“Statistical Oracle” Sees Big 2010 for Holliday, Carpenter
TOWER GROVE — Of all the statistical bubblegum that is packed into Bill James’ annual Handbooks, among the best to chew over and stretch to the point of popping is the predictions for the coming season.
ACTA Sports recently emailed me the 2010 predictions for several members of the 2009 St. Louis Cardinals — predictions that are featured in Bill James Handbook 2010 — and while it doesn’t take a “statistical oracle” (as The Wall Street Journal called James) to guesstimate what Albert Pujols will hit for the 10th year running, there are some interesting estimates for key Cardinals. They include:
- A virtual repeat for Chris Carpenter.
- A drop for Adam Wainwright that isn’t as dramatic as some will argue next spring.
- A modest second-year improvement for Colby Rasmus.
- A significant dip for closer Ryan Franklin.
- And a strong turn from free-agent outfielder Matt Holliday.
Each year, James, who has rewritten the book on using and projecting baseball statistics (several times, really), offers up “Hitter Projections” in tight, small-print grids that appear near the back of his handbooks. Last year’s essay/lead-in to the projections starts by pointing out that the 2008 Handbook predicted 33 players would drive in 100 runs or more. Twenty-nine did in the 2008 season, and 19 of them were the same players predicted by James. One that wasn’t: Cardinals outfielder Ryan Ludwick.
James had him pegged for 40 RBIs in 2008 — which is more than you did — and Ludwick finished with 113.
Some predictions from 2009’s Handbook are uncanny:
Predicted Pujols: .337 BA, .443 OBP, .633 SLG, 41 HR, 118 r, 125 RBI
Actual Pujols: .327 BA, .443 OBP, .658 SLG, 47 HR, 124 r, 135 RBI
**
Predicted Rasmus: .246 BA, .331 OBP, .422 SLG, 18 HR, 75 r, 54 RBI
Actual Rasmus: .251 BA, .307 OBP, .407 SLG, 16 HR, 72 r, 52 RBI
Others were off by a little bit (Ludwick’s almost drove in the 100 runs predicted) or by a few galaxies (i.e., Rick Ankiel did not come near the 31 home runs expected). But, overall, there were kernels of insight in every line predicted. Ankiel did, for example, strike out at nearly the rate predicted over a 500-AB season, and Skip Schumaker drove in 33 runs when he was predicted to drive in 34. The formulas that James uses to come up with the projections for hitters are based on the player’s individual production in their career, but also the career-arc of similar players with similar production throughout baseball’s history. It creates, essentially, a rate by which all players past production is funneled.
Pitching, as James acknowledges in the press release teaser, is more difficult to project.
“We used to believe that pitching performance was much, much less predictable than batter performance,” James says, per the email. “This is probably still true … due to injuries and other factors. Sometimes a pitcher gets hurt, and when that happens our projections for him are knocked into a cocked hat.”
Keep all of that in mind when looking at — finally — the projections for some Cardinals, from the Handbook:
Key Cardinals Hitters (by OPS), all around 500-600 PAs
Albert Pujols: 121 runs, 44 HR, 129 RBIs, 11 SB … .333 BA, 1.085 OPSMatt Holliday: 105 runs, 27 HR, 109 RBIs, 14 SB … .316 BA, .922 OPS
Ryan Ludwick: 77 runs, 26 HR, 93 RBIs, 3 SB … .270 BA, .826 OPS
Troy Glaus: 79 runs, 28 HR, 91 RBIs, 2 SB … .248 BA, .817 OPS
Colby Rasmus: 75 runs, 18 HR, 55 RBIs, 8 SB … .253 BA, .754 OPS
And, then the pitchers, with a couple market-corrections expected for two Cardinals:
Key Cardinals Pitchers (by ERA)
Chris Carpenter: 195 IP, 16-6, 154 Ks … 2.95 ERA
Adam Wainwright: 225 IP, 15-10, 178 Ks … 3.64 ERA
Ryan Franklin: 61 IP, 4-3, 33 Ks, 17 Saves … 4.13 ERA
I predict some teeth-gnashing with that last set of numbers. Discuss.
