Post-Game Talk: Old Joe the Science Bro

[kovazub94]

I love poorly used data.

If you want an accurate read of ANYTHING using early season data, you use the first 10 games played of every season, not the first 10 games of this year and then the full seasons of the years prior.

FWIW his SV% through the first 10 games of last year was .904 and that didn't include a 16 day break in the middle.

Even if you compare against the first 10 games of last season for a better apples to apples comparison, there’s really no story behind these limited data samples. It can show a wide range of results from very poor to excellent and anywhere in between, so only conclusion that can be derived maybe is that a player is on a (hot or cold) streak if the results are on the far sides of spectrum.