by Gregg Pearlman
Saturday, October 16, 1999
The Giants haven't made much noise since the season ended, having forgotten to participate in postseason play and all. The good news is that in their National League Division Series, the hateful Diamondbacks were dispatched in four by the marginally less hateful Mets, who themselves are down 3-0 to the Braves in the National League Championship Series. The Braves, sad to say, defeated the Astros, who I thought would make better copy.
Meanwhile, the Red Sox ensured the firing of Cleveland manager Mike Hargrove by taking out the Indians in five, after losing the first two games of their American League Division Series. (I wish they'd just call it the "Semifinals," or something.) In Game 4, they scored 23 runs -- unheard of in the postseason. The Yankees -- it hardly bears mentioning, in the same way that newspapers rarely supply in-depth reports on the daily occurrence of the sun rising -- surrendered one run to the Rangers in the three games of their Division Series, and they've taken a 2-0 lead over the Red Sox in the ALCS. Game 4 is just about to start, and the noteworthy thing about that is that Pedro Martinez is facing Roger Clemens at Fenway Park. I figure on a Sox victory, just because I think Clemens is going to disintegrate, with the Boston fans hurling such hatred at him today. (And no sooner had I finished typing that than Jose Offerman lined a triple to the sun field in left. And now John Valentin has followed with a home run that just crept over the Monster. It's 2-0, as the fans continuously croon "Ro-ger! Ro-ger!" The good thing about this is that Vin Scully isn't covering the game on Fox, singing along with the fans, horribly off key, as he did with the Darryl Strawberry taunters in 1986.)
I figured on boycotting the postseason, because the season itself, as well as the troll-generated idiocy in the Giants newsgroup, really took it out of me. But the fact is, I've blown off the postseason before, and didn't enjoy it much. And the fact is, I simply love this sport too much. And I must say that I truly enjoyed watching the Diamondbacks fail. Not that I would've chosen the Mets as my preferred team to beat them, but, well, right now, the Diamondbacks are just about neck-and-neck with the Dodgers on the Teams Gregg Hates list -- even though, as I've said before, I really have nothing against the players themselves. Well, you know all about that, if you've been reading EEEEEE! lately.
Game 4 of the NLCS is scheduled to start at 4:42, with the AL game having started at about 1:30 -- all of which is hard to understand. I mean, baseball games routinely take over three hours these days, and I have no real belief that the AL game will end by 4:42. Why not just start the NL game at, say, 5:15?
Either way, the Yankees and Braves are poised for a rematch of the 1996 World Series, which is kind of a drag. Sure, the '96 Yankees won their first World Championship in 18 years, and their first pennant in 15, but this will be their third World Series in four years, and just about everybody I know who isn't a Yankees fan has been sick of these guys since the '60s. The Braves, somehow, haven't been in the Series since 1996, but the only time they haven't won the division since 1990 (when they finished last in the NL West) was 1994, when the strike hit with Montreal in first place in the NL East. So everybody I know who isn't a Braves fan has probably been sick of these guys since 1991. I know I feel that way.
My only disappointment in the postseason so far, really, took place in Game 1 of the ALCS, when Rod Beck entered the game for the Red Sox and promptly gave up a game-winning home run to Bernie Williams. ("Don't do it, Rodney!" I'd been shouting, right up till he did it.)
Well, that's not the only disappointment. The main disappointment has to do with the fact that there's way too many wild-card teams still left in the chase. It's possible that the Mets and Red Sox could pull off miraculous comebacks, and then for the second time in three years, a second-place team would win the Series. But it was tough not to pull for the Sox in the Division Series, because they showed so much heart. And for that reason, and the sick-of-the-Yankees reason, it's hard not to pull for them in the LCS. (Another reason is that I sort of wish they'd win a Series already, just so everybody will stop talking about their huge championship-free streak.) And, of course, I wasn't going to root for the Diamondbacks under any circumstances, so I had to root for the wild-card team in the Division Series.
Perhaps the best thing about all this is that it's nearly impossible for fans of New York teams to get what they truly want, namely a Subway Series.
What I figure will happen is that the Yankees will beat the Braves in six.
The anonymous person ("Anonymous") says, "Keep him and keep getting 16-10 seasons."
Many of us see Rueter as someone who's posted good won-lost records despite high ERAs, low innings pitched per start, more than one hit given up per inning, etc. We wonder how much we can trust the guy.
Richard says, "I would be a lot more enthused about keeping him if he would just last longer in games. I wonder if he even averaged six innings this year."
"Just under 5-2/3 innings per start average this year," Stefan tells us. "Almost exactly 5-2/3 innings last year. Just under six innings in '97. Just about 5-1/2 over his career. Folks, I see a trend.
I still think keeping him is a good idea. Trading a guy who is 44-25 over the last 3 years would require a little bit o' 'splainin."
"Among all the talk about how Rueter always has a high ERA, more hits than innings pitched, and just gets lucky in winning 15 each year, we are ignoring the true characteristic of Rueter as a pitcher," Anonymous says.
"On the contrary," says Steven R., "everyone who obsesses about his won-lost record is missing the true characteristics of Rueter. He continues to be an extremely lucky pitcher whose perceived value is much greater than his actual value. For this reason, he would make great trade bait. I mean, how much evidence do y'all need? The guy's ERA this year was 5.41."
"Translate this to 2.5 in good outings (of which there were many) and about seven innings per outing, and 8.5 in his bad outings, which there were fewer of," says Ethan. "If you prefer more wins than losses, you keep a guy who has come through three years in a row and doesn't let bad outings get him down."
"This kind of situation is exactly what I think Support-Neutral Won-Lost Record (SNWL) is really good for," says Paul L. "It evaluates players' outings on a start-by-start basis, rather than the aggregate (as, say, ERA does). It helps to reduce the undue influence of bad starts on ERA.
"Unfortunately for Rueter, SNWL is not very kind to him -- he comes out at 10.5 wins, 12.9 losses, with 0.55 wins above replacement."
Now, those of you who haven't heard of SNWL just went "What?" But it's a measure developed by Michael Wolverton of Baseball Prospectus, in which he defines it thus:
"A starter's Support-Neutral W/L record (SNW/SNL) is his expected (in the statistical sense) W/L record -- how many games he would be expected to win and lose given his pitching performances, assuming he had a league average offense and bullpen behind him."
There. Easy-peasy. Right? Well, I think I get the concept okay, and that's all that matters for the purposes of this discussion: "Here's what a pitcher's won-lost record should be, with all other things being equal."
Ethan says, "I really think Dusty doesn't manage him well enough. Certainly you want to leave your pitcher in for a while even when he's getting hit, to see if he can overcome it. But when after three years he still just has outings where he's got nothing, keep a long reliever ready for his bad days (e.g., Miguel Del Toro).
"He isn't giving up lots of runs in the 14-16 games he wins each year. He routinely dominates some pretty good teams.
"One or two statistics don't tell all about every player, and Kirk should not be evaluated based on average earned runs yielded. Maybe I'm wrong, but isn't it the win statistic that counts, and not ERA?"
"If you prefer more wins than losses, you get pitchers who keep the other team from scoring runs," says Steven. "Let's put it this way, it's a good thing Rueter doesn't let bad outings get him down, because otherwise he'd be pretty depressed a lot of the time.
"How about this one? People complain about the quality start statistic because it's too 'easy.' [A "Quality Start," statistically, is defined as "at least six innings, while surrendering no more than three runs." Sounds simplistic, but the fact is, good pitchers tend to have quality starts far more often than bad pitchers. -- GP] Well, over the past two seasons, Rueter has made 66 starts. He's had 32 quality starts in that same period. So more than half the time, he doesn't even have a good enough outing to qualify for a quality start.
