Brawndo the Thirst Mutilat

I don't either, and I don't even feel like I need to make an argument that Villy is a better pitcher than he's shown so far this year (though that's true) to explain why. At the end of the day, both Villy and Dillard were going to pitch multiple innings today. I can't get worked up about the order in which they pitch said innings. In fact, if anything, it's probably smarter to use the pitcher whose performance and number of innings look more uncertain first. In the event he's getting lit up, at least you know that you've gotten nothing from him, and the other guy can go 5+ innings in a game that you are highly unlikely to win anyway.

Adjusting your expectations for recent trends is attempting to make a meaningful prediction, which in the first paragraph above you assert should not be done. When you say, "he is not pitching well right now," all you really know is "he has not pitched well in the recent past." In shifting this statement into the present tense, what you imply is that "he will continue to not pitch well in the immediate future," which is clearly a prediction. Try as they might, analysts have been unable to demonstrate that you can gain any predictive advantage by chasing hot and cold streaks over the short run. They are easy to identify in hindsight, but that's it.

I don't know, the Rays could be pretty motivated to move him...they have the pitching depth, and if he can't get back his velocity and his extreme strikeout rate or get his walk rate down into a more manageable territory, it's not out of the question that he might be totally broken. He's also about to start getting much more expensive. That contract could look awfully unmovable by this time next year if things don't improve, and maybe they don't think they're likely to get the good Scott Kazmir anymore this year in any event. Thing is, they're one of the smart organizations. If these are the reasons they're moving him, do you really want to bet on them being wrong?

Thanks for linking that. It was really interesting. It struck me that the one "sabermetric truth" Macha seemed to have picked up from his time with Beane was that lineup construction really doesn't make much of a difference. Which is true, but I couldn't tell if he was quoting Beane on this in genuine agreement or with a species of condescension, like "Look, if I listen to you math dweebs about this issue, you'll tell me it doesn't mean more than a handful of runs over the course of a season anyway, so why exactly should I do anything that'll make me look crazy to baseball people (e.g. bat Prince second)?" Also, I guess the entire organization really does have so little imagination that it wouldn't have mattered to them if Mark DiFelice had pitched a complete game shutout every single AA and AAA start he made. "He only has one good pitch, dang it, and that one is weird and slow...I've never seen anyone be successful starting in the majors throwing slow crappy pitches of various speeds that are set up by a weird slow pitch...wait, why did you just force a cough and say 'Tim Wakefield' and then force another cough...I see what you did there, and I don't like it."

I think that remains to be seen. He spent Gonzalez and Smith on a season of Holliday, and that gamble hasn't panned out because the A's are much lousier than he anticipated. He still might wind up with Brett Wallace from the Cards out of that, though. Brett Anderson is already pretty good, and they'll have twice as many seasons of him as they would have had of Haren. Cunningham is probably just a 4th OF, but I like Chris Carter's chances to be a pretty decent 1B for them as soon as next year.

Right you are, it was the ignominious Chuck Lamar. Might be the only trade he ever won. In concert with the acquisition of a post-injury Kris Benson at a cost that was, essentially, the entire Mets' farm system, it sparked this thread over at BTF (which might still have been baseballprimer at the time, I can't quite remember when the name changed...). However, somewhat shockingly, the Mets' GM at the time was not Steve Phillips, it was Jim Duquette. Which you need to know to fully appreciate the relevant Transaction Oracle post, which is also quite funny, unless you absolutely cannot laugh at mildly foul language. Still, half a decade later, one out of two ain't bad, TLB.

The thing is, Haren WAS extremely expensive for Arizona. They gave up 6 prospects, all of whom have since either already contributed at the MLB level (Brett Anderson, Greg Smith, Dana Eveland, Aaron Cunningham), or remain highly regarded prospects in the high levels of the minors (Chris Carter), or remain highly regarded young players despite being rushed to the majors a bit too soon (Carlos Gonzalez, who was probably seen as the closest thing to an "A" prospect at the time of the deal). That's right up there with the king's ransom Texas got for Teixeira as the two biggest prospect hauls anybody's gotten in a decade or so. When it would have been nice to get Haren is when Oakland got him from St. Louis...all that cost Oakland was what would prove to be two very forgettable years of Mark Mulder's services. Not that you can really blame the Cards for thinking that they were getting a better pitcher than they wound up getting.

