![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Clemson 84
Maryland 72
Who saw that coming? Oh yeah. I did. Damn. Well, time to get revved up for, no doubt, a disappointing road loss to a Mid-Major in the NIT. Probably some team like Virginia Commonwealth... They should have a "resigned" mood.
So, I shall depart from that again and go drop a quote:
James is probably best known as famous as the author of The Baseball Abstract, and leading exponent of Sabrmetrics. That is, the discipline of using statistics to analyze baseball.
But really, his comments are exportable to all manner of analysis and not merely statistical. When we watch a show, and look at scenes to draw conclusions - or make projections about future events, we're doing analysis. (Or, on occasion we are essentially engaging in religion and taking leaps of faith...)
In this particular case, James was discussing the old argument that there is no such thing as a "clutch" hitter - a key point where most statitical types deviate from the traditionalist perspective. The problem, always, is that it's near impossible to define "clutch" and to determine what actually is "clutch" and what isn't. The last at bat of the World Series, with a man on base and your team down by one is obviously clutch. A sixth inning tie in late May? Hmm.
But the bigger point is this. Sometimes, when the data you get from the analysis seems to indicate "nothing" it means that your study/test/case scenario was improperly constructed. Or that you have failed to set up your study. Not that the result is "nothing". A lot of analyses go wrong from the very beginning out of poor design. And now I'm rambling because I'm tired and whatnot, but just go with it. I likes thinking scientific thoughts about the Bill James. It's just unfortunate that I can never do a decent data analysis about subjects other than Baseball. The Econometrics folks are laughing at me right now.
Maryland 72
Who saw that coming? Oh yeah. I did. Damn. Well, time to get revved up for, no doubt, a disappointing road loss to a Mid-Major in the NIT. Probably some team like Virginia Commonwealth... They should have a "resigned" mood.
So, I shall depart from that again and go drop a quote:
We ran astray because we have been assuming that random data is proof of nothingness, when in reality random data proves nothing.
Random data proves nothing -- and it cannot be used as proof of nothingness. Why? Because whenever you do a study, if your study completely fails, you will get random data. Therefore, when you get random data, all you may conclude is that your study has failed.
---Bill James
James is probably best known as famous as the author of The Baseball Abstract, and leading exponent of Sabrmetrics. That is, the discipline of using statistics to analyze baseball.
But really, his comments are exportable to all manner of analysis and not merely statistical. When we watch a show, and look at scenes to draw conclusions - or make projections about future events, we're doing analysis. (Or, on occasion we are essentially engaging in religion and taking leaps of faith...)
In this particular case, James was discussing the old argument that there is no such thing as a "clutch" hitter - a key point where most statitical types deviate from the traditionalist perspective. The problem, always, is that it's near impossible to define "clutch" and to determine what actually is "clutch" and what isn't. The last at bat of the World Series, with a man on base and your team down by one is obviously clutch. A sixth inning tie in late May? Hmm.
But the bigger point is this. Sometimes, when the data you get from the analysis seems to indicate "nothing" it means that your study/test/case scenario was improperly constructed. Or that you have failed to set up your study. Not that the result is "nothing". A lot of analyses go wrong from the very beginning out of poor design. And now I'm rambling because I'm tired and whatnot, but just go with it. I likes thinking scientific thoughts about the Bill James. It's just unfortunate that I can never do a decent data analysis about subjects other than Baseball. The Econometrics folks are laughing at me right now.