"How about this one? Rueter's road ERA this year was 6.26, third worst in the entire National League.
"How did he win so many games? Same reason as previous years: he was third in the league in run support."
"For some reason he is always near the top in run support," says Ben F. "There has to be a reason why the Giants hit for him and not for someone else. Maybe because he pitches so quickly."
"It's the ears," Carlos guesses. "They have a soothing effect on his teammates."
I have the feeling that's as good a guess as any.
"Woody isn't consistent," says Anonymous, who's talking about Rueter as opposed to our Woody, of "Rookie Sleeves" fame. "He has good stretches and bad ones. When he is on, he dominates games. When he isn't, he gets nailed hard, fast, and often, like a three-dollar whore. But every year the good outweighs the bad, and his record looks good despite an ERA that muddles the good and the bad into a couple nonindicative stats.
"Sure, he gets good run support, but he does what it takes to win games. In low-scoring games, he keeps the other team down, and when the Giants do score lots of runs, he can still do what a pitcher needs to do.
"Plus the run support is also indicative of how much he helps himself at the plate. Despite his awkward swings, he makes consistent contact, and comes through in the clutch from time to time."
Mr. Bomb says, "Three points:
"Perhaps his high ERA is related to the high run support," says Ken K. "Say he gets out to a 6-1 lead in the fourth inning. Rueter isn't going to walk anyone. He might give up a couple of homers, but he isn't going to walk two guys and then give up a homer. He leaves the game in the seventh with a 6-4 lead. The Giants bullpen gives us a scare, but holds on for a 7-5 win. Rueter's ERA goes up, but he wins.
"I've also seen Kirk help the offense a couple of times despite his lousy batting stats.
"Or maybe he's just been lucky for the past three years. Hell, if he's that lucky, let's keep him around. I wouldn't trade him for one of those Dodger pitchers who have good run support, lower ERAs and still manage to have losing records."
"Why is everyone going to such much trouble over this one?" says Steven. "On the one hand, you have Rueter's performance: subpar ERA, allows lots of baserunners, generally pitches worse than many other pitchers. On the other hand, you've got a lot of what-ifs and maybes. Honest, the simplest explanation is often the best.
"[The idea that maybe he's been lucky is] an argument in favor of trading him. Luck cannot be expected to last... it's not real. Right now, you've got a pitcher with impressive won-lost records who might fool some GM who spends his spare time posting odd pro-Rueter messages on the Giants newsgroup. Trade him for someone who can actually pitch, before the secret gets out. Brad Radke's gone 12-14 each of the last two seasons... you really think Kirk Rueter is a better pitcher than Radke?"
"I don't think anyone here is saying that Kirk is a great pitcher, but we are looking for ways to explain the high ERA and the good winning percentage," says Ken. If a pitcher is up 7-0, he can afford to give up four runs. If a pitcher is up 3-0, he cannot afford to give up four runs.
"With the Giants' run support, I suspect that Rueter has been on the mound with big leads often and has given up runs that did not cause his team to lose.
"Dusty has probably left him in the game too long several times with big leads when he has tired out.
"I looked up Radke's numbers. He's given up fewer walks and has more strikeouts in fewer innings. He also has a 3.75 ERA compared to Rueter's 5.41. Plus he's two years younger. Sure, I'd make that trade in a second.
"That said, I'll still bet that Rueter's ERA goes down next year and that Radke's goes up."
(Actually, I bet Rueter's doesn't go down, at least not much. He'll be playing half his games in what I'm reasonably sure will be a hitter's park -- maybe a good hitter's park.
) James F. says, "If someone is continually lucky, they're just being them, and there's some factor you've missed.
"Not true," says Steven. "There is no such thing as 'luck,' for one thing... there is no one on the face of the earth who was consistently lucky throughout their life 'just being them.' Of course, there can be a missing factor, you're right on that."
I don't know that there's no such thing as luck -- it's a matter of philosophy, I guess. Faith, even. But it's not something one can reasonably quantify.
"But there is no clear evidence of anything beyond normal statistical variation going on here. It is nowhere near impossible for a person to have three 'lucky' seasons in a row. It happens sometimes.
"The thing we're trying to do is figure out what's up with Rueter. The possible explanations have included:
"Ten years in a row?" James says. "The probabilities are severely against it."
"Kirk Rueter has not been 'lucky' for 10 years in a row," Steven says. "He's only been in the majors for seven seasons, and didn't pitch more than 100 innings in a season until he came to the Giants. The proper frame of reference is the last three seasons. And since he posted a 3.45 ERA in 1997, I don't suppose he was all that lucky then, either. The last two seasons, Rueter had ERAs of 4.36 and 5.41 while posting records of 16-9 and 15-10. That's two years in a row."
"Possible," James says.
"Fine," Steven says. "It's possible. Until someone explains it as something more than 'maybe the hitters like Rueter more than they like all the other starting pitchers on the team the last two years and they somehow manage to hit better when he is the starting pitcher and this makes sense because XYZ,' then it remains an unlikely possibility."
[Now, the place where "they like him" theoretically falls down is: Jack Morris, the guy Joe Morgan cites as "knowing how to win," or whatever. Okay, I really don't know if his teammates liked him or not, but at least what we heard (via the press, of course) was that he was not a pleasant guy, if I remember right.
Here's what Morgan said: "Jack Morris is the perfect example of what I'm talking about. Morris finished last season with a 4.00+ ERA -- but he won 20 games! True, Morris pitched with a world champion team, but so what. All of those winning games were equal, whether the score was 1-0 or 10-9. The point is that Morris did what he had to do to win." -- GP]
"A fact," James says.
"No it isn't," Steven says. "At least, it hasn't been proven. His won-lost record is not proof that he 'pitches good enough to win.' It is merely proof that he's had excellent run support the last two seasons.
"All I can say is I think the first explanation is the most likely, given what we know. You don't have to do handstands to reach that conclusion, but you do have to stretch things to come to any of the other conclusions. I don't believe in 'luck,' I don't believe major league baseball players can force themselves to hit better for certain pitchers, and I don't think Rueter's record on a game-to-game basis justifies the assumption that he's 'pitching good enough to win.' That leaves the first explanation."
Which is pretty much where I'm at, too, for what it's worth. I just wonder if there's an X-factor -- something we just don't know about. Maybe several somethings. Or maybe we're looking for something that isn't there.
Thing is, "we," here, could well include everybody on Planet Earth. I doubt anybody in baseball really knows. If you asked the Giants' uniformed personnel, you'd get a bunch of different answers, probably mostly predicated on points 2 and 3, but probably with a handful (no more) of "I don't know's" thrown in. I'm guessing you'd get the same from the front-office people who are at all stats-savvy. If you asked a bunch of sabermetricians, I figure they'd cite point 1... but I don't know if you'd ever come up with anything definitive.
I kind of think it happens that way because it happens that way, and that there's no good explanation except "normal statistical variance." But I sure don't have an "answer," per se.
Tjames says, "You might want to take a look at Wolverton's (SNWL), which does as it says and removes run support from the equation, and breaks down each start individually, instead of in aggregate, which might be misleading given extreme outliers, and tries to predict a pitcher's 'true' record."
