|
Keith Woolner is an author of Baseball Prospectus.
Keith Woolner: Hi everyone, let's get chatting!
paul m (seattle): Is Quality of Batters Faced adjusted for league, park, all-time, ad nauseum?
Keith Woolner: Currently, the QBF report reports raw totals, not adjusted for league, park, or anything else.
John (Quincy, Mass): What's Pedro's stat line in 2007 and who is he playing for?
Keith Woolner: 5-0, a record 61 SV, 1.85 ERA, 90 IP, 110 SO, closing for the Yankees.
KK (The Internet): Your RBI opportunities report is another tool that helps to understand which players get the most chances. Should you take account of IBB's, for example, that takes away a player's chance to get RBI's? Will you make this report available from previous years? Again, thanks for you contributions for better baseball analysis.
Keith Woolner: Lots of questions about the RBI report...
First, previous years (since 1972) should already be there.
There are a number of ways the RBI report could be fleshed out, and once we get past the book-writing season, I may have a chance to go back and improve it.
Eliminating IBB, and perhaps walks in general, is one way to do it. Considering the specific mix of runners on 1B, 2B, and 3B rather than the aggregate number, in determining an expected (or league average) number of RBI's in those situations is another.
Aaron (West Hills, CA): Randy Johnson for Edwin Jackson, Choi, and a mid-tier prospect. Doesn't everyone win here?
Keith Woolner: Can the Dodgers really afford to give up an offensive player for an expensive, albeit top notch, starting pitcher, if they lose Beltre, and may trade Green? Johnson's making, what, $16 million next year? With so many free agent pitchers on the market, wouldn't it make more sense to go after a Pedro or Radke, and keep Jackson and Choi for the longer term?
Mike (Exont, PA): You've done a lot of research into catcher's defense and I think you also wrote an article about catcher's aging patterns. What tools would you use to determine Jason Varitek's value over the next 3-5 years? What's a ballpark reasonable contract?
Keith Woolner: I've written 3 or 4 articles looking at catcher game calling, without finding a significant effect. Most of the objective evidence currently available points to preventing the running game as being the primary defensive contribution of the catcher, and that's an area Varitek has not excelled in.
It's been awhile since I looked at catcher aging patterns, but if I recall, they tend to peak later, and decline sooner (the worst of both worlds).
There's at least anecdotal evidence for concerns about a large catcher aging poorly as well.
I'm not terribly concerned about Varitek being an offensive asset over the next couple of years, but as he gets further into his mid to late 30's he's a potential risk. 2/21 with a couple of club options might be interesting. I don't really have a sense for how the market is going to weight his contributions at the moment.
MM (()): I was wondering if you could explain how two of your sets of statistics
are calculated:
1) Expected Run Matrix
2) Expected Wins Matrix
In particular, I would like to know whether these were calculated
theoretically or empirically.
Keith Woolner: For both the Wins and Runs matrices, the values shown in the report are measured from play by play data, compiled for every game over the course of the season, looking at each game situation, and how many runs scored for the rest of the inning (or whether the team won the game).
A more theoretical calculation is done in the new SNWL and Reliever reports, where I estimate how likely a team of a given offensive strength is to score a certain number of runs over the rest of the game. A formula based on a model of run scoring is more appropriate here since it allows for tweaking the team's overall strength based on the actual opposing hitters the pitcher faces, which is important for short relievers.
carbolicsmoke (Toronto, Canada): Sorry if this is something of a stupid, newbie question, but have their been any studies done about the relationship between VORP and winning? Does the team with the greatest VORP usually win the most games, similar to the Pythagenport thing?
Keith Woolner: Not a stupid question at all...
I don't think we've published a study comparing VORP to team winning percentage, although I've looked at it informally myself. VORP uses an underlying model of run scoring (currently Runs Created), which has already been shown to relate to offense, and thus to winning. VORP is more of a way to figure out where the extra runs are coming from, relative to a replacement level baseline.
There are a couple of ways to compare VORP to winning percentage. One would be to simply add up the team VORP (both hitting and pitching) and run the analysis. Another, slighly more sophisticated way, would be to to add hitting VORP to the runs expected from a replacement level lineup, subtract the pitching VORP to the runs allowed expected from replacement level pitching staff, and use something like the Pythagenport formula to estimate team winning percentage.
For what it's worth, the work that I've done shows a very high correlation between team VORP (either method) and team winning percentage (above 0.9).
MPK (Franklin and Marshall College): What can the Astros do about replacing Lance Berkman for the 1st couple months of next season? Hope Roger Clemens sticks around for year?
Keith Woolner: Hope Jason Lane can hit?
cirwin (San Carlos, CA): Do you have any knowledge or opinions about the metrics MLB uses to determine the class of a free agent?
Thanks,
Chris Irwin
Keith Woolner: I don't have a link handy, but the system was negotiated between MLB and the player's union. For each position, players are ranked in various statistical categories (HR, AVG, RBI, Wins, Saves, etc.) compiled over the past 2-3 years. I believe their rankings in each category are added together to form a score. The top 20% scores are designated Class A, and so on down the line.
If you're implicitly asking whether the categories have any bearing on their actual (sabermetric) value, the answer is that "any similarities are purely coincidental."
