Brawndo the Thirst Mutilat

Since the offense knows the logical choice is to take the man at third it is also equally sensible to have the runner go to third to guarantee the run in those same circumstances. Why? Why not just see whether they can throw the guy at home out? If they can't, you get a run, and you still have a runner in scoring position. A bird in the hand is not always worth two in the bush. It depends on your odds of getting that second bird, and on how desperately you need at least one bird. In general, people who coach and analyze professional sports are incredibly too risk-averse.

Ben Sheets is not choking. Ben Sheets is not in need of a gut check. Ben Sheets is not sulking in the shadow of C.C. Sabathia. He had a .400 BABIP against in July. This is simply variance -- his season total of .286 is right around the mean nearly every pitcher's BABIP inevitably regresses to -- in other words, July was bad luck. His ERA results were heavily impacted by this, and combined with a lack of run support in general since the AS break, it has created a perception of a struggling pitcher where none exists. At this point, I'm just waiting for the first jerk to suggest that McClung should be given Sheets's spot in the rotation.

It is never worth it to sacrifice the out, except possibly in a tie game, late game situation. But if the single run has become that important, it is highly doubtful the idiotic "hey, look at me, stupid, throw me out at 3rd!" ploy is going to draw the throw. If the run seems that important to you, it will seem that important to the other team. Really, the stupidity of this concept is amply demonstrated by the fact that the opposing team consistently takes the free out that you're offering over the chance to throw the runner out at home and prevent any runs from scoring. If the free out is worth more to the opposition (every opposition) than the chance that the runner is thrown out at home, doesn't that tell you something about what should be more important to you?

Sabathia doesn't have to be Prior or Wood or Pedro for it to be abso-freaking-lutely imbecilic to run him out there for 130 pitches in a game that is already won. I don't know whether this will adversely affect him in his next few starts, or whether this kind of deal, repeated over and over down the stretch, will cause him to tire and lose effectiveness at some point in September or in the playoffs. None of the rest of you do either, and neither does Yost. I don't think that the over-worked pen criticism from last year and the over-worked rotation criticism from this year are in any way mutually exclusive. It is, in fact, fairly obvious that it is possible to go to too far an extreme in either direction. This is just my visceral reaction, and I'll grant I'm still pretty cheesed off right now, but dangit...Yost is the worst judge of high vs. low leverage, the worst judge of what sort of pitcher is appropriate in what sort of situation, and about the worst judge of how far to push any kind of effective pitcher (be it a starter or reliever) I've ever had the disprivilege to have to endure watching manage a Major League Baseball team.

It is my hope that you're also left with something of the incredibly rich character psychology and resulting differences in moral philosophy which make it the greatest story ever told by someone not named Dostoevsky. One of the several almost-was incarnations of this film would have included an updated setting. Personally, I'm glad that it seems they ultimately decided to stay as true as possible to the original. It's just too good to screw with. I don't think of Snyder as a total hack, but I am sorry that Terry Gilliam isn't making this film (even if both he and Moore concur that it is unfilmable, I think he would have done the best job...I think it would have felt like Brazil). Aranofsky was supposed to direct it for a time as well, and I'm a big fan of his films (plus, that would almost certainly have meant a mind-blowing original score by Clint Mansell). Snyder obviously isn't remotely in either of those guys' leagues, but I really like the look of the trailer, and ultimately I think that the quality of the screenplay will have more to do with whether I like the film than the quality of the director. David Hayter (voice of Solid Snake) wrote the first two X-Men films, which were decent, but he also the Scorpian King, which...um...wasn't.

DiFelice's MLE FIPs against LH for the past two seasons, since he has been throwing mostly cutters (he does have an 88 mph straight FB, a curve, and a change as well, all of which he used at least a few times at the MLB level this year) are 4.92 and 4.88. Not significantly worse than the 4.41 career MLE FIP. Of course, he's bunches better against RH hitters over that span (since his overall FIPs are barely above 3.00), but that doesn't mean that he's too terrible against LH hitters to use as a SP. If you wish, you can continue to regard major league LH hitters as entirely different animals relative to AAA LH hitters, such that no translation of minor league stats is believable or worthwhile. I find that a very strange article of faith.