Tjames lists the Giants' starters by SNWL:
SNWL Actual
Brock 5-8 6-8
Estes 11-13 11-11
Gardner 5-10 4-10
Hernandez 4-3 3-3
Nathan 5-5 7-3
Ortiz 12-12 18-9
"Additionally, Rueter ranks as the sixth 'luckiest' NL starter in terms of difference between actual won-lost and SNWL. Ortiz ranks as third luckiest, and the Giants, as a team, rate second-luckiest team in the majors, behind only Texas."
"You know, I read the stats tables and some of the article, and while I understand how you could call a pitcher 'lucky' based on this kind of analysis, I don't really buy calling a team 'lucky' based on the sum of the starting pitcher's SNWLs (or even throwing the bullpen's effect in there)," says Richard. "A pitcher can be evaluated in a vacuum, but a team is the sum of all its parts, not just its starting pitching. If a team decides to invest heavily in offense and it wins by scoring a lot of runs to support average or less pitching, that's not luck, that's a choice, and not necessarily a bad one per se.
Richard says, "The best I can think you can imply about a team in this type of study is that its starting pitchers are lucky as a group, but you can't make such a qualitative judgement about the team overall based on this. I pretty much have learned to believe in the Pythagorean measurement, and the Giants were well within a normal deviation for that this year."
"You're missing the point, though," Tjames says. "The SNWL team rankings are just looking at team run support. The Giants score more runs than most teams, and it turns out the ratio between the number of runs the offense scores and the number of runs its pitchers give up is the second smallest in the league, roughly. If that's not your definition of 'luck,' fine; you can toss it out and call it whatever you like, but that's really all that was being measured.
"SNWL isn't a stat that magically measures everything: it looks at one thing only: how to evaluate pitchers when you remove their run support from their stat line."
"Oh, I got that point, and I don't challenge it," says Richard. "I'm just not comfortable taking an evaluation of a part and implying that it completely translates to an evaluation of the whole. Maybe this is just semantics, though."
Anson suggests that Rueter's record has been so good "Because he works fast. Of course that's not the answer, but I think that's at least part of the answer. By pitching quickly, he's getting the Giants defense off the field quickly, which means the opposing pitcher has less time to rest in between innings. That's the only logical reason I can think of."
"Has anyone looked to see if Rueter's games go faster than other pitchers' games?" says Steven. "It would be useful to know this. Also, people talk about how Rueter works quickly, but since he also allows more baserunners than the average pitcher, it's possible he keeps the Giants on the field longer than other pitchers. We need more evidence on this one."
Ron M. says, "Before you dismiss a pitcher like Rueter based on his ERA, take a look at his earned run median. A median is a much better indicator than a mean when trying to assess the 'average' behavior of a widely fluctuating variable."
"Well, I dealt with this in part when I mentioned quality starts," Steven says. "Over the past two seasons, Rueter has had more 'non-quality' starts than quality starts. The quality start is not the best stat out there, but if anything, it's too easy to get credit for a quality start, and Rueter fails that test, too.
"But if you think this is the explanation for Rueter, check it out... look at his starts for the past two seasons, figure the earned run median, let us know what you find."
Richard says, "Well, we are talking about luck and whether Rueter has it, but I think luck is something you can only impute in retrospect. If he ends up his whole career with such a disproportionate ratio of wins to ERA, then I think you can assume he was very much lucky, but it's hard to make that assumption comfortably as you are going along.
"I think (I don't have the Excel file right at hand), that the Giants' record in Rueter's starts was three or four wins better than you would assume from a Pythagorean standpoint, for whatever that is worth. Same for Ortiz." (This is pretty well indicated by SNWL, too.)
"I think that sums up my feeling, but opposite," says Mr. Bomb. "I don't think any of us would say the 1997 outscored, pennant-winning team was lucky. They just lost big much more than they won big."
"I'm not convinced whether this logic (of big variations in the runs scored/allowed ratio against the won-lost record) applies to an individual pitcher's performance nearly as well as it applies to a team's," Richard says. "I'm not agreeing or disagreeing with you here; I'm just saying that I don't know if the variation is correlates nearly as well regarding 'luck' as it does for a team. In other words, big variances may mean the pitcher was even more lucky for all I know (if this is making any sense at all)."
"But, as long as they won more often then they lost, they were good," says Mr. Bomb. "Same with Rueter. That guy gave up some big-time runs once in a while, but more often than not (and much more often than not over the course of his career), he has pitched well enough to get wins."
"And yet," says Richard, "my problem with him is not won-lost or ERA but too-short stints on the mound."
"When they stop using the won-lost record as the primary method of ranking teams in the standings, I might consider [things such as SNWL]," says James. "Until then, it's irrelevant."
"I kinda agree," says Ian. "The win-loss record is like the RBI stat. You don't get a whole bunch of ribbies unless the guys in front of you get on base. Look at Randy Johnson. He's gets no support, ergo, i.e, per se, my brotha, crappy won-lost stats. Oh yeah, hey D-bags! In your face! Woo-hoo! U! S! A! U! S! A!"
But to James, Tjames says, "Thanks for playing. In other words, 'Don't give me none of that new-fangled debate stuff, I've already made my mind up.'"
"Irrelevant stats aren't going to make me change my mind," James says.
"You could at least provide a position for that," says Tjames, like 'SNWL is worthless because....' What do you want? Should we consult tea leaves?"
Steven wonders why SNWL wouldn't change James' mind. "Rueter's won-lost record made you change your mind, and it's more irrelevant than SNWL to the current discussion: What did Rueter contribute to the Giants?"
"I haven't followed this point by point," says Mr. Bomb, "but why is this: Rueter has a good record, and wins a lot of games, almost every season. Is he just always luckier or does he get better run support than other pitchers almost every season? I agree that you can discount record in some cases, and sometimes there are better indicators of ability, but this guy consistently wins games. I just think that maybe there must be some factors that lead to wins (especially in baseball) you can't quantify. Then again, I also liked Joe Carter."
"Before we start assuming that there is something remarkable in Rueter's won-lost record, we need a large sample," says Steven. "For the last two years, Rueter has had excellent won-lost records without the kind of ERA you'd normally associate with such won-lost records. Two consecutive 'lucky' seasons are not enough of a sample for us to throw statistical knowledge and common sense out the window."
"If you're looking at won-lost versus ERA, then he's been 'lucky' two years in a row," says Anson, "but if you're looking at his higher run support over his fellow pitchers, then it's three years in a row, I think."
"Is it?" says Tjames. "I'm not sure if we ever got a straight answer on this one. He had a decent ERA in '97; it would surprise me if his run support were anywhere near as high this year as in the last two.
"Actually, here's his run support per nine innings for the last three years:
ERA RS W-L
1997 3.45 5.62 13-6
1998 4.36 7.91 16-9
1999 5.41 6.63 15-10
"So it's just the last two years where he's really been 'lucky.'
"I don't remember who was on the Giants staff two years ago to compare him against his teammates, though. Estes' run support in '97 was 5.96, higher than Rueter's. Gardner's was 4.64. Who else did you guys have in 1997?"
"Okay, so Rueter didn't get the highest run support in 1997," says Anson. "But as far as his support in '97 not being as high as the last two years, that's because the Giants didn't have a high-power offense till 1998. Especially after the Carter trade (then getting Burks), when they totally went nuts in scoring runs. Benard ended up with the highest batting average in the NL for the second half, and Bonds had that ridiculous September.
"Question: How does run support work? Does it only count during the innings that a guy pitched in a game, or does it include the runs scored even after he's taken out?"
My take is that it does -- that is, run support measures runs your team scores in games in which you start.
