Baseball Prospectus - Chat with Keith Woolner

Chat: Keith Woolner

Welcome to Baseball Prospectus' Wednesday April 06, 2005 7:00 PM ET chat session with Keith Woolner.

Keith Woolner is an author of Baseball Prospectus.

Keith Woolner: Hi everyone, let's get started a few minutes early so we can work in a couple of extra questions...

Tell Me Lies (Houston): Do the Stros have a shot at making the NLCS again this season?

Keith Woolner: Sure they do. The Astros, despite losing Beltran, are not a horrible team. And in a division where no team looks like a sure-fire powerhouse, any team who catches fire at the right time could sneak in.

That's similar one of the points I made in my Setting the Stage article (http://www.baseballprospectus.com/article.php?articleid=3861) -- when teams aren't that far apart in talent, predictions are particularly messy.

dangor (New York): Your 23 Burning Questions Essay last year was one of the best articles on baseball that I ever read. I have a couple of silly ones, 1) Why doesn't every manager have his LOOGY play first base for a batter or two when the rightys come up in an effort to get him more playing time, 2) Why does the runner slide into second on a force out when it's first and third and two out?, 3) Why don't more pitchers throw the knuckler?

Keith Woolner: I think you're referring to my "Baseball Hilbert's Problems" article, that originally appeared in BP 2000, and was posted on the website last year (http://www.baseballprospectus.com/article.php?articleid=2551)

1) If the pitcher is athletic enough, that might not be a bad idea. It's kind of the next step in the Brooks Kieschnick direction. If a guy can play a position and pitch, you can get more value out of your roster slots. If you have two of them (one righty and one lefty), you could have both in the game and, to the degree the rules allow, swap them around.

2) If something happens on the force out at second, and the fielder drops the ball, you don't want to have overrun the base and get tagged out.

3) The knuckleball is a difficult pitch to master, and may take years of work to learn it. If I recall "The Diamond Appraised" correctly, a pitcher's mechanics are particularly crucial in throwing the knuckler too.

caspian88 (Salinas): Does PECOTA take into acount hitting when it projects pitchers and identifies their top comps? It seems to me like it should, in order to find a better comp. There have been quite a few pitchers I would count as being "good," being consistently at or above league average for a pitcher, and perhaps this is something to look into if it is not already included. Maybe good hitting pitchers tend to also be average to above average pitchers, like Bob Lemon, Livan Hernandez, Mike Hampton, etc.

Keith Woolner: Nate Silver is the PECOTA guy, and according to him PECOTA doesn't account for pitching hitting. As he says "There are enough variables to worry about as it is."

caspian88 (Salinas, CA): I recently read your article on gauging catcher game calling, and I am wondering whether you have done any recent work on that front. Also, is it possible that one or both of the following scenarios explains the phenomenon of there being little to no effect: either all of the bad game callers were weeded out in the minors, or even before, being moved to new positions or outright gotten rid of, or that game calling simply isn't that hard of a thing to learn? If the first scenario, then wouldn't that make it much more difficult for the idea you proposed - namely, playing Mike Piazza and Frank Thomas at catcher - to be feasible?

Keith Woolner: The trio of articles I've written on catcher game calling (http://www.baseballprospectus.com/article.php?articleid=432, http://www.baseballprospectus.com/article.php?articleid=436, http://www.baseballprospectus.com/article.php?articleid=1489) all looked at pitchers with particular catchers from year to year. But they all did so at the major league level.

That means that there could be a weeding out process in the minors -- and indeed a lot of catchers are converted to other positions (Josh Phelps, for one). In many cases, this is done for perceived defensive shortcomings.

I think I was pretty clear to point out that the Thomas/Piazza example was a stretch. And especially with the differences in controlling the running game that I presented in BP2004, even if game-calling isn't a particularly valuable skill, that doens't mean that any noodle-armed 1B can handle the position.

That being said, I am aware of, but have not yet read, another bit of research that may show catchers improve in lowering their pitcher's ERA with experience over multiple seasons. If it holds up under scrutiny, that would be a fascinating result.

Cris E (St Paul, MN): Have you ever twisted VORP so that instead of comparing within specific positions you compared within like positions (corner IF, OF, middle IF)? On a related note, ever try to create a utility position for players who appeared in several spots for more than a nominal amount of games?

Keith Woolner: VORP as it's currently formulated weights the positions the player actual appears at, and so to some degree handles players who appear at multiple positions.

That's not a complete answer, because the ability to play multiple positions well in the majors creates roster flexibility and there's value in that that VORP doesn't capture. In fact, a guy who can play 2B and 3B adequately is penalized by VORP when he plays 3B versus a player who only knows how to play 2B, because the offensive level at 3B he's compared to when he's there is higher.