Dillon's minor league splits since '05 show a reverse platoon differential. Note that the numbers in the link are MLEs -- so they are adjusted for park effects and level of competition. Cumulatively, he's hit for an MLE line of .269/.361 /.458 against RHP since '05. This represents a much more reasonable expectation for him in the bigs going forward than his actual MLB splits against RHP. Edit: Link doesn't work. Try again: http://www.minorleaguesplits.com/cgi-bin/pl.cgi?pl=434338 Ok, that one works, but you'll have to click "career" and then "MLE" to see the line I quoted above.

Jeff Sackmann has previous year minor league split data back up, so if you look here, you will see that there is nothing drastic or alarming about DiFelice's career platoon differential. Some key numbers: FIP vs LH: 4.41 FIP vs RH: 3.29 AVG vs LH: .256 AVG vs RH: .212 HR/9 vs LH: 1.30 HR/9 vs RH: 0.85 (this one's a bit wider than the others, but it's nothing like the impression you'd get based on his tiny MLB sample) By way of comparison, McClung's career MLB platoon differential is a little wider: OPS vs LH: .879 OPS vs RH: .720 AVG vs LH: .287 AVG vs RH: .239 Side note -- it's a bit sad that Sackmann's minor league splits now offer better info than anywhere I'm aware of where you can get major league splits. At any rate, can we please put the notion that DiFelice is some abnormal liability against LH to rest? It was a small sample, people. Never trust extreme results over small samples -- they always lie. As for where he ought to stand on next year's rotation pecking order, DiFelice's overall MLE FIPs for the last 2 seasons are 3.14 and 3.07. He might well be the second best pitcher in the organization next year. I'd certainly take Gallardo over DiFelice, and Parra is probably better too, but I'd sooner see DiFelice in the opening day rotation than Bush, Suppan, Villy, and (for the love of all that is holy and unholy alike) McClung. He's no worse than a league average SP.

The Great Productive Outs Debate of 2004, spurred by a Buster Olney article on ESPN, produced some of my favorite moments in sabermetric research. For example, here Cyril Morong demonstrates that for an entire season's worth of data (2004), if you hold team OPS constant, run scoring actually decreases the more productive outs a team makes. Larry Mahnken had a couple of gems that are still archived at Hardball Times -- check out this one. A couple of findings: - The top five teams in POP have scored 4.33 runs per game, have a .351 POP and a .392 winning percentage. - The bottom five teams in POP have scored 4.74 runs per game, have a .230 POP and a .534 winning percentage. The scary thing is that TWR might be absolutely right about why Yost wants Counsell in the lineup. Others have covered the defense angle -- there's just not enough of a difference to justify getting a totally anemic hitter this many ABs. Productive outs and contact hitting reduce to the same thing. If it has come to this...to a place where lineups are constructed in the pursuit of the sacred productive out in the middle of a playoff push...well, frankly, I'm just not sure I can really have any long-term hope for this franchise under current management. That goes for Melvin too, if he's just going to stand idly by and let Yost continue to do this. And the fact that the offense has been fine and the team has been winning with Counsell in the lineup, even though he's 0 for whatever, is not evidence that he's helping or that starting him is a good idea. I could use the exact same reasoning to argue that Braun better just sit out the rest of the season, because the team is undefeated since he went down and has yet to score fewer than 5 runs in any of the games.

Glaus just hit a 3 run shot on the very next pitch after a rather obvious intentional brush-back. 9 - 2 Cards. Don't really care who wins this game, but it sure was fun watching Zambrano get the snot pounded out of him.