Steven says, "If Rueter consistently posted won-lost records that were much better than his ERA performance, we'd have to wonder why. But two years ain't 'consistently.' If he retires with a career record of 200-120 and an ERA of 5.97, I'll reconsider. For now, there isn't enough evidence.
"The larger problem is that every single explanation for Rueter's won-lost record the last two seasons, outside of 'normal statistic variation,' require leaps of faith:
"'There is something about Rueter that helps the hitters do better at the plate.' The 'somethings' being suggested are all tenuous and lacking in any concrete evidence... he works fast, he's popular, etc.
"'He pitches to win.' This is mostly meaningless, yet people talk about it as if it's an established fact that such a quality exists, that it can be identified, and that Rueter has it. What does it mean to 'pitch good enough to win'? Can we identify this quality in anyone in particular? Can we predict which pitchers will continue to do well in this area, the same way we can predict with decent reliability that Tony Gwynn will hit for a good average and Mark McGwire will hit lots of homers. Is it something concrete, or just something in the vapor that we point to when we can't explain something?"
"It seems like this question should be answerable, in much the same way that the clutch hitting question is (at least partially) answerable," says Paul. "Do a study, see if certain pitchers have the ability to generate above-average run support year after year outside the realm of random variation. It should be doable. I'd do it myself, but I don't foresee having too much spare time in my near future."
Steven continues: "'He's lucky.' I'm still waiting for a definition of 'lucky' that isn't circular, i.e. "Rueter is lucky. Lucky is defined as having a good won-lost record with a bad ERA.' That just restates what we already know. This is a serious question: What is luck, how do we know what it is, how do we know it exists more in some people than others? And if you can answer that last one, you're a better person than I am.
"Meanwhile, there's a perfectly sound explanation for Rueter that doesn't require the mental gymnastics of the above: he's a mediocre pitcher who has benefited from excellent run support the last two seasons, run support that is entirely random within the context of the San Francisco Giants and has nothing to do with his inherent abilities as a pitcher."
"What are we getting from our statistical arguments? Nothing," says Ethan.
Not true. We're hearing about different approaches in evaluating players.
"I rethink and agree. One thing I feel I should add is that, overall, it felt to me as though, overall, Rueter didn't pitch as well as previous years. Despite what I may have implied in previous arguments, I can't excuse entirely that he got pounded on many occasions. Yet he still had many great outings, and spread them out well enough to not have been stuck in any long ruts."
"I think a lot of his problems this year may have had to do with the fact that he really struggled with the new strike zone at the beginning of the year," Ron says. "Looking at his stats for this year or for the last two years will mean you're looking at a sample which contains a streak which is almost certainly not representative of how well we can expect him to pitch in the future.
"I have no problem with people saying that we should look for a better pitcher if one becomes available. That's always the case with the guys in the bottom half of your rotation. But given what pitchers go for, I think the Giants probably have more pressing priorities."
Ethan says, "On top of his won-lost record, the team's won-lost record in games he started was extraordinarily good.
"The most important thing, obviously," says James. "Anyone have stats for this?
"I wouldn't agree that the team won-lost record is the most important thing," says John B. "Just another thing to consider. The Giants were 21-12 in games Rueter started this year.
"I'd like to know one thing for now: Top lefthanded starter winning percentage since 1980 (minimum 75 decisions):
1. Randy Johnson 160-88 .645
2. Kirk Rueter 70-39 .642 3. Andy Pettitte 81-46 .638"I think there has to be a better explanation for the above than simply 'luck,' or 'good run suppport.' I agree with Stephen that there's no such thing as luck. And I don't know that Rueter has had unrivaled run support throughout his career. Personally, I suspect that Rueter is a good pitcher. Maybe not great. Certainly good, though."
Steven wants to know why there has to be a better explanation than "luck" or "good run support." John says, "Well, because it doesn't make much sense. To me, it seems more logical to think that perhaps Rueter, like Pettitte and Johnson, has been a fairly effective pitcher much of the time. Are you saying that you believe Rueter's presence on this list to be nothing more than a freak of nature?"
"On the contrary, normal random variation makes plenty of sense," says Steven. "It's not uncommon for someone to 'beat the odds' as Rueter has done for 60-some starts over a couple of seasons. Two years of starts isn't as large a sample as you seem to think. It's beyond exaggeration to describe this explanation as 'a freak of nature.' What doesn't make much sense is arguing that a guy with a subpar ERA, who allows lots of baserunners, doesn't strike many guys out, doesn't pitch very many innings per start, arguing that he's a good pitcher because he has gotten good run support."
"Well, the word 'good,' used to describe a pitcher, can mean a lot of different things to different people," says John. "Or maybe my range of what I would consider 'good' is broader than your range.
"In any event, Rueter's run support was nearly half a run less per start than that received by Shawn Estes. It was exactly the same as Russ Ortiz' (132 runs scored in 33 starts). So, I don't think anyone has demonstrated that good run support is responsible for Rueter's success. Also, I don't see where not being a strikeout pitcher precludes one from being a good pitcher. Another surprising thing is that Rueter had the second-best baserunners/inning ratio, on the Giants' rotation:
"Nevertheless, you're right in pointing out Rueter's high (and rising, for that matter) ERA, and his tendency to not make it into the seventh inning. These are concerns. And as you suggested, I would welcome Brad Radke (not that he's available) to the rotation, over Rueter, with great enthusiasm. But yes, I do think Rueter is a 'good' pitcher. Not the best on the staff. Not an untouchable, by any means. But, when he's on, he has his way of being effective. He hits spots. Hopefully next year, that will once again include the inside part of the plate. He changes speeds. He works quickly and gets a good rhythm going. And, in my opinion, he competes."IP H BB BR/I Nathan 90.1 84 46 1.44 Rueter 184.2 219 55 1.48 Ortiz 207.2 189 125 1.51 Hernandez 199.2 227 76 1.52 Estes 203.0 209 112 1.58 BR/I = baserunners per inning. (I realize this is rough: no hit batsmen, fielder's choices, errors, etc....)
"Two things about this [lefty winning percentage] stat as defined above," says Greg L.:
"The truth of the matter is that I think the strike zone, as called this year, hurt guys like Glavine, Rueter, et al., who depend on that ball being two inches outside the strike zone being called a strike. He doesn't throw very hard and doesn't have what one would consider exceptional stuff. And he doesn't pitch deep into games. Whatever his actual value, I would say his perceived value is higher. And those are the kinds of guys that I think a smart GM would use in a trade to upgrade his team.
"I happen to think that a pitcher's won-loss record is totally irrelevant."
"Totally irrelevant?" says John B. "As in, the stat should not even exist where pitchers are concerned? As in, being a 20-game winner is completely meaningless? In every instance?"
"Yes," Greg says. "Wins are a team stat. How do you compare the wins of a five-inning pitcher like Kirk Rueter against a guy who regularly pitches into the seventh inning of every game? The fewer innings a pitcher pitches, the more his win is a team win. Heck, in baseball, a pitcher can get a win without throwing a pitch. I'd call that a poorly defined stat that has no real meaning.
"Many of the guys who win a lot of games are good pitchers. However, it's not the wins per se that prove that they are good."
"So would you say that the stat had some meaning 30, 40 or 50 years ago, before the advent of widespread bullpen use?" says John. "I mean, you're right in pointing out the difference between someone who goes five-plus, and someone who goes 7-1/3. But in the end, the starting pitcher typically has far more to do with whether his team wins or loses than any other player on the field. Still, you do take how many innings a guy generally pitches into consideration when looking at the won-lost record, and you take the won-lost record with a grain of salt. But you don't completely ignore it. Or I don't, is what I mean. I'd call it well-defined, but misunderstood, and generally overrated. But still worthwhile in its own way.
