to reject the null hypothesis. As of November 8 2010, we reject the notion the Raiders suck, and not sucking is a very important step forward.