Don't take this as any huge disagreement or anything -- the general point of BtA's post is spot-on in my opinion -- but I can assure you that being a SABR dork in no way precludes you from having an emotional connection to the game. So could anyone who was carefully observing my alcohol consumption patterns Monday through Thursday last week. With respect to the question of projections presented as true talent estimations, I've done this several times, because that's what they are. Note that I say "estimations". What non-SABR fans don't realize is that they do the same thing all the time. When someone says, for example, "J.J. Hardy is hitting .000 over his last 20-whatever ABs -- bench him!" they are looking at a statistical description of the past, applying present tense language ("is hitting"), and making an argument that a decision should be made with a view of likely near-term future results. To make this argument, s/he necessarily must believe that J.J. Hardy's present true-talent hitting ability (if only for the near-future) is better described by those 20 ABs than by (for example) a 3-2-1 weighted amalgam of Hardy's last three seasons, adjusted for age and regressed toward the league average hitter. But wait -- Ned didn't bench him and he went 2-5. Then Ned benched him. Now what? Is Hardy still cold? Is this the beginning of a hot streak? On my side of the fence, it was a fool's errand to try to identify Hardy as "currently" hot or cold in the first place. None of us would deny that streaks happen -- but I would deny that statistical descriptions of recent streaky performance are of any use in predicting future performance. Why things get a bit tense around here in-season on this score is no great mystery. On any given day, the 4-5 most active threads are usually the ones which were either started or resurrected based on yesterday's results, or last week's, or at most the last month's. For fear that my voice will otherwise be drowned out by the thunderous cacophany of "Weeks is a bum!" and "Start Kapler!" posts, I know I find it awfully tempting to dig in a bit more fiercely than is perhaps called-for. I really try not to hurt anybody's feelings, and I'm quick to apologize if I recognize I've stepped into smarmy jerksville, but it's really tough not to sometimes. I don't really think anything can be done to rectify this situation -- for most fans, it seems the last few weeks are either all they really remember, or all that summons BtA's "emotional connection" for them. I think the best the site can do is try to keep the tone civil and semi-constructive, and I think the mods do an excellent job of at least trying their darndest to do that.

Prior to yesterday, Kapler was 0 for his previous 12 PA in the preceding two weeks. No hits, no walks. That included 2 starts. Lo and behold, it turned out that this mini-slump had zero predictive value for his performance in yesterday's game. Just like yesterday's 3-4 has zero predictive value for his next AB, whenever that might be.

MGL, from that link rluz provided: "Griffey is a replacement player no matter how you look at it. How often does a team make a trade to GET a replacement player?" I think that the Griffey-bashing here has been relatively measured and restrained. Sure, he can still hit a little. Just not enough for a DH. And not necessarily any better than the player he's displacing, Paul Konerko. That's if the White Sox are intelligent enough to know that Griffey is a DH now. If they plan to regularly play him in the OF, this trade is a disaster for them. If they actually let him play CF for more than a few innings, this trade is a lot closer to "good night, White Sox" than "good night, Twins."

I suppose this trade could really help the Dodgers, but it seems equally likely that Kemp or Ethier will be sitting on any given day as Jones or Pierre. In fact, I really can't see BOTH Jones and Pierre riding the bench very often at all. The second there is a losing streak of more than 2 games, Jeff Kent, Jim Plaschke, and Lee Simers will have a collaberative conniption. The fact that Ramirez will displace Kemp or Ethier a fair amount of the time cuts into the impact this has on the NL West race rather drastically. I still think the D'Backs are the best team in that division. I think people are overestimating the return the Pirates got a little bit. They did okay -- I still like LaRoche, and Moss might be an adequate starter at a corner OF position for a bad team if he's lucky. I don't think either of the pitchers they got will ever be useful parts of a quality staff. Red Sox are the clear winners to me. They shouldn't have had any leverage at all, given the soap opera going on with Manny, and Theo still managed to come away with the best player in the trade, with the best contract, for 1 1/3 seasons, and he didn't give up anything he cared about. I think they're better off with Bay than Manny for the duration of '08 anyway.

I watched every game in this series up to the end, missing only one half inning in transit to a bowling alley on Monday. At least, at the close of the final atrocity, the fine folks at the National Geographic Channel have seen fit to provide the perfect program for me to watch at the moment, given my mood. It is called "Garbage Mountain". And, yes, the creators are being quite literal with that title.

