tag:blogger.com,1999:blog-31545676.post7610877875180866314..comments2019-07-12T10:02:52.556-04:00Comments on Sabermetric Research: Answer to: a flawed argument that marginal offense and defense have equal valuePhil Birnbaumhttp://www.blogger.com/profile/03800617749001032996noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-31545676.post-20449688143674333562018-12-21T10:13:37.108-05:002018-12-21T10:13:37.108-05:00The REAL question is, what difference does one run...The REAL question is, what difference does one run over a season make. In other words, when subtracting a run, as you stated, you're not subtracting a run from a specific game, allowing a possible change from a zero run game to a -1 run game. You're subtracting 1 run from a 721 run average 2018 major league season, and redistributing those 720 runs in a new gammma distribution with a slightly smaller mean, and variance probably still a multiple of about 1.1 times the mean.<br /><br />https://stats.seandolinar.com/mlb-run-distribution-neg-binomial/<br /><br />I'm sure such results would be a shade less "extreme" than removing a specific run from a specific game.<br /><br />I don't disagree with your point about reducing opponents runs being more effective for a better than average team, increasing your own run production being more effective for a worse than average team.<br />Unknownhttps://www.blogger.com/profile/15999469754997104765noreply@blogger.comtag:blogger.com,1999:blog-31545676.post-69902485669379420452018-12-04T12:35:23.320-05:002018-12-04T12:35:23.320-05:00Of course. Very elegant explanation. In true rando...Of course. Very elegant explanation. In true random selection, every element must have an equal probability of being selected. Of course, you then need to sample from runs if you are subtracting (or run opportunities if you are adding) rather than games. I should have gotten that. :)Scott Segrinhttps://www.blogger.com/profile/17707976016425293072noreply@blogger.comtag:blogger.com,1999:blog-31545676.post-33075995510534521672018-12-03T22:37:23.126-05:002018-12-03T22:37:23.126-05:00I understand the argument/conclusion from simply a...I understand the argument/conclusion from simply a "counting stat".<br />I was too focused on the "how's".Unknownhttps://www.blogger.com/profile/08881822855187378673noreply@blogger.comtag:blogger.com,1999:blog-31545676.post-52751619463226830932018-12-03T16:37:21.287-05:002018-12-03T16:37:21.287-05:00A much simpler way to say it: runs are a counting ...A much simpler way to say it: runs are a counting stat. It makes no sense to adjust in whole increments a counting stat within an individual game, where this stat might not even occur once.Anonymousnoreply@blogger.com