Comments of the day so far
We racked up two great ones in under 20m.
The more of this I hear, the more it sounds like scapegoating. If the Mariners were winning, we wouldnâ€™t be hearing it, and of all the reasons theyâ€™re losing, Johjimaâ€™s relationship with the pitchers canâ€™t possibly be as important as team defense and the lack of hitting and the absurd levels of veteran entitlement and McLarenâ€™s inability to manage a game.
Scapegoating always works like this. Things go bad, a little finger-pointing stars, and then it avalanches as the pent-up frustration gets focused on one target. And itâ€™s usually an outsider.
Johjima is beginning [to] become the fall guy for this season. Is that really reasonable?
Steve T, here:
there is every reason to believe that that can effect performance
Bzzzt. Thanks for playing.
Yes, there are factors like this that affect performance. We can tell, because we can MEASURE THE PERFORMANCE.
Saying stuff like â€œBedard wasnâ€™t sharp today because he couldnâ€™t get comfortable with his catcherâ€ is ANTI-KNOWLEDGE. Itâ€™s not just untrue; itâ€™s blocking the view of the truth. It steps in and â€œexplainsâ€ an event with reference to stuff thatâ€™s completely unknowable (Bedardâ€™s mental state) without first trying to establish whether the event even took place or not.
If you want to make a serious study of the results of pitchers working with different catchers, go ahead. But do it the right way: make sure youâ€™ve got enough data, make sure youâ€™re accounting for other factors, and make sure youâ€™re asking the question in the right way and looking for an answer in a place where it is possible for answers to be found.
Grabbing a tiny handful of numbers from somewhere â€” ah, it doesnâ€™t matter from where, we just want a number here, so it looks like weâ€™re doing science, right? â€” tells you nothing. It tells you WORSE than nothing, because it shuts down your mind and stops you looking for other possible explanations â€” including what is always the most likely explanation in cases like this: random chance.
Your â€œrobotâ€ example doesnâ€™t apply. If Iâ€™ve had conflict with my boss, thereâ€™s no numerical way to adjudge my performance as a result. With Bedard there is. Howâ€™s he doing at getting hitters out? Leaving aside the sample-size problem, I didnâ€™t even see that number there. You rarely do, with pitchers â€” what IS Bedardâ€™s out percentage, anyways? [hint: 1-OBP].
This kind of baloney statistic really gets my goat, and it makes real statistical analysis look bad.