THT Mailbagby Bryan Tsao
December 12, 2006
When we first announced the mailbag a couple weeks ago, I felt like a teenager waiting for his first date. What if she doesn't show up? What if she only writes questions that we can't answer? I began compulsively checking my email every few minutes for new mailbag questions and cursing Gassko for coming up with such a stupid idea in the first place.
Well, two mailbags in, I can safely say that our readers are likely the love children of Jessica Alba and Aaron Gleeman.
Seriously though, thank you all for taking the time to write in, and while we can't publish all the questions sent in, rest assured that all of your letters were read by our staff. So keep 'em coming!
Now, onto the questions.
I've got a question about the conversion of relievers to starters. The Boston Red Sox are converting Jonathan Papelbon to a starter after an incredible season as closer. The market values mediocre starters (Ted Lilly/Gil Meche/Vicente Padilla) signifcantly higher than excellent relievers (Justin Speier).
I am wondering what does it take for a reliever to potentially be a good starter.
The Blue Jays have an excellent set-up man in 23 year-old Brandon League. I am wondering if you think he would make a good starter. His repertoire consists almost exclusively of a hard (94-96 mph, but up to 100 mph) sinking fastball, although he also has a slider and a change which he never uses. He has an incredible groundball-to-flyball ratio of 5.71 and I feel he is a very similar pitcher to Brandon Webb and Chien-Ming Wang. He is very efficient and has low walk totals. His minor league stats are even remarkably similar to Brandon Webb's. He also seemed to get better the higher the pitch total as a reliever last season.
The Jays are desperate for starting pitching, would moving League to the starting rotation be wise, and what characteristics of a reliever could make a good starter?
- Daniel Y., Toronto
John Brattain: I can't answer what what it would take to make a reliever into a starter or what prevents it from happening: arm durability, too dumb to learn a third pitch etc., but as to the League question; League was a starter in the minors (70 starts), but the Jays use him a reliever at the major league level.
I've never heard an official explanation as to why.
My theory is that the Jays possibly overvalue the closer role and projected him as "closer of the future." Now that they've got B.J. Ryan locked up for four more years, I can't see a reason why they wouldn't move him into the rotation.
It might explain why general manager J.P. Ricciardi has been so comatose this year in shopping for starting pitching. League has a minor league ERA of 3.58, gave up fewer hits than innings pitched, was downright stingy in giving up gophers, walked less than three and struck out almost seven per nine innings.
I see no reason why he couldn't be a starter.
David Gassko: There has been a lot of research done on switching relievers to starters recently, including this piece by THT's Steve Treder, and this piece by yours truly. Steve took a fascinating look at the difference between relief and starting for pitchers who have done a lot of both, and also studied what traits comport with making a good transition from relief to starting.
What I found is that a slightly above-average starter is actually as valuable as an elite closer. There is also a great study in The Book that finds that the difference between starting and relieving is almost a full run. That is, a reliever should expect his ERA to go up by about a run when he goes from starting to relief.
Generally, it's thought that a pitcher needs three good pitches to succeed as a starter in the major leagues, but I think that a good sinker baller can definitely get by with less. League's main issue is a low strikeout rate, but he does have a decent walk rate, which does make him very similar to Wang. Of course, the difference is that in his whole career (major plus minor leagues), Wang has walked 2.14 batters per nine innings, while League has walked 2.81, 30% more. Nonetheless, if League can keep his groundball rate up and his walk rate down, he can be about an average starting pitcher, which obviously has a lot of value.
Hi, there are a lot of articles out there comparing the "too much" five-year, $70 million contract the Red Sox have offered Johnny Damon last season. I've heard more than a few sportswriters claim that Damon is a better player than Drew, and that the Red Sox should really regret letting Damon get away.
Looking at their 2006 and career numbers, however, I can't agree. If you compare their 2006 EqA, Drew comes in at an even .300, Damon at .281. As for career EqA: Drew .305, Damon .267. Looking at their 2006 WARP1, Drew gets a 7.3, Damon a 5.0. Looking at 2006 WARP2: Drew 8.6, Damon 6.7. OK, that's a bunch of numbers.
Still, overall Drew looks like the better offensive player. Some point to Damon's speed, but Drew has a better arm and can steal fairly well, too. Drew will also be 31 in 2007, while Damon will be 33. The one area where Damon really seems to trump Drew is durability. I won't argue with that. Obviously, we can't know how either player will fare in next few years, but I'm curious who you guys would take.
- Mike, Maine
Dave Studeman: I would rather have Drew on my team than Damon. As you point out, all of Drew's basic batting sabermetric stats were better than Damon's last year. For instance, Drew's GPA was .309 in 2006; Damon's was .290. Damon's year was also flukier than Drew's. He hit a career-high 24 home runs and had the second-highest slugging percentage of his career. I wouldn't look for him to repeat those numbers next year. Without the 24 home runs, his numbers will be even less valuable than Drew's.
