Jim Albright / the japanese insider
Japanese Possibilities for the MLB for 2003
The major league equivalents used in this article were arrived at by a comparison of matched at bats for all position players who played in Japan between 1992 and 2001 and also played in the majors at some point. This yielded 41,290 matched at bats, the equivalent of full seasons for approximately 8 teams, one set being in the major league parks facing major league pitchers, and the other in Japanese parks facing Japanese baseball pitching. If you want more details on the method of matched at bats, please see my article estimating Ohs career home runs, which explains this process in detail.
One thing I did was to make the matches come from the nearest season(s) in time, but in the case of ties, I chose the later season in the case of the major leagues, and the earlier one in the case of Japanese baseball. I did this because the predominant scenario is a player played in the majors, then went to Japan. Once he went to Japan, his career either continued in Japan, or ended without him returning to the majors. This means that the matched seasons for Japan are generally of the same player, but a little older. The rule I used on ties for seasons equally close in time is designed to counteract this effect rather than emphasize it.
The factors derived for the hitters were as follows:
The above statistics were multiplied by the appropriate factor. Then those adjusted stats as well as games and at bats were multiplied by a factor of 162 divided by team games in order to bring the player up to a 162 game base. I left out runs scored and RBI essentially because when players move between the leagues, they often change batting slots, which can have a profound effect on these stats. I left out stolen bases because most of the position players who have played in both places are slow sluggers, which limits the value of any comparison of that data between the two leagues.
The pitching adjustments start with the inverse of the batting adjustments. This is a
correct approach because in the case of batters, we were comparing their output in
the two separate sets of conditions. When we use the inverse, we are simply comparing how
Japanese pitchers in Japanese parks did against these players to how major league pitchers
did against the same players in major league parks. The number of opportunities
are the same. The approach yields the following results:
The factor for runs allowed and earned runs allowed is 1.242, calculated by averaging
the following four factors:
Were not yet done with the pitchers, though. Except for 2002, I have actual batters faced totals in Japan. I multiplied this figure by the 162 divided by team games figure to bring the pitchers up to a 162 game schedule. If I had done equivalents for earlier Japanese pitchers, I doubt Id make that seasonal adjustment due to the extreme loads they were already pitching. However, in the past 15 to 20 years, pitching workloads in Japan have moderated to the point where such an adjustment seems appropriate. For 2002, I estimated the batters faced by mulitplying the innings pitched by 2.836 (the 1992-2001 average in Japan for the calculation of (AB H) divided by IP) and adding walks, hit batsmen, and hits to that figure. Once we have the adjusted batters faced figure, we deduct the projected hits and walks (I didnt project hit batsmen) and then divided the result by 2.821 (which is the major league average for approximately the same period of (AB-H) divided by IP) to yield the projected innings pitched.
I then adjusted saves by the projected innings pitched divided by the actual innings pitched. I also had a factor to reduce the number of saves if a reliever (defined as pitchers with less than 3 IP per game pitched) had an ERA over 3. This last factor rarely had a significant impact.
All that remains are wins and losses. We start by adding wins and losses together and multiplying that total by the projected innings pitched divided by the actual innings pitched to get the adjusted decisions figure. We then calculate a projected winning percentage by the formula
21.64 divided by (21.64 plus ((projected ERA times 1.099)squared))
This is an adaptation of Bill James Pythagorean theorem, where 1.099 is the major league average ratio for the period of runs to earned runs and 21.64 is the square of the average number of runs scored. The projected winning percentage is multiplied by the adjusted decisions to get the projected wins, and losses are calculated by (1 projected winning percentage) times adjusted decisions.
One big caveat I must make is that I havent made park adjustments to this data, which could seriously skew individual results. The reasons I did not do so are 1) with respect to Japan, the data isnt available to me, and 2) with respect to the majors we dont even know in most cases if the player will even come over, much less which team he will sign with.
B. The Batters
Ive limited my conversions for batters to the period from 1996 and later so we dont have to worry about the effects of the 1994-1995 major league strike. First, Ill give three batters who have already come over to the majors so you can get a feel for how well or poorly the system works. Remember, the system deals with what players have done, not what they will do. However, batting records have significant predictive value.
First, well look at Ichiro:
I'd say the method accurately sees him as hitting for a very high average, but medium to low power
The method realistically assesses him as a rather weak hitter for an outfielder.
The method sees that Taguchi is too weak a hitter to get a starting outfield job in the majors. The Cardinals needed spring training to make that determination.
With the help of Michael Westbay of japanesebaseball.com, I compiled a list of the leading free agents in Japan along with those players reasonable likely to be posted in late 2002. Michael also provided many of the stats which are the basis for these conversions. I included the player's birthdate and his 2002 salary based on a conversion rate of 1 million yen equals $83,000.