GBS (See more, IN): Chocolate ice cream or vanilla?
Keith Woolner: Vanilla, but skip the ice cream, and just drink the extract straight. Woo!
ncbncbncb (Vancouver, BC): how about also providing a composite Expected Win Matrix table containing all the data for the entire period 1972-2004? A single season doesn't provide enough data points for most of the fields in the table.
Keith Woolner: That's another common request -- as you note, the single season tables have situations that occur rarely, such that the sample size can't really be trusted to give an accurate estimate of the team's chance of winning. I'm looking at adding aggregated-across-all-years reporting to the stat reports, but no promises as to when it will get done.
Dave (Fresno): At the end of their respective careers, who do you think will be the better shortstop: Tejada or Crosby. Crosby will likely never reach Tejada's career RBI numbers because Tejada refuses to draw a walk with runners in scoring position and thus greatly expands his RBI opportunities. However, it seems to me that in most other categories Crosby figures to be as good or better, particularly in OBP. What would be your prediction?
Keith Woolner: Bobby Crosby, age 24: .233/.319/.426 as a rookie
Miguel Tejada, age 24: .275/.349/.479, in his 4th full major league season.
I like Crosby, but he's got a ways to go to join Tejada's company, even with an OBP edge.
Andy (Raleigh): Your wrote an article in 2001 about BABIP. In it, you suggested that perhaps successful veteran pitchers have learned how to get maximum benefit from their defense. What might a pitcher do to get this advantage, and have you done any further research on this idea?
Keith Woolner: I think you're referring to this article:
http://www.baseballprospectus.com/article.php?articleid=883
I think what I wrote was that one reason that I might have gotten the BABIP results that I got, was that by looking at pitchers with substantial careers, I may have biased the data unwittingly. One way that could have happened would be if pitchers who don't learn how to pitch with their defenses wash out sooner, before reaching the playing time thresholds I was using.
I've done some work since then, and there do seem to be other ways that pitchers affect batted balls, more strongly that BABIP by itself would indicate, primarily in ground-ball vs. fly-ball tendencies, and the propensity to give up extra base hits. Another question I get asked from time to time, but have not studied, is whether catchers affect a pitcher's BABIP at all. I think Chris Dial has analyzed the reverse of this (does a catcher affect a pitcher's defense-independent outcomes), and basically confirmed my original research, but I'm not aware of anyone specifically looking at a catcher effect on BABIP.
kgknapp (Palo Alto, CA): Keith,
Any chance that BP would develop a fantasy product to compete with diamond mind baseball, whatifsports.com, scoresheet, purebaseball? They all have problems, and I think there's a market there for something better.
Keith Woolner: While BP as a whole hasn't undertaken a fantasy product along those lines, I've done some idle brainstorming of my own on the topic. We're always interested in hearing what kinds of products would be of interest to potential BP customers, so if you have thoughts on what you'd like to see (on a fantasy game or otherwise), please email us.
billy wegman (madison, wisconsin): Should the Brewers trade Lyle Overbay? Prince Fielder is going to be arriving soon...
Keith Woolner: Trading from where you have depth to shore up weaknesses is a sound strategy. However, Fielder's still pretty young, and there's no need to rush him and start his service clock ticking. And it's unlikely that the Brewers are on the verge of competing immediately. If they get a sweet offer for Overbay, there's no reason not to take it, but I wouldn't move him prematurely.
BJ Upton ((Tampa Bay)): So, do I have a position to play next year with the big club? Or is Lou going to be bouncing me around like a ping pong ball for the rest of my career? Heck, maybe it's true -- my younger brother is a better player.
Keith Woolner: Who can say which way Lou will blow, but blow he will. I'd keep you at shortstop until you conclusively show you can't handle it.
Besides, you won't spend your whole career at Pinella's mercy. You'll be a free agent at age 25 at this rate, so you've got plenty of time post-Tampa if you so choose.
DavidCrowe (Canada): Where does pitcher abuse stand now - beyond the comments that Bill James made in his recent book what do you make of Roy Halliday's tired arm? Does the human element overwhelm any statistical trends?
Keith Woolner: Bill was kind enough to allow Rany and I to write a rebuttal in his book, and I think we pointed out some legitimate concerns about the work he did. But regardless, right now it looks like the pitcher abuse battle has largely been won -- at least at the major league level. Abusively high pitch counts are down significantly, and awareness of pitcher workloads is improved.
What remains to be examined, and potentially championed again, is to get a better handle on minor league and college workloads. There's really no difference in the physical load borne by a major league pitch versus a minor league pitch -- and ignoring (for lack of data) minor league pitch counts in considering a pitcher's lifetime injury risk misses a lot.
Furthermore, pitcher usage as a whole (rather than just pitch counts) need closer examination. Rany's done some great work advocating a return to 4-man rotations, which would offset some of the innings lost to stricter pitch counts, and increase starting pitcher value without impacting their health.
Reliever workloads also need some serious research. We're probably 10 years behind understanding bullpen usage patterns compared to starting pitcher usage patterns.
Keith Woolner: Thanks for the questions everyone! And congrats to Bobby Crosby and Jason Bay on their RoY awards.
|
