How much trust do you put into previous results?

[Bluehorseshoe]

I'm looking at last year's past results for MLB and there is such a deviation on one thing from previous years, I wonder what the point is in using it as a guide. If something is that far off, why bother?

There are so many different variables from season to season. The balls are juiced. The players are juiced. Now they're not juiced. New stadiums.....etc.

Feel free to attack me on this one.

[mintybetmachine]

MATH FOR THE WIN

[mintybetmachine]

or look back many years

and the steroid era doesn't matter that much in my opinion because EVERYONE was doing it so it kind of canceled out

[benjy]

A big deviation is usually either an outlier (which does happen) or a signal that something significant has changed (like the factors you've mentioned). Determining which is the case is often impossible from data alone in the short term.

It really depends on what you are looking at. For instance, home teams will still win more than away teams. Yet a big change in home win rates occurred after WWII (likely due to easier travel). An interesting discussion is in this sub-forum and on the links included.