<p>Bat... what about, well, Bats? It was all our batters in the minors falling on their faces that was so disheartening, did they turn it around in the 2nd half as well? Any standouts?</p>

<p>Admin - the site is chewing up 50% CPU time on a quad-core Intel i5. And then it won't let me edit to format a bit. Any fixes coming?</p>

<p>Couple of posts in the hopper ... just waiting for it to carousel around and post ... thanks homies...</p>

<p>yes...that post hurt my brain muchly...so...more please! I like a challenge.</p>

<p>I would be up for a Konspiracy Korner, but I'm still trying to think about the slit experiment. That has bothered me for about three weeks.</p>

<p>sorry to those whose heads I've hurt with my admittedly dense shouts...LOL...as I'm now at a loss for how to proceed, I'll spare the crowd any further torment</p>

<p>Per MLBTR, Ted Simmons has been removed from his position in the M's front office. <a href="http://www.mlbtraderumors.com/2015/10/hisashi-iwakuma-three-year-deal.html" target="_parent">http://www.mlbtrader...</a></p>

<p>A rumination: how often does the best batter in the big leagues "beat" the worst pitcher in the big leagues? What are the odds of the best batter in the game reaching base (generating a positive run expectancy added) in that situation? The way I have the Elo function scripted right now, I am setting those odds at roughly 64%, which seems reasonable to me. Run expectancy is a measure of the typical number of runs that will score given the starting base/out state, so it already accounts for the probability of reaching base (which is significantly different from 0.5)...this allows us to get a 50/50 frame of reference even though #Baseball is not symmetrical in this way...what ends up happening is that successes are worth more than failure costs (outs are worth an average of -0.25 to -0.29 runs while even a walk is worth +0.35 runs on average and the typical homer is worth +1.4 runs). The problem, as I see it, with Elo for #Baseball, is that the difference between failing and succeeding is too small to create a large gap between the great players and the bad ones - this lack of separation is reflected in the seasonal Elo ratings themselves...and thus, we cannot measure player performance against "better" or "worse" competition and see patterns.</p>

<p>I have a theory I'm testing today that may explain why, when relievers are included, the pitching correlations go down generally, and specifically are lower in the highly-leveraged relief pitching era and why hitting Elo doesn't seem to add any value - it may be that weighting events by leverage index is a bad idea...it may be that Elo picks up on real differences in a player's ability to perform against higher levels of competition (or inability at such), but that leveraging the events to try to capture "clutch" performance is messing up my ability to see it. So I'm rerunning Elo calculations today with leverage index removed. Will let you folks know what I find.</p>

<p>early result of paired player correlation trials...I've examined every two-season pairing for players with at least 100 PA of PBP data in any two consecutive seasons, comparing the Elo rating they would expect to receive based on the mean of their linear-weight generated batter winning percentages with the Elo rating they actually received in both seasons in the pair and correlating them. I find no significant relationship for batters, but a VERY significant relationship for pitchers. The typical r-value for hitting pairs was 0.1 and ditty-bopped as low as -0.05 and as high as 0.25 from season to season. But for pitchers, the typical value was 0.4 and it didn't ditty-bop much at all, except in a multidecadal pattern that peaked around 0.5 in and around 1973 and declined to around 0.25 today. If I exclude player pairs for pitchers where, in either season, they failed to face at least 400 hitters (so I'm mostly looking at starters), the multidecadal trend disappears entirely, and the R values increase to consistently above 0.5. This implies that leverage index is not the cause of the significant relationship at all, and that the strong relationship is more likely due to real skill either in clutch performance or in performance against better competition.</p>
<div class="indented">Reply - anonymous - 10/6/15 6:01pm<br>Thanks for posting, Matt. Sounds compelling. Wish I could spare the time to help you on this, but I enjoy reading about your work.

Also, ditty-bopping is a great term I've never heard in this context. Could come from the style of dance, as in the numbers dancing with high variance, or from the urban dictionary definition of 'the way you walk,' like a random walk. Both are suitable visuals of what is happening!</div>

<p>His nickname is JeDi, DaddyO...Angels fans already coined it...and it is good.</p>
<div class="indented">Reply - DaddyO - 10/6/15 11:23am<br>JeDi. I like it.</div>

<p>But OF can be mixed and matched, and I expect it to. I am deeply concerned about catcher and what happens if Zunino is lost. He needs a capable veteran timeshare. I like A.J. Pierzynski if he's ready to move into that role, but he may still see himself as a starter, and his 2015 should make him coveted to do just that for a team in need of a stopgap until a prospect is ready. I guess it would be nice to glean Dipoto's stance on offense from C, and if he follows the existing mantra of defense first, and whatever he can do at the plate is icing on the cake.</p>

<p>FYI, I put a brief article up in the #Football section. Nothing brilliant, just my reaction to the game last night.</p>

<p>That's where I am, Matt. Mostly because I think that's the way the M's will operate. I love Pompey, but #Toronto is probably not going to cut a guy like that lose without us ponying up large pitching talent. A guy like Jay is a great fit, then. MId-level $ and likely swapable. I have been pretty confident that the M's will let Morrison go in favor of Trumbo at 1B, but I am less sure today, for some reason. I wouldn't be surprised at a Morrison/Montero deal in '16 now. And I won't be too surprised if the only returning OF starter of ours is Miller, with Cruz going to DH. I"m not sure we don't trade Smith. I don't see BOTH Miller and Smith in COF spots. For the past 4 seasons, Smith has OPS'ed .521, .621, .744, .571 against LHP. in '14 and '15, Miller was .542 and .513. Keeping them both with platoon pards consumes a bunch of lineup space, unless a utility guy like O'Malley is one of the platoon guys. You're going to keep an O'Malley-ish guy up anyway. Of course, you could keep Cruz in RF but I'm not sensing that is the way DiPoto wants to go. Flores doesn't seem to have much of a split differential and neither does Kivlehan.Guys like that might get a look, being "athletic" and all.</p>