I have thought about positional groupings, particularly for LF and RF, since RF offense tends to be higher, yet is thought of as the more difficult defensive position. Since the skill sets for corner outfielders are similar (other than RF needing a better arm), it's possible that general athleticism among the pool of corner outfielders correlated with both offense and defense, which would mean VORP underrates RF, and overrates LF.

Simon (Vegas): Who is the next name in this sequence: Roth, James, Woolner, _______

Keith Woolner: Sponge Bob.

Tom (NYC): I don't care....Rivera is my #1 choice to close vs. Boston until he retires. Gladly.

Keith Woolner: Though I'm a Bosox fan, I have to agree. Unless something is physically wrong with Rivera, there's no reason to think he's gone all Schraldi on us. The Red Sox have a good lineup, and he should be expected to blow a few saves against them. And as a Red Sox fan, I'll enjoy each one immensely.

Cris E (St Paul, MN): Quickie: What's you favorite place to sit when watching games (other than behind the plate), and why?

Keith Woolner: When I lived in the Bay Area, I used to go to A's games several times a year, and always liked the $5 seats in the upper deck behind home plate. You have a good view of the entire field, somewhat akin to the angle in a baseball video game, but aren't as far away from the batter-pitcher action as a bleacher seat would be.

If I get spendy and can't get seats behind the plate, I'll take them on the 1B side. You aren't blocked from the pitch by right handed batters, and there tends to be more action at 1B than 3B, so you're closer to what's going on overall.

Cal Ripkin Jr (6' 4" above the earth): What is this I keep reading about tall guys not being able to catch? It's been bugging me and I was hoping for something meatier than "It just hasn't been done."

Keith Woolner: As I understand it, the problem isn't that tall guys can't catch, but that they can't last long doing so. They physically break down sooner.

This makes some sense from a physics perspective -- if you envision the knee as the pivot point, longer thighs and potentially greater upper body weight further from the pivot create more torque, which stresses the knees. Thousands of repetitions later, you have no cartilage. Will Carroll can correct me if I'm way off base.

That being said, the question reminds me of the movie Crazy People, with Dudley Moore running an ad agency inside a mental institution. One of their slogans was: "Sony: Because Caucasians are just too damn tall."

Fausto (Emeryville, CA): I have a question about DIPS. If you take a look at Voros's original article at Baseball Prospectus, he chooses a set of data that is really a little bit anomalous. Does it seem to you that people seem to be a little bit overzealous in claiming both the significance and accuracy of this particular tool?

Keith Woolner: DIPS aficionados have been overzealous, but that's not Voros's fault. I think his work in becoming popular in the sabermetrics community has been oversimplified.

In fact, I wrote a counterpoint to Voros when his work originally appeared on BP: http://www.baseballprospectus.com/article.php?articleid=883, and he and I both answered some reader questions soon afterwards: http://www.baseballprospectus.com/article.php?articleid=890 that addressed some of the nuances of the argument. Of course, those have largely been lost in favor of the catchy "pitcher have no control on balls in play" misstatement.

But I think the backlash to DIPS has been a similar overreaction. That pitchers have noticeably smaller ranges of ability in preventing hits on balls in play was a major realization. And I think that in the intervening years some people have forgotten how much in dark we were beforehand.

marlette (reno): Not-a-question: I'm pretty sure Whitey Herzog used to put Worrell in the outfield for an out or two, while the lefty (Dayley? I was eight) pitched, to save the bench spot. Question: do the Red Sox "have MRivera's 'number,'" as the Steves claimed earlier today? More to the point, is there any evidence that prolonged exposure to one team damages pitchers' performances, rather than the hitters'?

Keith Woolner: Good point on Herzog -- I was almost exclusively an AL fan back then, so had forgotten about Worrell.

As to the question of prolonged exposure favoring one or the other, my intuition (and, I believe, the common wisdom) is that experience favors the hitter, because they are always in the reactive position. Any time you're in a position to better anticipate what's coming before you see it, the better off you'll be. I don't know of a systematic study of the topic though.

On a semi-related note, I think someone (Steven Jay Gould? Pete Palmer?) had previously noted during eras where the rules of baseball were relatively constant, the hitters gradually adapted to the environment, and offense increased, whereas new conditions tended to favor the pitcher. Obviously that doens't apply to something like moving the mound back to 60 feet, but the principle of experience tipping the scales towards the hitter is similar.