My mistake. First time I'd noticed this thread. Good early call on the Royals dangling Bannister.

After the PR nightmare that was the trade for Yuniesky Betancourt, if Dayton Moore were to essentially admit defeat by so quickly trading for another MLB starting SS, he may as well fire himself.

I don't really disagree -- he probably isn't enough to make the difference between missing the playoffs and making the playoffs -- but then again, the vast majority of the time guys like Halladay and Lee aren't enough to do that either. Last year's experience with Sabathia was the exception, not the rule. If anybody's curious, this was Bannister's first year of arbitration. He settled for $1.7 mil. He's having a very nice year, so he's probably looking at a decent raise, but is unlikely to get more than $2.75-3 mil or so next year and $4-5 mil or so in 2011, depending on performance. He's basically a slightly cheaper Dave Bush with one more year of team control left. If we'd been sellers instead of buyers at the deadline last year, how much of a return do you think we'd have required for Dave Bush?

Probably little more than a throwaway line, after all the title of the article is "Brewers' trade options appear to be limited", and I would certainly need to know what kind of return the Royals are looking for, but this guy is one of my very favorite human beings in the game of baseball. Why, you ask? He's smart. Scratch that, he's really, really smart. Last year, he had some bad luck on BIP, and the Royals were dumb enough to prefer, of all people, Sidney Ponson in their rotation at the start of the season, but since returning from AAA, he's been darn good this year, and that's with a much more normal .288 BABIP than the fluky .266 BABIP from his other good year. The awesome thing about Bannister is I'm sure he's aware of all of this. The Royals are probably thinking it's better to sell high on a RHP with a sub 90 mph fastball than wait and watch him have another stretch like last year torpedo his trade value. They may not be asking for a ton (after all, he's no Sidney Ponson). I've got my fingers crossed.

League average first strike percentage was 58.7% in 07, 58.6% in 09, and is 58.1% so far this year. Given that pitchers have a tougher time locating the strike zone early in the season, it's safe to assume it will creep back toward where it's been the last two years over the course of the rest of the season. I'm assuming you got Braun's numbers from fangraphs -- if so, I should point out that all you have to do to see the league averages for a lot of the stats they have there is find the "show averages" button right under the heading for that group of stats (e.g., in this case, right under the "plate discipline" heading on Braun's season stats page).

This shouldn't be a job. I eagerly await a time when I can cheer for a team that doesn't believe it should be. I think the team would have given Swindle an extended look, rather than calling up Narveson in the first place, if they didn't believe they needed somebody to perform this totally inane function. That would have been a very good thing. Swindle is a much more effective pitcher than Narveson, and there's every reason to believe he's somewhere around as good as Stetter, who has already made everybody forget all about Brian Shouse. That said, we shouldn't pretend we learned anything about Narveson's ability level from the 10+ IP he racked up as a Brewer. The big jump in his K rate moving from SP to RP this year (while still much too small a sample to get terribly excited about) is at least a reason to think he might be a useful pitcher. The chances he could be really good are vanishingly slim, but he's probably not as sub-replacement level as it's tempting to think given the lousy MLB results (after all, a .381 BABIP will tend to torpedo your chances of keeping your ERA down). I will say that I think they made the right choice if it was Narveson vs. Chris Smith, which it probably was.

For everyone throwing out the "but what he said is true!" defense for Braun's comments, I'm reminded of a scene from the Big Lebowski, which I obviously cannot quote in full given the language rules around here, but if you've seen the film, I shouldn't need to: Walter: Am I wrong? AM I WRONG? Dude: No, Walter, you're not wrong, you're just... You know how the rest goes.

For those that think DiFelice has only one effective pitch, did you know that Tim Wakefield's fastball, clocking in at a mighty 72 mph, is the 5th most effective fastball in the majors this season?

Right, I think -- hence, when a player crosses the 1000 / 2000 actual PA line, their actual results begin to inform your true talent estimate more than the league average split does. I guess I may have made it sound like you can stop regressing once the 1000 / 2000 line is crossed, which isn't true. It's just that for anybody with less than 1000 / 2000, knowing their actual results is less valuable than merely knowing the normal platoon split. For anybody with more than 1000 / 2000, knowing their actual results is more valuable than knowing the normal platoon split.