"And I'm not saying that Rueter's won-lost record proves that he is good. A good won-lost record proves nothing, in my opinion. But I do think the won-lost record can hint at some things that other stats like ERA may not make apparent. I think that in Rueter's case, the won-lost record is a hint that there's more to him than meets the eye, just from looking at ERA. You have to look at Rueter start-by-start. Does that necessarily make his 15 wins this year comparable to Tom Glavine's 14 wins, or Shane Reynolds' 16 wins? No way. Is it possible he was just lucky? Sure it's possible. But I also think it's possible to attribute his success to something other than random chance.
As to the suggestion that Rueter possibly pitches differently, depending on the score, John says, "I've seen absolutely no evidence to suggest that a pitcher pitches exactly the same way, no matter what the score is."
"Fair enough," Greg says. "However, have we seen any evidence that pitchers don't pitch exactly the same way, regardless of the score?
John doesn't answer this -- or, if he does, that's not the bit I'm putting here. Instead, I have him saying, "The above numbers cover 161 starts, and seven seasons. How many starts does a pitcher need to have before it can be considered an adequate sample size? I think [Steven is] saying that Rueter's presence on this list is the result of his 'beating the odds'; that it's mainly a matter of normal random variation (i.e., he's lucky). Certainly that's possible -- and I can see where you'd come to that conclusion. His ERA was high this year."
"[Again], Kirk Rueter's won-lost record is even more irrelevant because he usually pitches six or fewer innings in the games he starts, which means he's even less responsible for the team win than his already irrelevant 'win' that he gets would imply. His 'wins' are dependent on the bullpen and offense (a.k.a. not Kirk) keeping the lead after he leaves the game."
"This is a valid point," John points out, validly. "This is one of many concerns I have about Rueter: his deplorable lack of stamina. Still, if you're going to dock him for his high ERA, it might also be fair to point out that his lack of stamina actually may be contributing to his higher ERA. Put it this way, Ortiz runs out of gas in the eighth, has runners on first and third. Who comes in? Embree or Johnstone. On the other hand, Rueter runs out of gas in the sixth, has runners on first and third. Who comes in? Tavarez or Spradlin."
"Dammit, don't make me think about things like that," Greg says. "But you make a good point. A pitcher that can pitch deeper into games can help himself just by giving the ball to better pitchers."
"Also, his run support can even contribute to his high ERA," says John. "Look at his final start. He goes into the sixth, leading 7-1. He runs out of gas, but because he has such a large lead, Dusty leaves him in long enough to give up three runs. I could be wrong, but it just seems like every time Rueter has a 2-1 lead in the sixth, he sets the other team down without scoring. But give him a 7-1 lead, and he gets sloppy. Just my opinion.
"I may not care that much how many runs a pitcher gives up, as long as it's fewer runs than his own team scores. But I do care about how many innings a pitcher goes. A guy who can generally get you into the eighth inning has so much greater value than a guy who only goes six, and after that it's pot luck. This is what really drives me crazy about Rueter. I think he pitches pretty well most of the time. This year he allowed three or fewer runs in 20 of his 33 starts. That's not bad. But how much more valuable would that be if he were to do that while averaging seven or more innings per start, instead of the less than six that he averages? Rueter's abbreviated starts really put a strain on the bullpen, in my opinion. If there was some way to get him to average, say 6-2/3 to 7 innings per start, I'd lose all my ambivalence about him."
"I agree with everything you say here," Greg says. "In fact, I think he contributes an awful lot to the bullpen strain problems of the last three years. Estes the last two years, too."
Ethan says, "As far as stats go, he managed to deliver a high amount of wins for the team despite being inconsistent, and getting nailed more than in years past. This impresses me.
"I also concur somewhat with the soothing-ear hypothesis."
"Say you have two pitchers," Greg says. "One of them (we'll call him Mark Gardner), has a RA [that's run average, as in earned run average, but with the unearned runs thrown in. -- GP] of 99.00. However, the Giants hitters score 1,000 runs or more in every start for him that year, so Gardner finishes with a record of 25-0 (and because Joe Morgan is the sole voter for the Cy Young award that year, Gardner wins the Cy Young)."
"Talk about piling on," Richard says. "He probably would have gone undefeated if they had just scored more than 200 runs a game, although Gardy might have taken a run at a loss once or twice.
"Next," says Greg, "we have another pitcher. We'll call him Gregg Pearlman. Gregg has a RA of 1.00. However, the Giants don't score a single run for him all year. He ends the year 0-25."
"Man, poor Gregg never gets a break, does he?" says Richard, who knows whereof he speaks, I'm sad to say.
"Gregg is the better pitcher, period," says my new best friend Greg L.
"Gregg, I probably wouldn't start packing for spring training just yet," Richard offers.
Well, here's where Greg's hypothesis kind of falls to the ground: I don't have any control over my fastball -- we're talking Steve Dalkowski without the speed; My curve seems reasonably controllable, but it's a big ol' roundhouse, and I suspect it'd get hit some; and I like to throw a knuckleball, but only because it's fun, not because it's any good -- I call it my "change"; plus I don't know how to throw any other pitches. Plus, I always get this twinge just below my right shoulder after about two innings, so I doubt I could even last as long as Rueter, most days. But I'll tell you this: It sure pisses me off that my boys just won't freaking score for me! I mean, I had a 1.200 OPS. Do I have to do everything myself? Dammit!
"A pitcher has control over only the runs scored against him, not for him," Greg says. "Whatever minimal effect pitching to the score might have on a pitcher's RA would average out over a 162-game season.
"Rueter has simply played for some very good offensive teams since he's been on the Giants (thus, the deceptively good and irrelevant won-lost record). However, since Rueter is likely not going to be traded anywhere (and particularly not to the Marlins, where the erroneosity of this argument would be clearly illustrated), he will remain a Giant."
"I'm not so sure I agree with this presumption," says Richard.
"And since the only way I can prove the trueness of the above to those of you firmly in the opposite camp is for either Rueter or the Giants' offense to start blowing large chunks, I hope that this statistical aberration continues nigh unto 2010," Greg says.
"Or even beyond," says Richard. "If it could extend to a couple of other Giant pitchers, that would be okay too."
Another factor here is, since Rueter's lasting into the sixth inning, that means he's getting scored for early on. And with the Giants winning a huge proportion of those games, that speaks well for the lads "in the clutch," now, doesn't it?
"Or how about, Gardner allows five runs in every one of his 30 starts, and finishes the year with a 6.00 ERA," John says. "Gregg pitches a shutout in all but one of his 30 starts. Unfortunately, his one bad outing (vs. the Dodgers, in the final game of the season, with the pennant on the line of course -- thanks Gregg.) results in 1,000 runs allowed, and leaves him with an ERA of 35.29. Once again, Gregg is the better pitcher. But this time, the ERA stat lies, not the won-lost record."
Hate when that happens. I just couldn't pitch through all the errors you SOBs made behind me. Not that I'm bitter. Creeps. Okay, fine, so they called them all "hits," just to screw up my ERA. Thanks, Baseball Gods. Thanks heaps. Thanks to you and my "teammates," I'm gonna get murdered in arbitration. Thanks, everybody.
"Though ERA has a tendency to even out over time for starters, it is not my favorite stat," Greg says. "It has weaknesses that have been discussed into the ground here."
