from Amazon
Tuesday, March 06, 2007
Hardball Times Preview Book
When one of your favorite websites posts a tip jar, it behooves all to drop a few bucks in there every year. It’s a gratitude for service. The gang at Hardball Times certainly has provided me with more favorable service than the local pub. But, Hardball Times is not content to put out a tip jar, but are instead providing you with a goodie bag:
http://www.hardballtimes.com/main/article/its-the-hardball-times-2007-season-preview/
I’d say, regardless of the quality, drop the ten spot, and get it. Don’t look at it as to whether “is it worth it?”.
But, what about non-regular readers? This blog entry is for you. Is it worth it? Beats me. But I’ll tell you the highlights and you can decide:
1. A quick intro to the forecasting system. There’s not much detail in there, but that’s the case with all non-Marcel forecasting systems. We’ll have to check in 6 months to see how they stack up. But, the guys who did the system are bright guys who “get it”, so it’s probably as good as anything out there. And, you get a spreadsheet with all the forecasts. It’s unclear how the distinction was made between starters and relievers, which to me is a huge issue. The forecasts also aren’t separated by starters/relievers. Maybe next year.
(Note to Studes: kinda dangerous to have the xls and pdf files share the same directory.)
2. You get a two or three pages for each team of “Team In A Box”. I’ve only read one so far, from my favorite blogger (Lisa Gray, or should I say: lisa gray), and the style appeals to me. It’s very Bill Jamesian, and very much for the internet age. If you know your writer, you’ll invest your time in whatever he says. But, those days are gone, and the “Team In a Box” format works for me.
3. The layout is very nice. Frankly, Studes should handle all baseball stat books that come out.
4. The cool part is that fielding is included, though I suspect the aging for fielding can use some more work. For example, Ichiro is rightfully at the top of best fielder in baseball for 2007 as a +17. But in 2009, he’s a 35yr old at +16. On the other hand, his OPS goes from .764 to .679 in 2 years. Jeter’s fielding improves each year (albeit from -25 to -23), and his hitting takes such a beating that he is a below-average hitter in 2 years (as a 35yr old in 2009). The fielding aging is very light, while the hitting aging seems very excessive. Overall, it probably works. So, while I can, prima facie, trust the 2007 forecast for hitters, I can’t for 2009. I’d have to do more work to evaluate those.
I actually prefer the WAR title, rather than WARP, since there is no such thing as WARP. You can’t do a wins above replacement, without considering position. Therefore, the position is a requirement if you are discussing replacement. Therefore, WAR should capture position, and it does.
5. If we select all nonpitchers with at least +.40 WAR and all pitchers with at least +.14 WAR, and you sum their PA and IP, you get pretty much the annual MLB totals for those categories. This gives us the pool of players from which to do further analysis. One of which is to sum their WAR, and we get 612 for nonpitchers and 483 for pitchers. That total is 1095 wins, which is +36.5 per team, meaning +.225 wins above replacement per team. That makes the replacement-level team .275. It’s a little low for my taste (I prefer .300), but certainly justifiable.
I operate with +583 for nonpitchers (+.12 x 162 x 30), and +437 for pitchers (+.09 x 162 x 30).
Of course, the THT results could simply have been the product of how I selected the pool to begin with. After all, I could have sorted based on PA instead, and selected those guys. If I did that, the total WAR would have been under 400, and that’s way too low.
Instead, and this is my preference, if I select all nonpitchers with a WAR of greater than zero, and a PA of at least 380, I get the right total of PA, and the sum of WAR is +558. Selecting pitchers with a WAR of greater than zero and at least 53 IP gives me a sum of WAR of +469. That total is now 1027 WAR, which is +34 WAR per team, implying a replacement level of .290. That sounds pretty good to me.