This guy would likely come cheap, but a 33 year old outfielder with no power and poor on base percentages has very little value.
Has moderate power, and generally good on base percentages. However, you've got to be concerned about a 34 year old hitter who had the kind of dropoff he had in 2002.
As I understand it, he served prison time in 1998 for tax evasion, which made him rusty in 1999. Hits for a decent average and on base percentages with moderate power. At age 31, he doesn't offer a long term solution to a team's thired base problems, but at $2-3 million a year, he's probably worth a shot if you need a short term solution to a problem at the hot corner.
It's now official that he's coming to the majors, and you've got to like a .300 average, .400 on base percentage, and over .500 slugging. It would probably be best to have him bat third in the majors, to keep him focused on using his broad range of skills rather than see him swing for the fences and possibly mess himself up in the process.
May be posted in 2002, a free agent in 2003, or stay in Japan until the end of his career. Nobody, not even apparently Kazuo himself knows as I write this. Still, he's likely one of the best five shortstops in the world of baseball at the present time, along with A-Rod, Nomar, Jeter, and Tejada. The 1997 and 2000 triples estimates are probably too high due to the size of the adjustment factor on triples. However, he will likely push double digit triples totals in almost any major league park.
He's a better player than Kokubo, but far more expensive. I'd be reluctant to put up $4.5 million or more a year he likely can command in Japan. He's a player for whom economics, not talent, is the reason he will likely stay in Japan.
Not good enough to start at first base in the majors
The method sees him as having been able to be a starting major league outfielder from 1997 to 2000. It also sees he has experienced significant and consistent decline over the past 4 seasons. If I were a major league GM, I'd pass on any 30 year old who 1) was not of major league starting quality in his last season and 2) shows such a pattern of decline.
The only season he even resembled a regular platoon outfielder was 2000. He's now 31, and in 2002, he didn't play well enough to merit a major league job. Needless to say, if I were a major league GM, I'd have no interest in him.
Pitchers records have far less predictive value than hitters records do, in large measure because pitchers so often hurt their arms. I should also note that the translation method never lets a pitcher pile up wins on good run support or losses on poor run support. Therefore, won-loss records are somewhat unrealistically bunched.
I also ran the records for two pitchers back to the 1992 start of the data used to determine the conversion factors. Those pitchers are Nomo and Irabu, both of whom would really give no proper basis for comparison if I held to starting the conversions in 1996. Of the four most prominent Japanese pitching imports to the majors, only Sasaki clearly matches the conversion of his Japanese accomplishments. Nomo was worked hard in 1990 and 1991, and it was already starting to show in 1992. He had a dramatically reduced workload in 1994, which seems to have restored his arm in time for his major league rookie season of 1995. Irabu probably had the talent the conversion suggests, but undermined it with poor training habits. It is also possible he pitched in a very favorable park when he was in Japan, which would help his stats. The conversion for Ishii does capture the ups and downs he experienced in the majors in 2002.
Now for the pitchers who are free agents or who are reasonably likely to be posted in 2002:
He might come over for fourth or fifth starting pitcher money, and could well be worth that much. However, his 2002 drop in innings pitched would concern me and makes me think such a move might be a gamble. Certainly, you'd want to check out why the drop occurred and carefully investigate how healthy his arm is.
He has flowered in a relief role and could be quite valuable as a setup man and possibly even as a closer.
His record doesn't suggest he'd be successful in any pitching role in the majors except for batting practice pitcher.
His 1999 and 2002 seasons look decent at first glance, but none of the other seasons do. If you look closer, he's always giving up lots of HR and over a hit an inning. That's not the kind of qualifications you want. Since he's 33, I wouldn't get excited about his major league potential.
His record suggests he could find a middle reliever role in a bullpen, but not much more than that.
He did have a poor 2001 season, but other than that, he would seem to be good enough to be a solid setup man in the majors and maybe even a closer.
Hurt his leg in 1996, and went through several years where he wasn't of major league quality before putting it back together in 2002. Since he's already 34, if I were a major league GM, I'd only be interested if 1) he would take an incentive laden contract with little guaranteed money, and/or 2) he would be willing to become a setup man (or a closer if he pitched well enough). I don't know if he could or would make the switch effectively, and I'd certainly want my scouts' opinions on the feasibility of such a move. The reason I find this possibility intriguing is Kuwata was a fabulous pitcher in 1986-1990 and in 1994 in Japan. This means he has a real upside, but only if he stays healthy. The setup or closer role could be used to provide a greater probability of keeping him healthy and therefore improve the likelihood of tapping his upside potential. However, I suspect economics strongly favor his staying in Japan.