Szym, given Morgan's presence on the blogosphere, having a career projection up for him might make a little depressed if he came across it.

Dan: Bostock played the first three years of his career with Minnesota, not the Angels.

Dammit!

The projections are still correct, but I simply copied and pasted the "CAL" too quickly (I had to manually do it because for some reason with one of macros, "CAL" rather than "LAA" crashes Excel, I problem I have to figure out).

Also the years are wrong for Ensberg. 2003 was the big breakout year, not 2002.

I know this because my projection for Ensberg in 2002 was ground zero for a bit of projection existentialism I had. Thanks to a couple of huge years in 2000 and 2001 (in a terrible hitters park in 2001) the 2002 projection for Ensberg was fantastic and I got a few inquiries as to whether I thought Ensberg would be that good. He wasn't as he hit .242 with 3 homers in a 132 at bats. Of course the following year he went nuclear and posted even better numbers than I had projected the year before.

So I posed the question to myself: was the 2002 projection for Ensberg wrong, or was it correct in the spirit that it was intended (IE, Ensberg can hit a baseball)? Never really could settle on an answer...

OK, everything's fixed and re-uploaded.

As long as we're doing injured thirdbaseman, I'd love to see Eric Chavez after 2005 or 2006.

I only see one year of real Lyman Bostock (1975). Am I missing something?

According to your projection, Bostock, if he had lived, blew a chance at the Hall of Fame by taking the Angel's money. Also two rings.

Crap crap, I messed up the copy and paste *again*!

Sounds like you could use a better workflow.

OK, let's see if I avoided screwing up Bostock's projection for the 3rd time.

I need a better brain directing the workflow.

I've run a bunch of these, but this is the first time I've had this much of an issue. Kinda embarrassing, actually.

Dan, something else that would be useful for these projections would be a column listing the player's age. When I eyeball the ages of these guys agains the data, I'm wondering if your aging curve is the same for the different decades--seems as though Bostock declines more quickly from his peak than Ensberg.

Voros says, "So I posed the question to myself: was the 2002 projection for Ensberg wrong, or was it correct in the spirit that it was intended (IE, Ensberg can hit a baseball)? Never really could settle on an answer..."

Many years ago, when I was still doing sabermetrics books, I came up with something that I call the "iambic development theory." It's an expansion of the concept of Sophomore Slump. The idea is that a hitter has a good year; then the pitchers spend the offseason looking at film and making adjustments, so he has a down year. Then he makes adjustments and has a rebound year, and so on until reaching the peak at or near age 27. Just looking at Ensberg's minor league career and his few major league years before his peak, he fits the theory very well. The swings from good to weak years is pretty extreme, but the pattern fits all the way through to his peak year at age 29 (which is old for the pattern, but Ensberg got a late major league start).

I don't know if anyone factors this into projection methods, but it might help. It doesn't apply to every hitter, but it seems to apply to about 75% or so. Essentially, IMO, you didn't factor in that 2002 was due to be a weak year for Morgan, and it turned out that Morgan has very wide swings between his good and weak years.

The reason I never formally published this, BTW, is that I could never figure out a resolution to one odd dilemma. In iambic patterns, you have guys who have their good years at odd-numbered ages, like 23, 25, 27. Ensberg is one of those. Then there are the guys who have the good years at even-numbered ages. Of the latter - the "even-numbered iambics" - some have their peak year at age 26 and some at age 28 (and some few at other ages). I never could figure out a way to separate the 26-peakers from the 28s. Obviously, that makes a LOT of difference in a projection. So, I've never published. - Brock Hanke

Morgan's reaction, from his blog:

I have checked that out Steve and that is fun to speculate. You know what? I am really cool with the fact that I hit 110 hr’s in the big leagues with nothing but Norwegian genes. I am really thankful that I got to play as long as I did. Nothing is an accident, although it was really tough to go through the bad times, I am thankful that I had them because it gives me another reference point for players who are going through the same experience.

Morgan

Had Ensberg met that projection, he'd have been the 46th player with at least 300 HR and more than half of his career EBH being HR (51.0%). Of that group, Mark McGwire (duh) has the highest career percentage of his EBH being HR, 69.3%. The top 5 are McGwire, Harmon Killebrew (64.6%), Dave Kingman (62.5%), Cecil Fielder (60.7%), and Ralph Kiner (59.1%). Adam Dunn (57.0%) is 9th. Other active players above 50% are ARod (55.3%), Griffey Jr. (52.9%), Troy Glaus (51.6%), Jason Giambi (51.5%), Paul Konerko (50.78%), and Gary Sheffield (50.75%). Ryan Howard, with 222 career HR, stands at 60.2% and will certainly be in this group two years from now. If Carlos Pena (202 HR, 53.9%) plays three more years at his recent level of production, he stands to be in it too.

-- MWE

It's not so much that Ensberg is projected to have a more gentle decline than Bostock as much as Ensberg is projected starting at a later age, so there's less uncertainty for Ensberg at every matching age compared to Bostock.

To take an extreme example, look at Kevin Millwood and Tim Lincecum. Lincecum is a much better pitcher than Kevin Millwood, but Kevin Millwood's odds, as of today, of stepping onto a mound healthy at age 35 are higher than Lincecum's because Millwood's already survived the decade of attrition that Link hasn't.

Cool "what-if" projections.

It'd be neat if you kept doing these with guys like Lou Gehrig (although I guess it would be problematic as to when the disease really took hold), Roberto Clemente, Joe Jackson, Darryl Kile, Thurman Munson, Griffey without all those injuries, Bo Jackson without his football injuries, Tony C. and so on....

I did do Thurman Munson awhile ago and Conigliaro as an early experiment (I hadn't engineered ZiPS to do long-term career projections back then).

They're fun to do from time-to-time. On my 2011 build of ZiPS that I'm going to start working on soon, I'm making it more dynamic with the years so that I can do career ones more easily (right now, I have to copy and paste a historical player's stats to a separate spreadsheet to translate the numbers into 2006-2009 factors and after doing the projection, paste the numbers a second time on the second spreadsheet to move them back to the appropriate context).