"Just to be completely arbitrary, I'm going to break Rueter's season down into four areas," says Tjames:
You'd generally expect wins from his quality starts, and losses from his bad starts. That would give him a 9-6 record if he had a decision in each game. The games where he's most affected by his run support, therefore, are the good starts and the mediocre starts. That's 18 games in the balance, essentially decided by how well the Giants' offense is working that day.
Now to give some perspective, here's Ismail Valdes, using the same method:
"And here's Randy Johnson, who, if you're going by won-lost, should have only been two more wins and one fewer loss better than Rueter in 1999:
"I guess what I'm getting at is that if you're going to nitpick a Rueter's alleged superiority over a Valdes, the difference becomes crystal clear when you look at a truly excellent pitcher like Johnson. Why didn't Johnson win 28 games, then? Well, mainly because of this stuff; take a look at some of his no-decisions and losses:
Date Decision IP H R Score 4/05 ND 7.0 5 2 6-8 4/25 ND 8.0 5 3 5-3 4/30 ND 7.0 5 2 3-2 6/25 L 9.0 5 1 0-1 6/30 L 8.0 7 2 0-2 7/05 L 8.0 4 1 0-1 7/10 L 7.0 3 2 0-2 7/15 ND 8.0 6 0 2-3 7/25 L 7.0 8 2 1-2 8/06 ND 8.0 5 2 2-4 8/11 ND 8.0 6 3 7-5 8/31 L 8.0 4 2 1-2 9/18 ND 6.0 8 2 8-6"It's kind of hard to win when your team doesn't score runs when you're on the mound. But what else can you ask for in a pitcher? Johnson takes his team deep into games and keeps the opponents scoring to a minimum. After a certain point, 'pitching to the score' and 'knows how to win' are irrelevant. The pitcher is either effective, or is not.
"But to cut to the heart of the Rueter problem now: The main thing I notice about Rueter after looking at each of his starts: he's completely inconsistent. He gives up seven runs one day and then follows it immediately with a gem; the next time out he's hammered, or comes out early. There is no pattern with Rueter, except that he tends to leave games very early while holding a lead. He pitches well enough to win, in the sense that he tends to hang in there five innings to get a win. All the rest is mediocrity punctuated by few and far between moments of brilliance."
As for Valdes, John says, "I don't think I ever said, nor do I believe, that Rueter is a superior pitcher to Valdes. Valdes is a more skilled pitcher than Rueter. The numbers bear this out. I mean, have you seen the dookie Rueter tosses up there? Put it another way: If I were the Giants' GM, and Kevin Malone were to call and offer me Valdes for Rueter, straight up, I'd do that deal in a second. But I do think Rueter was more successful this year and, in my opinion, that was partly due to Rueter being a better competitor than Valdes was. All right, I know you're all rolling your eyes right now. I know you can't measure this kind of stuff. But that doesn't mean it doesn't exist, or that I can't offer up an opinion about it. And okay, I'll admit that maybe this is an unfair comparison. It's a lot easier to compete when you're on a good team, tougher when you're on a bad one. Rueter is on a team that generally is playing well, and scores lots of runs (four runs per game in Rueter's starts). Valdes is on a team that is struggling. Just curious, but what was Valdes' run support? Anyway, put Rueter on this year's Dodgers and Valdes on the Giants, and I'm guessing that Valdes would be transformed into a gritty competitor too. Maybe a better comparison would be Rueter to Estes:
Rueter Estes Ortiz Nathan
Quality starts 9 13 13 4
Good starts 11 7 14 8
Mediocre starts 7 9 1 1
Bad starts 6 3 5 1
Starts w/ up to 3 ER allowed 20/33 17/32 23/33 9/14
Giants' record in his starts 21-12 13-19 22-11 9-5
"Why didn't Estes have a better record than Rueter? He should have at least had the 13 wins from his quality starts, shouldn't he? Instead he's 11-11. It goes back to my opinion that Rueter is a better competitor. Maybe Estes is the guy the Giants should trade. Actually, I suspect he'd have more trade value than Rueter.
"What you say is true, but what happened to Johnson was just weird. No question that Johnson is a great pitcher, and deserving of the Cy Young, in my opinion. But I don't think his results compare with Rueter's. Rueter's run support was not at all out of line with that received by other Giants starters. Johnson's I believe was way out of line, in comparison to other Diamondbacks starters. In fact, I'd guess that Johnson's run support had to be down among the worst in the majors. Amazing, considering how strong the Diamondbacks offense generally was this year. Again, I think run support stats would be helpful here, if we had them."
My own opinion in all this is that won-lost record isn't value-free, just that too big a deal is made of it. However, I can't say I know what to make of it -- which is probably the thrust of this whole discussion.
John says, "I guess what I make of it is is this: a good won-lost record, that is at odds with the team's won-lost record, or the pitcher's own ERA, or the run support he gets, might... might... be an indication that the pitcher is a 'gamer,' and he 'knows how to win,' or he (fill in your own cliche here)."
Okay, I will: "... is Steve Carlton that year he won, like, 27, and the Phils won, like, 59."
"Conversely, a poor won-lost record, especially one that goes against the grain of the other relevant numbers, may be a hint that the pitcher will pitch well, but do something to lose the game in the end. It's just something to file away when evaluating a pitcher. I mentioned Valdes. Another example is Mark Langston, who for a while during his career, was known as a very talented pitcher who would find a way to lose his starts in the end. The quote I remember about him, from his manager at the time, Dick Williams, was: 'Mark Langston doesn't have gut one.' I think that was true at the time, but I do think he became a more effective pitcher as time went on. I'm not using Williams' quote as an explanation for the home run Langston gave up in the Series last year. Personally, I happen to think he struck Tino Martinez out.
"Here's a question -- can a pitcher 'bear down,' or does he pitch exactly the same way all the time, no matter what the score or situation is?"
I suppose ideally he'd pitch the same way all the time, but it doesn't happen that way -- whether you're talking about approach, mechanics, or actual performance.
"Well, where mechanics are concerned, I'm sure every pitcher seeks nothing but consistency. I guess I'm talking about concentration. Some pitchers seem to lose it when faced with pressure. Others seem to kick into another gear."
This is true. Atlee Hammaker, for instance, went into a tizzy when faced with a speedy baserunner at first. Just couldn't cope.
Which just brings into account the idea of how much a guy's mental makeup comes into play. I think we've kind of written Shawn Estes off, for instance, as a talented nutball -- but is that really the case? Is he just not tough enough?
"Who knows? I mean, we all have our opinions about this type of 'intangibles' stuff. The problem is, it's just opinions. There's no way to prove or disprove anything anyone says in this whole realm. Because of this, I think there are some who feel that the whole topic should not come into any evaluation equation. Others get so caught up in all the mystical stuff, that they lose sight of hard facts. What I'm arguing is that Estes' mechanics, his ability to throw strikes, and throw his curve at any time in the count, his ERA and strikeout-to-walk ratio, are important to look at. But so are his mental approach, his heart, and his emotional toughness. These are all factors, in my opinion. Problem is, one set of factors can be quantified and studied. The others are pretty much a mystery, and can only be guessed at. But that doesn't mean they don't exist, or that they aren't important factors in whether a pitcher is successful or not."
And the part we *can't* quantify, we attribute to "heart" -- right or wrong.
Ron brings the conversation back to ERA:
"I posted a fairly lengthy article trying to explain just how misleading ERA can be when trying to judge a pitcher, because it's a mean, and means are unduly affected by outliers. In it I tried to show how a pitcher could have an ERA well over five, and still put his team in a position to win three-quarters of the games which he started -- assuming average (4.5/game) run support.