Robert R (Milwaukee): I know some research has been done on catcher's defense, but has BP ever thought about keeping track of the performance of batteries during the season? Things like hits, runs, earned runs, pick offs, SB, CS, etc. Obviously, there would be sample size issues, but then again people still print BAs this early in the season. I think the theory being that while catchers might not have an overall effect, some combos just might be more effective than others as they cover up each others weaknesses, ie. strong armed catcher with pitcher with weak more, or the opposite, or they enhance each other, catcher good at blocking balls in dirt with sinker ball pitcher.

Keith Woolner: I mentioned my catcher defense studies earlier. In one of them, I specifically compared two catchers who worked with the same pitcher over two years. No matter how good one combo was in the first season, there was no tendency for that to continue in the second season.

That focused purely on hits and walks allowed, though, not on SB/CS/PB/WP. And we know that some catchers have a pronounced effect on the running game, and probably have differing abilities to block pitches in the dirt. That *may* show up in ERA or run totals, though as yet I haven't seen convincing evidence of that.

Whether a given catcher is even better that his average talent level when working with a particular pitcher is even more debatable, IMO.

tcfwine (Philly): What's with my fantasy starters? Santana, Woody Williams, Pedro and Kip Wells! Gave up 15 1st inning runs!!!

Keith Woolner: And here I thought everyone knew the laws of fantasy baseball cause-and-effect.

It's precisely because they are on *your* team that they gave up so many runs. You've doomed them to seasons of mediocrity. Trade them now, or feel forever guilty.

(Could you also please send along the rest of your roster? I'd like to know who to avoid in my own fantasy league.)

Rennie (Downey): Keith, your article in Baseball Prospectus 2005 was very dense, but I really enjoyed it. My question is about the role of the statistical analysis person in a front office. Right now about half the clubs have hired someone like Keith Law, Voros McCracken, etc. How much are these advisors listened to, which ones are most successful? It seems like this is nothing new, and Mike Gimbel got canned for not having people skills. Are the people skills that important, what are they, and who has both the statistical knowledge and the people skills? Thank you for all the great material and for answering my question.

Keith Woolner: First off, glad you enjoyed the BP05 article on Win Expectancy. I feel like I should apologize to some degree, but it was exceptionally mathematical, even for me. :-) It started off as an article to explain some of the changes made to SNWL and the reliever reports, but I ended up covering so much about the basic concepts of win expectation that I changed the focus on the article, perhaps to the detriment of its readability.

As to the role of statistical advisors, the amount of influence they have varies. Some are virtually ignored, others have great influence. It depends on the attitude of the rest of the organization, and of the GM specifically.

IMO, people skills are critically important for a statistical consultant, and I think some organizations either don't realize that, or equate antisocial attitudes with mathematical aptitude. Ideally, the stats advisor is a operational consultant, and one who understands the business and real-life decision making processes beyond run scoring models. You've got to be persuasive, not just right.

jonlewallen (Washington, DC): Is data available regarding pitch counts for pitchers either in Cuba or the Japanese Leagues? I'm skeptical about Cuba, but Japan seems possible. Could pitch counts be a reason that older pitchers from these leagues seem to struggle when they get to the US - they simply have pitched too much? I'm thinking specifically of Cuba, where my impression is that they ride the one "ace" pitcher for all he's worth.

Keith Woolner: I agree with your guess that Japanese leagues are going to be a better bet for pitch counts than Cuban leagues.

That being said, I don't think the problem with Cuban pitchers has been overwork, as much as the level of competition is just not high enough. Clay Davenport has done some work on translating Cuban stats, and though the sample size is small, and my memory unreliable, I think he believes it's comparable to the NY-Penn league.

Cris E (St Paul, MN): In your NL picks you have the Giants finishing second in the West but Bonds not appearing in your top five MVP spots. Comments?

Keith Woolner: Jason Schmidt goes Old Hoss Radbourne in the 2nd half, and the 1-man rotation wins 50 games.

(more seriously, Bonds no matter how good will miss enough of the season to give a reason not to vote for him, and after four years, I think the voters are ready to give it to someone else for a change.)

jwb0581 (Dunbar, WV): Keith- As a big fan of our local small college team, I'm curious about the track record of Division II and III college players who advance through the Minors. Thoughts?

Keith Woolner: I don't know of any specific research looking at the track records of big baseball programs vs. smaller ones. There's probably something to the intuition that the top-name programs attract the best high-school talent, and then they get to compete against other top programs and thus are more polished coming out of college.

Intuition is often wrong, though. If anyone knows of work in this area, I'd be interested to see it.

Keith Woolner: That's all I've got time for today. Thanks for all the great questions, and come back for James Click's chat on Friday!