May 2, 2003
Return of the Cripple Stick
I've said it before and I'll say it again: reader feedback is one of the great parts of this job. I'm lucky--everyone at BP is lucky--to write for an audience that provides thoughtful, articulate responses. It keeps reading e-mail from being a cringeworthy job, and turns it into an informative experience.
I read every piece of mail I get, from two-liners to two chapters. That I don't respond to it all is a flaw in my game, like my inability to handle a good fastball or make the throw from the hole. I'll keep working on it.
For today, here's a sample of the feedback, and a promise to spend an afternoon next week going through my "Reader Mail" folder and getting back to you.
I first came across Voros McCracken's hit rate theory last season. It took me a while to get used to the idea, but after some deep thought it stuck with me.
This year over on Ron Shandler's BaseballHQ site, there is an article by John Burnson that shows some preliminary study of groundball/flyball ratio and pitchers' hit rates. Burnson
ran data for 2002 pitchers who hurled at least 60 innings and came to the conclusion that a fly ball pitcher is more likely to have a lower hit rate than a ground ball pitcher. The regression trend line runs from an expected hit rate of .282 for a pitcher with a .51 GB/FB ratio and ends at .310 with a 3.65 GB/FB ratio.
It's an interesting read, and I was wondering whether anyone at BP had looked into such a correlation?
My understanding of Voros' work is that yes, groundball pitchers allow a higher batting average on balls in play than do flyball pitchers, but that DIPS ERA works because flyball pitchers allow more extra-base hits. The two effects cancel each other out.
Just wondering if the Voros research on pitcher hit rates would prove conversely true for hitters. It would seem to make some sense, but you guys would be certainly better to analyze that than me. It could provide a way to see if a hitter's batting average was due to rise or fall and could be predictive.
I don't know, I was just thinking it through and was curious if it had been looked at. The same variables would seem to apply as to pitchers: batter speed, team defense, park environment, contact rate, and line-up situations. Any thoughts?
There's no converse of DIPS for hitters. Hitters do control outcomes on balls in play. If you think about it, this makes sense: Someone has to control it, or balls in play would be completely random.
On a Variety of Power Hitters
You wrote: "The thing is, [Ryan] Klesko isn't a bad outfielder. He got that reputation while with the Braves, but it was largely a function of his size."
Baseball Prospectus 2001 has Ryan Klesko as -3 runs and -2 runs in left field in his last two seasons with the Braves, which conforms with my recollection that he improved to average-minus. He is a brutal, brutal first baseman, and I may be letting that fact polish my memory of his work in left field.
I have to disagree here. I watched a whole lot of Braves games (on TBS, not live, admittedly) during the Klesko conversion, and at first he was just awful. I mean, we're talking "makes Lonnie Smith look good" awful. Even after he stopped making the glaringly obvious misplays, his reaction times on flyballs were measurable by sundial. His arm was mediocre and erratic, he didn't play the hops or the walls well, he didn't charge the ball well...what was to like?
He certainly improved over that initial (abysmal) standard, but I don't think he ever advanced to the point of being "not a bad outfielder." Do the stats tell me I'm wrong?
-- Dave Tate
What attracted you to Adam Dunn, other than his youth and patience? Do you expect him to rebound from a poor 2002 (when his average and power declined, especially in the second half)? Won't his batting average kill you? It looks like Sammy Sosa, Brian Giles and Bobby Abreu (if you want to stay young) all went for around the same value.
I was insanely high on Dunn coming into the season, believing he would be a .300/.400/.550 guy, and in that lineup, get well over 100 runs and RBI. His performance has been spotty, but the bigger problem--I still think he'll be fine--is that the Reds' injuries mean he won't have the context-dependant stats.
He'll be a tough protect at $37, unless he proves to be at least a minor threat on the bases.
I disagree with your assertion that using Craig Wilson as the Pirates' backup catcher is a good thing. I think it only magnifies the problem created when both Kenny Lofton and Reggie Sanders were signed, squeezing him out of the outfield. Wilson needs at-bats. As the only backup catcher, he will probably only bat when filling in for Jason Kendall. The safe thing for a manager to do is to keep his last catcher on the bench in case something happens to the guy in the starting lineup. I have to believe the Pirates will follow this theory.
In retrospect, you might be right. I had assumed Wilson would play occasionally behind the plate, and still get time on the corners. He's started three times at catcher and six times in the outfield, most of those coming since the Brian Giles injury.
It's still good roster construction, but McClendon has to get Wilson more playing time. Wilson can help by hitting when he plays, but that's hard with sporadic at-bats.
I was reading the Tim Kurkjian article on ESPN.com, and he made a reference to Alex Rodriguez as the greatest shortstop ever.
