A caution: part of the problem with trying to do this kind of comparison is that there are so few sample points every year. There are a ton of pitchers and hitters who move up and down between AAA and the majors, but in any given year there might be five players who go from NPB to MLB. One of the things that may be throwing off projection attempts so far is that the players moving over may not be a good, even representation of Japanese baseball talent in the same way that AAA has.
Also, this is me applying rough statistical tools that involve league difficulty, translations, and other related stuff. Jojima’s ability to adjust, his approach, and all kinds of non-stat things are going to affect his performance. This is intended for entertainment purposes only, so please… no wagering.
Updated note: in going through this exercise, my starting point was older estimations of the relative difficulty levels of Japanese baseball to the major leagues. While this is a good starting point, there are new ways to do this: see the comments for details.
So take Kazuo Matsui (please!).
2002: .332/.391/.617 (Seibu)
2003: .305/.364/.549 (Seibu)
2004: .272/.331/.396 (Mets)
2005: .255/.300/.352 (Mets)
That’s way under what he was projected to hit by most analysts who took a crack at it. (Dave, btw, said “The MÃ¢â‚¬â„¢s should jump for joy that they didnÃ¢â‚¬â„¢t get Kazuo Matsui. HeÃ¢â‚¬â„¢s got serious issues with his swing, and unless he makes some adjustments, is going to be a groundball machine. ThereÃ¢â‚¬â„¢s just no way he hits for any real power with his current hitting mechanics.” This is because Dave is awesome.) So in this exercise, I tried to error consistently on the low side, to try and avoid that kind of over-estimation.
Here’s Jojima’s last couple of years for the Fukuoka Daiei Hawks:
He was born in 1976, so aging’s not going to be a huge issue over the next couple of years. Jeff at Lookout Landing took a super-quick look at a couple of other players who’ve come over and said
Thus, Johjima’s .309/.381/.557 2005 batting line translates into a .270/.319/.413 performance in 2006.
Clay Davenport wrote a much longer, detailed look at a set of Japanese players in 2002 (here) that has some lessons as we attempt this.
Say, for a second, that there’s no adjustment in coming over. Averaging a player’s three-year performance is actually a great way to do projections (it gets us almost all the way there: PECOTA, the STATS, Inc. lines, and all the other systems are fighting over that last gap).
Raw 3-year-based projection if he stays in Japan: .326/.404/.602
That’d be nice if we could get it. That’s almost Pujols territory.
I’ll take some rough cuts at this. A straight Japanese baseball-to-MLB difficulty adjustment, based on Clay Davenport’s work on league difficulty, I come out with an estimate of .300/.380/.575 (pre-park-adjustments). That’s Mark Teixeira territory from a catcher. That’d be pretty awesome.
However, let’s also go for a much more pessimistic view and knock him down further. This is entirely reasonable, given a glance at recent Japanese players who’ve made that transition. I don’t want to be too doom-and-gloom, but skepticism is warranted.
I took the league difficulty numbers and then applied a blow-to-the-sternum adjustment, assuming that Jojima’s performance would take a significant hit. At that point, it became difficult to come up with a realistic-looking projection. Assuming that he retained the same offensive shape, it came out at .275/.355/.555.
That doesn’t look right, even for a low projection. If his average gets knocked down, we can assume that he’s not going to continue to get walks at the same rate, either. So toss that aside for a second, and let’s look for a more reasonable projection for an awful year. Kazuo Matsui only worse: his average gets chopped along with his walk rate, and almost none of his power comes over as well: .270/.310/.425.
Mariners catchers in 2005 hit .216/.253/.313. That’s not a joke. Upgrading from the Mariners to a horrible, worst-case Jojima is like the difference between 2005 Beltre and 2005 Sexson.
Before park adjustments, then —
Total collapse: .270/.310/.425
Crunching some more numbers and assuming that Safeco blunts his power a little, I came up with a lower line of .300/.340/.500. I still wonder if I should go back and see if that power line is too high. A .500 SLG in Safeco? Really? For $5m? Heck yeah.
Now, whether any of those are high or low remain to be seen, but I hope the exercise has been interesting. It’s interesting that applying a NPB league adjustment that’s harsher than what’s been used in the past still gives us a line so gaudy it’s immediately discarded. While it’s obviously eye-poppingly gaudy, it’s not a best-case scenario in representing a career year in the same way that the worst-case line there is built to show a collapse.
I’ll be revisiting this later, looking at possible comperable players and how they did.