"Before you dismiss a pitcher like Rueter based on his ERA, take a look at his earned run median. A median is a much better indicator than a mean when trying to assess the 'average' behavior of a widely fluctuating variable.
"For example, a pitcher who pitched complete games in which he game up, respectively, 8 8 7 6 4 4 4 3 3 3 3 will have an ERA of almost five, but assuming he's on an average team, offensively, you could reasonably expect him to go 7-4 during those games. His Earned Run Median is a much better indication of how much he brings to a team than his Earned Run Mean is."
This is a good point. I think it may even apply more to relievers, for whom a single bad outing can be devastating, statwise. I mean, you go eight scoreless innings over six appearances, then give up five runs in zero innings pitched... and suddenly you're at 5.63.
"Unfortunately, I don't think that MLB addresses this with any stat -- except for wins and losses, which many people are loath to rely on because of the run-support element."
This kind of goes along the lines of something I've tried, here and there, regarding closers: a kind of point system for save situations, based on "difficulty factors." What Ron's talking about is kind of like giving points based on the quality of the outing.
"People who think that ERA is the be-all end-all statistic when it comes to assessing pitchers really don't understand stats very well. This isn't to say a low ERA isn't a great thing -- of course it is -- but a high ERA doesn't necessarily mean what your intuition tells you it means."
Well, I don't know that anybody is arguing that ERA is the be-all end-all. It's probably the best "readily available" stat, but I agree that a high ERA doesn't necessarily mean "generally bad."
Obviously won-lost record is relevant for teams. For pitchers, I don't know that it's particularly relevant, or at least indicative of performance, if for no other reason than because of the way wins and losses are defined. For one thing, why five innings? Why not nine? Or one? Or zero? The win stat itself can be kind of dangerous -- how many people have hurt themselves trying to go five? (Ask Orel Hershiser, back in 1990.) How many pitchers turn 8-0 leads into 8-6 leads in the fourth, then spend the rest of the season muttering after being taken out "without getting a chance at my W"?
Tony LaRussa, I think, might have been onto something a few years back when he did a sort of "double rotation" with the A's -- the starter would go four, the next guy would go four... that kind of thing. This might be a great approach (if modified somewhat) if there were no such thing as individual wins for pitchers, but as it is, ego gets in the way: "Chicks dig the W."
And if we had a better sense of just what effect pitchers had on run support, we'd have more of an idea of just how valid a pitcher's won-lost percentage is, but my feeling is that individual won-lost percentage means very little, and what you have to pay attention to are mostly ERA, walks per nine innings pitched, hits per nine innings pitched, opposition OPS (if you have that data), stuff like that -- especially because individual wins and losses are defined pretty arbitrarily, and sometimes a pitcher gets a win or a loss for stupid reasons. I mean, to tell me a pitcher "knows how to win" when he gives up eight runs in five innings -- that's pretty silly.
A lot of people feel that "run average" -- i.e., runs per nine innings -- is a better measure than ERA, because even though a lot of unearned runs aren't the pitcher's fault, sometimes an early-inning error leads to, like, an eight-run inning... and how is the pitcher not responsible for that? I mean, he's the one who couldn't pitch through the error.
Strange how people don't pay much attention to RA -- it's easier to calculate than ERA: no subtraction.
Steven says, "I think I've got a huge blind spot here. I simply can't imagine how a pitcher can have an effect on the number of runs his team scores, beyond what he does at the plate."
Well, this is really how I feel, but I'm game. I just want to know how, for instance, "pitching to the score" could possibly be legit. My sense is, it's not -- but if it is, I'd love to know how.
"Maybe when a pitcher has a five-run lead he'd throw more fat pitches instead of nibbling for the corner strikes to avoid giving up free passes?" says Anson. "This, of course, doesn't apply to Estes."
I don't know that I respect the philosophy, though. I don't buy the notion that any lead is completely safe; you can't just run it into the line three times to kill the clock, then punt.
How often do we hear about a pitcher losing concentration with a big lead? I think you have to pitch like you mean it. Sure, challenge hitters... but you have to do that anyway.
Yet again, part of the problem has to do with the way wins are defined. I really don't know how much thought went into it. Nor am I sure how to make the win stat better. But it's got holes.
"I don't think anyone will argue with that," says John. "Then again, ERA has holes too. Not as many as won-lost, I'll grant you. Perhaps it could be said that won-lost record is to ERA what batting average is to OPS. But that would be doing batting average a disservice. There, ya see? I'm not saying that won-lost is all that wonderful. I'm just saying that it has some value, and can be worth noting. In Rueter's case, I think this is true.
"If there are pitchers who have the ability to bear down in certain situations and even certain games, I think that could provide some explanation as to why a guy might have a good won-lost record, despite his other stats being poor."
I suppose it'd help to know, say, ERA in wins vs. ERA in losses. (Yes, I would expect the latter to be higher....)
I also wonder if there have been year-by-year studies done on individual pitchers' run support. I wonder how often it happens where a guy gets great (or rotten) run support consistently. Or, is it more common for a pitcher's run support to fluctuate from season to season?
Plus, I really don't know what psychological effect, if any, a pitcher can have on his teammates in terms of run support -- I mean, "We like Woody (Rueter, not the newsgroup's and EEEEEE!'s own Woody) so we're going to score up the wazoo for him... but that Shawn Estes makes us crazy, so if he thinks we're gonna score for him, he's got another think coming!" How often does that happen? (Really, it'd be more like, "We like playing behind Rueter because he works quickly and throws strikes; Estes makes us lose our concentration.") I mean, while all that may be true, I wonder what effect it really has on the offense."
Richard mentioned the suggestion that maybe Rueter gets better offense behind him because he works so fast and the position players get back to the dugout quicker. "I don't know if this is true or not, but I am wondering if maybe that is why he flames out so early in games. Maybe his fast pace affects his own recovery time between innings, which would wear him out faster."
Another good question.
Clearly there's a call, then, for a study involving run support and game length. This would be a bit tough to compile....
Steven says, "Has anyone looked to see if Rueter's games go faster than other pitchers' games? It would be useful to know this. Also, people talk about how Rueter works quickly, but since he also allows more baserunners than the average pitcher, it's possible he keeps the Giants on the field longer than other pitchers. We need more evidence on this one.
"Since Rueter walks far fewer batters than Estes, Gardner, and Ortiz, his games must go faster than the rest," says Ben.
It might help to have an idea of how many pitches these guys throw per nine innings. That might give some indication of game speed. (Well, it would with Rueter, I guess, because we know he works quickly. But it wouldn't surprise me to find out that, say, Armando Reynoso throws fewer pitches per nine innings, while still being the slowest worker around, or close to it.)
In any case, let me wind up with my touchy-feely feelings on the subject, which re that won-lost record doesn't tell the story, but I'm not prepared to say what really and truly does -- I mean, as Ron said, ERA doesn't tell the whole story, either. There are a bunch of factors -- many of which have shown up in this discussion -- that need to be considered, including per-start quality and run support.
I had an interesting discussion about this thread with Greg Spira of Baseball Prospectus. Evidently there's quite a correlation between run support and won-lost record -- I would guess (though I don't remember if I asked) that this is a higher correlation than that between won-lost and ERA.