Depending on how long ago that discussion occurred, it may have been that MGL's comments reflected the prevailing opinion among hardcore stat geeks at the time. Bill James was the originator of the idea that batters have no individual tendency toward unique platoon splits in his 1988 Baseball Abstract. By 2005, James had concluded that he had been wrong. The linked SABR article is really good -- if anybody even remotely interested in sabermetrics hasn't read it, I highly recommend it. Essentially, James is reminding us that absence of evidence is not necessarily evidence of absence, and that just because something is extraordinarily difficult to detect doesn't necessarily mean it does not exist. I'm not sure how long the sea change on this question took, but I doubt that MGL would still maintain that no player ever has a true talent platoon differential that is wider than the league average. I'd also just like to note that it really isn't possible to arrive at a hard and fast "certainty line" in terms of how big a sample is needed for something like this. The best you can do is say that you do better assuming league average results in projecting future performance (regressing 100% to the mean, as it were) than using prior statistics until you get a sample that is around x PA, or IP, or whatever. That's probably obvious, but I wanted to make sure no one mistakenly read the contention as if 2000 and 1000 PA were some sort of bright line division between uncertainty and certainty. When you have a smaller sample, typically what you want to do in order to estimate true talent is regress part of the way to the league average. The more information you have, the less the need to regress.

Actually, both RHB and LHB can have persistent platoon splits as an individual trait that are wider or narrower than the league average, and even sometimes in the opposite direction. The problem is that the statistical noise in this data is so very large that an absurd quantity of PA are required before an individual hitter's personal platoon split becomes more reliable than simply assuming a league average platoon differential going forward. The Book gives the required number of PA against LHP before this occurs as 2000 for RHB and 1000 for LHB. In his entire major league career, Bill Hall has so far amassed just 841 PA against LHP. So Hall is not yet even halfway to having faced enough LHP that we can assume (statistically speaking) that his so far observed wider-than-normal platoon splits are a persistent attribute of Bill Hall as a hitter. To give an idea of just how long it takes before a RHB amasses enough PA against LHP for his individual platoon split to be trustworthy, Derek Jeter just got there last year. Albert Pujols should arrive sometime late in the 2011 season if he stays healthy. Hall isn't going to get there, ever, in all likelihood. This is part of the problem with studying this issue -- if you are a RHB that cannot really hit RHP, you will tend to play your way out of the majors at some point, no matter how thoroughly you crush LHP. Note that I'm not even saying that Hall can't hit RHP -- only that if he continues to fail at it on the scale he has the last few years, he'll never be given the opportunity to attain a large enough sample size that we can be confident that his failure to hit RHP was actually a persistent phenomena rather than a transient one. Oddly, partly because there is simply a greater spread in LHB ability to hit LHP, if we're willing to use minor league data in place of MLB data (not necessarily safe, and I'm not sure it's been studied), it's probably a safer bet that Gamel will have a narrower-than-usual platoon split than that Hall will continue to have a wider-than-normal one. Gamel has over 500 PA against LHP in the minors, more than halfway to reliability.

It would be .720. If you could find somebody who hit a single every time, his wOBA would be .900, reflecting the fact that sometimes a single is better than a walk.

Amend that to "has gotten crushed" and I wouldn't disagree. However, that isn't remotely enough data to conclude that we should expect a wider than normal platoon differential going forward. In the minors, his platoon differential was pretty typical. He was extraordinarily effective as a AA and AAA starter. There were, no doubt, some lefty-heavy AA and AAA teams he faced in the 200 IP he threw at those levels over the course of 2007 and 2008. To be totally honest, 200 IP isn't remotely enough information to draw any firm conclusions about platoon splits either (as I understand it, you need about 600 IP, or 2000 TBF for platoon differentials to stabilize enough to be really useful), but it is more information than 46 IP. There is no reason whatsoever, statistically speaking, to think that DiFelice is so vulnerable against LHB that he couldn't start. It's one thing if you want to make the argument that starters need to throw more than one pitch, and from a scouting perspective a slow cutter that moves pretty much like a slider shouldn't be effective against LHB, but you cannot win the argument simply by citing DiFelice's MLB splits. I don't buy the scouting argument either -- I doubt he was throwing 83% cutters as a SP in the minors. He may have used that as his main pitch, but he also can throw a straight fastball into the 85-89 mph range, and has a curve and a change as well. He should probably be using those pitches more against LHB even as a short reliever, but I think that they have him pegged as some sort of freakshow one pitch wonder, which is, in my opinion, an unfortunate lack of imagination on the part of the organization (or Kendall, or DiFelice himself, or whoever it is that has decided he should just go out there and throw cutter after cutter).