-30-
Derrick Goold said he was going to Mizzou for capital-J journalism, but after growing up in the Time Zone Baseball Forgot he was really drawn to MU sitting between two major-league cities. Goold joined the Post-Dispatch in 2001 after working for The Times-Picayune and Rocky Mountain News, covering sports from LSU to NHL and every level of baseball in between. 
I am surprised by the bullish projection for Glaus. He would be a good buy-low candidate if those numbers prove true.
I think Rasmus will beat that projection but even if he doesn’t I would be happy for the modest offensive improvement at his age. With his fantastic range he will still be a very valuable player.
I think the 3 pitching lines you posted are pretty accurate estimates. With Carp it is all about health. IF he stays healthy those numbers are probably a little bearish but certainly acceptable. I expect Wainwright to beat those numbers…but I do expect some regression from him.
As for Franklin…he had a great first 4 months and then regressed to what he is. He is a productive bullpen piece who fades in the second half of seasons. He is really not closer material but until one of Motte, Boggs or Sanchez steps up and takes the title away from him, Franklin will likely remain in the closer role. If the Cards do not re-sign Holliday or waste their money on Bay, then maybe they will spend it on Billy Wagner.
Put me on the pile that has expected Glaus to bounce back. The dude (when healthy) can hit. That furthers my desire to bring him back at a one year 5-mill contract that will help both parties. No way he wants (or gets) a long term deal after this season. Unfortunately, I hold my breath everytime Carp walks off the mound during an inning so we all definitely take his prediction with a grain of salt. His injury is so rare and unknown, we could only hope he stays healthy for the duration of his contract.
I’m exactly afraid of a season like that from Franklin. Closer is a HUGE issue for this team right now and the front office doesn’t want to acknowledge it. They’re happy with the pocket change they are paying Franklin and are saving on not having a front-line closer. Motte isn’t much of an option there either, in my opinion. Hawksworth might be the best, in my opinion, but I think he deserves to get a chance to start.
Adam 220 IP, 16-8 185 K’s 2.97 era
Carp same, but era around 2.65
Franklin, I will agree with the innings and saves as I think Motte will assume the role sometime after the break, but I’d say an era mid 3’s.
Any thoughts on what it would cost to sign Glaus?
I’d say 90 runs, 21 HR’s, 71 RBI’s with a .277 avg for Colby.
All this is based upon a very precise method of calculating future performance. I call it gut feeling.
Wow I guess they play the games just to see if he is right. Could save billions and just not play and go with the predictions. Sorry just not a convert of using math and stats to value a player.
Kinda funny James didn’t predict Jeter taking a time out to tell C.C. a joke when he got in a tough spot to settle him down or the lack there of any Cardinal walking to the mound and settling Franklin down after Holliday dropped the ball.
Bah,.. nevermind I love stats and Tony’s endless logic of matchups and, and,…the Mayan calender ends in 2012.
Derrick,
Thanks for sharing. Any idea how much stock (if any) the FO takes in this data when making decisions?
Well, does James necessarily say that Franklin will be the closer for all 2010? I believe RF was extended for two more years? That ought to be plenty of time to condition and transition one of several other candidates for the closer role and to have Franklin take over the ‘Springer’ role of getting one or two key guys out. Get a double play. I don’t think the Cardinals believed they had the next Sutter when they resigned him. They saw a solid reliever.
And that is what one ought to take away from those stats if you believe them, which I don’t. I think stats guys say that a closer is no different than any other reliever and all they do is count mph and swinging misses. I am one of those people that believe that there are a number of decisions made by managers that start with the 9th inning and work backwards - so the 9th is, by definition, a statistical outlier.
james k: “Sorry just not a convert of using math and stats to value a player.”
I agree that some of the stats people can be pretty arrogant, and that no predictions are foolproof, but honestly, what do you suggest is a better way to value players? Really, unless you can watch every single game played at every single level (from low minors up through the majors) and remember every pitch and every play with photographic detail — AND predict the future based on these extraordinary powers — what alternative is there?
jdog,
The Cardinals have their own analytical staff and metrics that they use. They’ve cooked up some of their own mathematical approaches to evaluate and value players.
dg
-30-
I know stats are probably the best method for alot of it. I just try to remember it was a game once and now it is a business. Its the romantic in me that recalls the days of a crusty ol scout saying “kid I can’t promise you everything more than a bad bus ride and the roar of the crowd if your willing to give it all you got”.