Bonds was one of the best defensive outfielders in baseball when he still had his wheels. Even old, totally immobile Bonds got good enough jumps and took good enough routes to balls to remain a statistically average LF most years. His gold gloves were most certainly not a joke.

They award 3 outfield gold gloves, but in most seasons all 3 go to guys who play CF, or maybe 2 CF and a RF. Barry Bonds won 8 in the 90s. I can't think of a another LF who has won one recently.

Griffey may get the chance to DH until 2010 (if he indeeed wants to) due to his name and reputation, but if he does, whatever team gives him that chance is dumb. He's going to be 39, and he will have hit too poorly for a DH 2 out of the last 3 seasons.

The top 3 are really the only viable candidates at this point. I would put the current order as 1. Jones 2. Berkman 3. Pujols. The only others I think have a plausible shot at catching that group are Utley and Ramirez. But the writers only vote on triple crown stats, unless they can't find a viable triple crown stats candidate. In which case, we learned last year, they vote on the quality of the story (Rollins). Thus, Braun could certainly win one of the more undeserved MVPs in recent memory with a strong finish - he is a triple crown stats kind of guy on a winning team.

Don't see it. Bill James's "Favorite Toy" puts that chance at 2%. I suppose, technically, that is a chance. Barring lots of injury problems or a Dale Murphy early 30s decline, I think Pujols will get 3000, but I'd bet all the money I will ever earn against 4000.

So you assign more weight to the most recent 25%, but you cannot discount the other 75%. That's exactly how Marcel projections work, and the monkey still likes Weeks better than Durham on offense for the duration of the year: Weeks: 248 / 354 / 412 Durham: 259 / 335 / 412 On defense, all of the advanced metrics like Weeks better than Durham. Dial had him 5.5 runs better on his all star break leaders update. THT's numbers have Weeks near the bottom of the pack in RZR, but he's second only to Utley in OOZ plays, putting him middle of the pack overall among regular 2B. Durham has both the lowest RZR and fewest OOZ plays of any 2B who has played more than 250 innings. If you lower the threshold to 150 innings, only Ronnie Belliard has been worse. The only argument I've seen anybody make for Durham's defensive superiority (including Haudricort) is some variant of the increasingly irritating "I've seen Weeks, and Durham can't possibly be worse," which only betrays a lack of imagination on the part of the author. We're all bummed about the loss yesterday, and we all love to have a scapegoat. But even in a year where his AVG is way down (incidentally, AVG is about the least stable, and therefore least predictive, stat that anybody really cares about), Weeks excellent walk / HBP skills, his above average IsoP, and his middle of the pack defense have kept him right around league average for a regular 2B. If he can start hitting just a bit more like he's hit in the other 75% of his career, he will be significantly better than any reasonable expectation for Durham.

Villy's updated Marcel projection ERA for remainder of '08: 4.32 Torres's updated Marcel projection ERA for remainder of '08: 4.07 Before you scoff at the Torres projection, realize that he's 36, he's already thrown as many IP as he did all of last year, and as bad as he was last year, his peripherals are actually worse this year. I think that's actually a pretty fair estimate of Torres's present true talent run prevention ability. They're much closer to the being the same pitcher than they are to being separated by as wide a gulf as you used. And those pitch splits are going to be way too small a sample to learn anything meaningful.

Even if Wolf is a class B, do you really want to risk offering him arbitration? He might accept.

But, aside from a tiny window of his career now over 4 years removed, Mota has never been a good pitcher.

What do suggest is a better measuring stick? Lots of choices here. FIP or xERA if you insist on looking at single season stats. 3 or so year FIP and xERA trends, and the constituent stats that determine them (pitching peripherals). Updated Marcel projections. A weighted combination of the better projection systems (Pecota, ZiPS, CHONE, etc.) and this season's results. Pretty much anything is better than saying "Torres's current ERA = Torres's current true talent run prevention ability".