A good example is John Burkett, who went 22-7 in 1993, then 6-8 in '94 while having essentially the same season otherwise. This is the thing I mentioned a while ago, where Al Rosen (Dammit! Not Al Rosen, Bob Quinn! I made that mistake when I talked to Greg, too....) wondered just why he should be expected to give a raise to someone who goes from 22-7 to 6-8, hence the trade to Texas. Greg did a quick check and saw that Burky's run support went from, I believe, 5.5 to 3.3. But did Rosen take this into account? Of course, not! He wasn't even around! But Quinn didn't bother with this, either.
Here's at least a partial list of factors that should be considered, and I'd love your input if you can think of any other factors:
And for the record, I'm not exactly on the "Trade Rueter" platform, though the idea intrigues me, if only because his wins alone might intrigue the kind of GM who'd ignore the high ERA and low innings-per-start.
What it really boils down to is that there are factors we haven't identified, factors that no one has identified, and factors that have been identified but are red herrings. Hence, I guess, the variety of approaches. I think Michael Wolverton's on the right track with SNWL, which -- unless I misunderstand the intent -- sort of tries to state "overall pitching performance" in terms of won-lost record, in the way that Clay Davenport's Equivalent Average (EqA) tries to state "overall batting performance" in terms of batting average. That is, SNWL isn't really about wins and losses; nor is EqA really about batting average. The only problem is that I don't quite get either measure; nor do I think I can improve on them.
First, there's been a bit of a shuffle on the coaching staff. First-base coach Carlos Alfonso has become the "administrative coach and video coordinator." Pitching coach Ron Perranoski has been named "special assistant to the general manager." Taking their places on the lines, respectively, are Robby Thompson and Dave Righetti.
Now, first, I called the Righetti move right after his brief stint with the big club this season, when Perranoski had to leave the team due to the death of his mom. Now, I'm not going to take total credit for it, as many other Giants fans called it too (probably), but the pitchers seemed to communicate well with him, and it's been clear for a while that some change was needed. Plus, as rotten as he was for the Giants, Rags was one of those guys you couldn't help rooting for.
The Robby move was probably equally predictable, and at least as welcome. Bear in mind that the Giants brought them both on at the same time, about a year ago, and I'm sure it's no coincidence. (Really, the key element here is that Robby'll be wearing a new number, because he'll be making coaches' money and won't be able to afford to buy number 6 off J.T. Snow.)
The concern is that Sonny Jackson evidently hasn't had his duties altered any. "Keep him off the lines" would be my advice. Clearly he's an old pal of Dusty's -- he was in Atlanta for Dusty's first three or four years -- so unless Sabes wants to dump him, it won't happen. I was relieved that he was made bench coach in '98, though, and see no reason for it not to happen again. Ron Wotus was no worse a third-base coach last year, and indeed he may get way better. Sonny doesn't seem to be improving at all.
In fact, Giants fans, let's try to keep track of his baserunner gaffes next year -- be fair, now! I bet we'll see way more than we hope.
Meanwhile, the Giants acquired Nelson Castro, a minor league shortstop, via waivers from Anaheim and added him to the 40-man roster. He replaces reliever Bronswell Patrick, who was outrighted to Fresno. "The 40-man roster now stands at 37, pending some sort of deal," says Marty. "Any speculation?"
Well, a lot of teams don't keep a full 40 anyway, but there has been some speculation about Mueller getting traded, and you'd also get that impression from reading between the lines in the Chronicle. Maybe if someone wants Mueller bad enough, Julian Tavarez can be a condition of the deal being made. I'd miss Mueller... but I would assume -- not trust, assume -- that he'd be replaced with someone with more pop.
Meanwhile, I know nothing about Castro, and I'm mildly surprised about Patrick... but that doesn't mean he's through (not that I care much). (I've since learned that Castro, from the looks of his numbers, is Just Another Middle Infielder. We already have at least one of those.)
"Anybody with the cojones to throw a 3-2 changeup to Sheffield... I think I almost want to see him stay," says Marty.
Richard says, "Don't know how to interpret this comment, but when Sabean was asked by Ralph Barbieri on his show yesterday about whether Tavarez would be back, he dodged the question, then laughed and said 'I think you know the answer to that.'"
The Giants have also sent Steve Connolly, Robbie Crabtree, and Steve Soderstrom outright to Fresno. Connolly's the Rule 5 guy they brought to spring training, the former Red Sox and A's guy, right? And Soderstrom... well, that's no big surprise. But I thought they were really high on Crabtree, and I guess the fact is, he ain't young for his level, and he didn't tear it up.
As for who could replace Mueller, well, as near as I can tell, the Giants don't have anybody ready at third -- Edwards Guzman? I'm kidding. Steven says, "My guess: Vinny Castilla. He's getting old, his stats have the Coors inflation, he's coming off an off-year for him, his team acts like they want to move him. Castilla has Brian Sabean written all over him."
"And you know, Castilla outside Coors is still going to be good for 25-30 homers a year and 100 RBIs on a team like the Giants," says Tjames. His OBP will suck, he won't be half the player he is at altitude, but he'd do just well enough to convince most folks that Sabean made the right move.
"Maybe that's Sabean's true 'genius': making mediocre guys look 'not half bad' instead of 'barely half good."
To a certain extent, Sabes will be driving even more blind than usual, because he doesn't know what the park will be like yet. For all we know, it may be death on righthanded power hitters -- meaning that Castilla's the only choice.
What's scary, however, is that there's been at least mild speculation about Ed Sprague. Yih.
Also, you should know that longtime trainer Mark Letendre has left to work for the umps, and he'll be replaced by conditioning coach Stan Conte.
In addition to some player bios and (rather limited) stats, the site also has scans of the Pilots' '69 and '70 Topps baseball cards. I have the '70 set -- the whole set, I mean; completed it in '89, I think -- so those cards aren't new to me, but I hadn't seen many of the '69 Pilots before. Kind of cool. The bios (there are maybe 20) also include black-and-white photos you probably haven't seen.
Oh, and there's a transactions page that's interesting, too. One thing I'd love to see an explanation for is how a player could be acquired "conditionally" from another team.
I think it's a great site for Ball Four fans, just because it helps attach faces to some familiar names.
"When no one is on base (an intrinsically boring state of affairs at all times) and the hitter hits a line drive into the gap, either the second baseman or the shortstop is allowed to pick up second base and run with it into the outfield as soon as the ball passes the outfielder."
"Cool," says Jeff D. "Stolen bases as a defensive stat."
Jim says, "The hitter/runner then has the option of either (a) tackling the base stealer before he gets to the center-field wall (which, if successful, will result in an inside-the-park home run; if not successful, the runner must return to first with a single) or (b) cutting directly across the diamond after reaching first base and trying to reach third base before either (a) the ball gets there or (b) the pitcher body-checks him to the ground (both knees must touch the ground). In the latter case, the hitter returns to first base with a single, the jeers of the crowd ringing in his ears.
"These slight modifications add the best features of football and hockey. Baseball broadcasters will then shout more often, like those who cover football and hockey games do (shouting, per se, causes fans to be more interested). The game will be saved."
"I've always felt that there had to be a place in baseball -- somewhere -- for the use of bolas," says Jeff. "Besides appealing to South Americans, it would greatly add to the excitement of the game if offensive and defensive players could be entangled and brought to the ground at any moment by the skillful toss of an opponent. The 'routine play' would become a distant memory."
What could help, too, is if certain fans -- I would be one of these -- were issued blow-guns loaded with mild tranquilizers. These, of course, we would aim at our teams' opposing pitchers just as they're about to deliver the ball. Every pitch would be accompanied by a soothing "Tfoo!"
Copyright ©1999 by Gregg Pearlman
Last updated 10/18/99 Gregg Pearlman, gregg@EEEEEEgp.com