Before Alex Rodriguez came along and Barry Bonds became BARRY BONDS, I generally listed Wagner right behind Babe Ruth. Now, those two may edge Wagner aside--Rodriguez in a few more years--and I've also reconsidered Ted Williams over the years.
This got me thinking about Honus Wagner, and I think he might be the most unappreciated player in history. The guy hit for average (35th-highest of all time), power (eighth all-time in doubles, and 101 home runs before the live-ball era), speed (third all-time in triples, 10th all-time in steals) and he played one of the three toughest positions in the game. In fact, since he played before the large glove and smooth infields, shortstop might have the most difficlut postion.
Now obviously I have not looked at all of these statistics in depth, but usually when one leads the field in so many categories, greatness is not far behind. In addition to being unappreciated, could Honus Wagner be the best all-around player in history?
With all the things that go into tearing apart the careers of the greatest dozen or so players in history, I don't know that we can say, definitively, who No. 1 is. The game has improved so much in 100 years that I would believe that the edge would go to more recent players, but that's a larger discussion.
I do think, in general, Wagner gets short shrift. He's the greatest shortstop ever, and if he wasn't Babe Ruth, he's in the mix with Bonds, Williams, Ty Cobb and Willie Mays for the spot behind him. The fact that he played in the deadball era, and his underappreciated defensive skill, work against him in these discussions, however.
Maureen Dowd's Favorite Pitching Staffs
I saw the article about left-handed heavy pitching staffs and found it quite interesting. One group I thought might make the list was the Blue Jays in the late 1980s. I remember at one point they had a rotation with Jimmy Key, John Cerutti, Mike Flanagan and Jeff Musselman, all lefties. From the right, they had some time from Dave Stieb and Jim Clancy. Did the percentage of left-handed starts not wind up being that high for them? They must have been close to 100 left-handed starts in '88 or '89.
I went digging. According to baseball-reference.com, here are the left-handed starts by the Jays in that timeframe:
-- Dave Kirsch
Jim Clancy, Dave Stieb and Todd Stottlemyre generally got enough
work to keep the Jays above 60 right-handed starts.
Sabermetric Stuff, Sorta
I came across a program posted in BP the past week which allows you to see the average number of runs a team will score in a season based on the players/lineups a team plays. I ran a couple of simulations using the PECOTA predictions for the 2003 Cubs, and was startled (well, not really that startled, considering I would've guessed) by how many runs the Cubs lineup would gain by simply switching Mark Bellhorn and Alex Gonzalez in the projected lineup--something absurd like 40 runs a year, I believe.
As a general rule, 10 extra runs equals one win. It's a little higher in the current era, maybe 10 1/2, but not as high as it was a few years ago at the peak of the post-1993 offensive surge.
I see these sorts of statements all the time, though--"such-and-such a team would gain 30 runs a year if they did so-and-so," but how many "gained runs" do you project to "gained wins"? Four runs per win? Five runs per win?
The concept predates my involvement in sabermetrics, but I know it's somethig Bill James came up with in the 1980s, and that it still holds true now.
It seems as though the outcomes of the last few World Series haven't fit in the sabermetrically correct model. The Angels are a prime example here. I believe this is because a sabermetrically-inclined team will succeed in the regular season, where 162 games allows the statistics to work themselves out, but not in the postseason. Any postseason team will face the opposition's best pitchers two or three times, resulting in one-run games where defense is truly at a premium. Games like this don't really mesh with a team like the A's, who used baserunners and walks to clobber opponents in the regular season (just as many times against a team's #4 or #5 as they did against a #1 or #2), but who in the postseason may have trouble small-balling it against a team's ace. Any merit here?
Also known as "The Joe Morgan Theorem."
-- Michael Goldman
The problem with this is that the A's haven't been outscored wildly, or had a ton of troubles scoring runs. They've lost three five-game series in the fifth game. That makes them 6-9 with a decent run differential in 15 games against top competition.
Not to mention the fact that the Yankees have been the best example of the "take and rake" philosophy, and they've had the most success. Their 1998-99 teams played this style and won handily.
This is a popular, but flawed, theory. The A's lack of playoff success comes down to a handful of specific mistakes, and doesn't serve to indict the team's assembly or concept.
I have a question about your AL East predictions. Why is that the Yankees fielding will sink them, but the Red Sox fielding (which is worse) won't?
Primarily because the Yankees' defense is worse at more critical positions up the middle. The Red Sox look to be pretty bad on the corners, but solid up the middle, and in fact are varying their lineup to a certain extent based on the starting pitcher. This should help them minimize their defensive issues.
-- Adam Dlugacz
Joe Sheehan is an author of Baseball Prospectus.
Click here to see Joe's other articles.
You can contact Joe by clicking here