Okay, that may have been hyperbole, but a guy can dream, can't he? Also, I thought the reason that they were kicking around getting rid of draft pick compensation is that its existence is the reason that changes to the draft have to be collectively bargained. My understanding is that with draft pick compensation for free agents eliminated, any changes the league wishes to make to the draft could be imposed unilaterally, since the MLBPA would no longer have any stake in it. I'll admit I'm not 100% sure that I'm right on this, but as I said, that was my understanding. Now, it could be that the players would refuse to allow the owners to get rid of the existing compensation system because they like having a say about how the draft is conducted. Sort of doubt it though, since the current system acts as at least a slight artificial drag on free agent salaries (having to forfeit a draft pick is, after all, a disincentive to signing a free agent. Unless you're the Sabean-run Giants circa 2001. In which case, getting rid of the pick is apparently the reason to sign someone as terrible as Michael Tucker in the first place. But I digress...). And no matter how clever their agents are about trying to convince the players that draft pick compensation is a good thing, I'm pretty sure that the players (or at least their union) could figure out the aforementioned.

I'm not remotely as big a fan of NFL or NBA as I am a baseball fan, but it is my understanding that both of those leagues have arrangements where teams don't really negotiate with draft picks at all -- you're drafted where you're drafted, and that pick is due x dollars over y years this year. Something like this is what I'm talking about. The drafted player would presumably still retain the right to refuse to sign and re-enter the draft a year later. But the only calculus a player in this position needs to perform is "do I think I'll be picked higher next year or the year after that?" since next year's offer would similarly be irrevocably tied to where he was drafted. "Signability" would be gone from draft lexicon immediately, except for HS guys who might want to go to college unless they go in the first couple of rounds. This would obviously help small market teams because, currently, they mostly adhere to MLB's recommended soft slots, and are therefore forced annually to pass on some of the draft's best players. Scott Boras would become instantly irrelevant. As for trading draft picks, I think you're spot on that the price for acquiring rental players would drop considerably. I also suspect that teams would often opt to pay in draft picks rather than current prospects, though selling teams will probably always prefer to get guys that are as major-league ready as possible. But the market would sort that out, and it's difficult to see an advantage in it either for small or large market teams over the system we currently have.

Exactly. He is annually among the top handful of hitters in MLB in percentage of first pitches taken. It has never stopped him from hitting before. For better or worse, that's how he likes to approach an at bat. It isn't like opposing teams are just figuring this out now, either. He's actually seen a lower percentage of first pitch strikes this year than the last two years by around 6%, as I posted in the Sveum thread.

If the reason they eliminate compensation picks is in order to institute a hard slotting system for draft pick bonuses and contracts, the good of this for small market teams significantly outweighs the bad. They should also allow teams to trade draft picks. That way, if your goal is to turn an impending free agent into a couple picks, it shouldn't be too hard.

Hitter BABIP is not luck-driven to the same extent that pitcher BABIP against is luck-driven, but you should still expect more variance in hitter BABIP than you do in plain old batting average (since strikeout rate and homerun rate are two of the more stable elements of hitter performance), and even plain old batting average isn't even reliable enough to stabilize over the course of a whole season, much less two-fifths of one. As far as whether opponents' advance scouts have caught on to anything in terms of Hardy taking every single first pitch, he's been in the top 5 hitters in MLB in percentage of first pitches taken ever since he broke into the league. If they are only catching on to this now, those are some truly DUMB human beings whose negligence of their basic job function borders on criminal. Additionally, it simply isn't true that they are throwing him a higher percentage of first strikes. That data is avaliable at fangraphs. For his career, he's at 56.2% first strikes. It was within 1% of that each of the last two seasons (his two really good offensive years). That has actually been a couple percentage points below league average, for anyone wondering. This year, Hardy has seen the first pitch go for strike one just 52.0% of the time, which is nearly 6% below league average. I too wish he'd stop doing it, but I've been wishing it his entire career. It is frustrating to watch, but he obviously believes it is the correct approach for him and it is difficult to argue that it hasn't worked pretty darn well so far.