The other factors favor Damon; he's speedier and sturdier (and plays a more critical position), while Drew is younger and has a better arm. Still, I would pick Drew.
David Gassko: I agree. Even given their differences in playing time, Damon is about three wins better than replacement going forward, while Drew is about a win-and-a-half above that. It's also much more likely that Damon under-performs his expected playing time than Drew, since I'm already expecting Drew to miss a large chunk of playing time, while Damon is projected to stay healthy. More can go wrong with his projection.
Adjusting for aging, Damon is expected to be something like eight wins above replacement over the remainder of his contract (after being at around 2.5 wins above replacement last year), while Drew is expected to be around 20 wins above replacement.
Moreover, we have to remember that Drew is being paid for later years than Damon, which means the present value of his contract is going to be discounted more. In 2007 dollars, Damon's deal was for four years, $49.86 million (he got 4 years, $52 million, not $40 million as you wrote), while Drew's contract is five years, $58.38 million. So Damon is being paid something like $4.75 million per marginal win, while Drew is being paid around $2.92 million per marginal win, or about half of what Damon is getting.
So the J.D. Drew deal was about twice as good—some pretty nice maneuvering by Theo Epstein.
Thanks for the analysis of baserunning by Bill James on pages 291-295 of the new Bill James Handbook. I would love to see you do a similar analysis for the throwing arms of outfielders. Being a despondent Dodger fan, it would be interesting to compare Juan Pierre's base running rating to a comparably scaled rating of his "success" in keeping baserunners from taking the extra base. Likewise for other outfielders with weak arms who are "valued" for their speed (or "slow" outfielders who may have strong arms). It would be nice to eventually be able to bring this all together for outfielders (baserunning, range, assists, holding baserunners).
- Mark, Bakersfield, California
Dave Studeman: We didn't generate those baserunning stats in the Bill James Handbook. They were calculated by the good folks at Baseball Info Solutions.
However, you can find John Walsh's outfield arms rankings for 2005 in this article, in which Pierre is assigned a -3 runs. Also, in this year's THT Annual, John ranks all outfield arms from 1959 through 2005 and calculates that Pierre has the second-worst arm of regular major league center fielders during that time, trailing only Bernie Williams. He averaged -5.4 runs per 162 games.
Thanks to your stats glossary, I finally understand what goes into a players batting average on balls in play. However, I would like to know how, if its possible, to convert a players batting average into his expected average based on his BABIP. For example, if a player bats .300, but has a .400 BABIP, how could I calculate what his batting average should have been?
- Dan, New York
Dave Studeman: We're actually pretty proud of our glossary, and I'm glad you found what you needed for BABIP.
There really is no straightforward relationship between BABIP and batting average. The difference between the two depends on a specific batter's home run and strikeout rates. Quite often, a batter's average will be lower than BABIP, but there are plenty of exceptions. For instance, Albert Pujols' average was .331, but his BABIP was .294.
To put it mathematically, if a player's ratio of home runs to strikeouts plus home runs (HR/(K+HR)) is higher than his BABIP, than his overall batting average will be higher. And vice versa. I hope that helps.
I'm a 6th grade teacher who is currently setting up a "Baseball Statistics" club. I'm interested in teaching and showing students (ages 11-13) the depth of baseball statistics. (Your site will of course be a great reference for us.) How would you go about introducing kids to stats? I'd like to get these kids away from that ESPN/Joe Morgan way of thinking about the game and look at what really matters. If you were introducing someone new to your way of thinking about baseball, what do you think are the most important things they need to know?
- Andrew, Dedham, Massachusetts
Sal Baxamusa: This sounds like an interesting idea. The main thing that I would keep in mind is that statistics are not a new idea in baseball (I believe Bill James and many others have said something to this effect before). Ever since Henry Chadwick, people have been creating a numerical record of baseball. The most fundamental ideas to "our way of thinking" is separating the statistics that tell us something important from the statistics that don't.
For example, most people know about the Triple Crown stats: batting average, home runs, RBIs. I would classify them somewhat telling, very telling, and not very telling. Similarly, the big pitching stats—ERA, wins, strikeouts—are somewhat telling, not very telling, and very telling. Without delving too rigorously into concepts like the Three True Outcomes or Defense Independent Pitching, I think you could very easily get your kids debating whether and why a particular statistic tells you something about the player. That's the first step to understanding why we've seen different statistics come to prominence in the last ten years.
The more I think about it, the more I think that this is fantastic idea for young students. Learning to identify important bits of information and discarding the red herrings is central to good critical thinking. Additionally, discovering sabermetrics for the first time will give your students the opportunity to unlearn a lot of stuff. That's hard to do for adults; my guess—and you of all people probably already know this—is that your students will be more open to recalibrating their thinking. It also sounds like a heck of a lot more fun than the recycling club.
Bryan Tsao is the editor of The Hardball Times website. He welcomes comments, questions, and suggestions for both himself and the site via email.
<< Return to Article