Go Rockies!

I'm sick of the talk about the "Law of Averages," which implies that the Rockies' recent success will necessarily end because such an above-average performance is improbable. People say casually "all streaks end," but this is an inductive argument, meaning we think it's true because that's what we've observed in the past, but there's nothing that forces it to be true. The Rockies' streak very well may end, but we must be careful how quickly we declare it to be "statistically impossible" or "highly improbable," and therefore, inevitable to be short-lived.

With baseball especially, we like to exhaustively analyze statistics and try to break the game down as much as possible into a predictable formula, which is valuable, but has certain limitations. From a statistical perspective, baseball games are not random experiments with independent outcomes; if they were, we could easily say that the Rockies' streak is improbable. But baseball games depend on a number of non-random factors and interdependence, so the best we can say is that the Rockies' streak is uncommon, not improbable. In fact, if we have assumed some supposed "average" or expected performance from the Rockies, then the only thing the last 22 games have done is is to allow us to reject that expectation. Simply put, the Rockies are a better team than people who imply this "average" give them credit for.

To put it succinctly, the Rockies have won because they've played better baseball than their opposition, and they'll lose when someone plays better than them, not when some "Law of Averages" catches up. This isn't a very flashy or sensational prediction, but there is nothing to suggest they'll necessarily lose four of seven games because they have to catch up to some "average."