Home Page

Baseball Analysis Home   Jim Albright / the japanese insider

Was Oh Better Than the Projection?

by Jim Albright

I received some correspondence from a gentleman named Don Lemmon who took issue with my projection of Sadaharu Oh's career. Two things made his correspondence unique. First, he was taking me to task not because I had evaluated Oh too highly but because I had underestimated Oh's talent. He's the first person to treat me to that particular criticism, which certainly made him unique. Secondly, he offered to do his own projection, which, naturally enough, he would regard as more accurate, so long as I agreed to see it posted. I invited him to do his projection, and I promised to post it, under the condition I reserved the right to point out areas where I thought his projection was flawed. He then sent me the following effort. I composed my response, and sent this document to him in the interest of fairness. After a little back and forth, the following article came about.

We'll start with almost all of Mr. Lemmon's projection, which will be followed by my response. My response will include a projection which I believe to be slightly more accurate than the one appearing in my articles on Oh. The projection used for my earlier articles can be found at full length here. For reasons I will discuss later, I will not use this revised projection in the earlier article. However, I do believe that this latest effort is the best projection I can do at least at the present time.

Mr. Lemmon wrote: I don't want to get too technical here, as we all know without a doubt, being exact is absolutely impossible. It's all speculation as to how well Oh would have done in a career against Major Leaguers or how different the results could have been if I spent more than the 2 hours I did doing this by being tedious. I simply began my projection by eliminating Oh-san's first 4 seasons. I did this because it wasn't until he changed styles, got a good coach and quit partying that his performance was noticed and anyone, including MLB scouts, could see his promise. If these four years were the American minor leagues, we know MLB would have called him up ASAP after or in the middle of season 4... I did not figure in a half season on these stats from these years, but if I had, I think he would have bested Aaron.

Next, it was easy enough to assume (as Mr. Albright did) that the man would have been in 20 more games a season those remaining 18 years of his career. Oh was in 130 games a season, full seasons generally speaking, so by adding 360 games and a modest amount of at bats, 1187 alongside a .300 average we really aren't making much of a stretch. As far as adding home runs in those additional chances, forget that idea. We all want to think he would excel, but let's say he didn't. I want to be fair and not go into this thinking he would have hit 1000 home runs in 10,000 at bats. Nonetheless, I do believe as he learned to read Japanese pitchers, he'd have done the same here and we aren't here to turn Oh into a God.

As far as existing home runs, I considered Japanese stadium sizes and that our parks are 300 to 350 feet out at the sides from home plate, 400 to 420 down the middle and 8 to 35 foot high if you want to hit one. This eliminates a good 100 home runs from Oh's record on top of the 75 I removed during his 'rookie' or 'minor league' years. Why? They weren't over 330 feet. We couldn't be sure if he'd have hit them over a fence here in the states.

What do we do with 100 home runs that fell short? If we called these 100 fair balls doubles, assumed he scored or drove in 1/4 of those times he hit one, he would benefit a little in the batting department but lose 75 or so runs and runs batted in. He may actually lose more on the runs batted in stats, maybe twice as many as he lost in runs, so I considered this.

I also highly doubted he had half the 450 intentional walks or anywhere near the 2250 overall walks left either. We aren't as walk happy here and only Ruth, Bonds and Big Mac are there to compare such a rationale to. I estimated this gave Oh 950 more hitting opportunities. Again, trying to be rational, I gave him a .260 average in these at bats, just like he hit for real against the MLB in All Star games. Let's not forget that with 360 extra games we should allow for him to score or drive in a run every other game, hit a double on par with his average years, possibly a triple instead each season for giggles, and then a modest number of additional walks as it only makes sense. You can't say I haven't been fair nor forgotten to cover anything so far, can you? Ah, yes, the home runs....

Speculating with 2140 overall added at bats, if he had them in Japan, Oh would have averaged 107 more four baggers if we used the % of his real ones that traveled more than 360 feet. Oddly enough, we are standing at 691 homers without giving the man anything extra to work with. Maybe that's too much, 107 more would have him at 798 career round trippers. Let's cut back further. If we added a homer for the same % he hit them over 400 feet while in Japan, that'd be just 44 more. He would be at 735 over his 18 years. Is that fair? Aaron hit 755 in 23 seasons. Ruth, 714 in 21....

We know that in 110 Japan vs USA exhibition games Oh hit 25 home runs in 338 at bats. If we used these figures and compared the homers per game to the 2700 games in my theory, that's 613 homers Oh stands a chance at. If we compared things by how many he hit per at bats coming to the plate 9900 times, he would have 730 homers. And that's just 18 seasons. I am certain he would have played a 19th season if he had 690 standing, if for no other reason than to pass Ruth. I know if he had 730 total he would have tried to pass Aaron at 755 with another season. It's just as safe to say he would have stuck around to pass Mays at 660 or played it just to play if he had only 613. I projected Oh would have been right at 3000 hits either way too so really, only home runs would be the issue at hand.

For a 19th year I added his actual game stats against MLB players to the projections just to say he did play a final season at age 41. That's not too much to figure in, he only hit .260 against in those games. Besides, the Japanese do not play on rest. They practice all day long even on game days. We in the states know the value of recovery and becoming stronger with recuperation. The best hitters in Japan aren't like Ty Cobb who retired hitting .367 for a lifetime. They are career .320 hitters at best over there. You also cannot tell me the Japanese pitchers are easier to hit. There are too many stories of how the Japanese held their own or select pitchers embarrassed our Major Leaguers at times. Still, Oh hit .300 for his career, comparably that may as well have been .344 here in the States...

If we dropped his first 2 and last 2 years against the MLB pros where he was too young and too tired, he hit .300, not .260, and he a homer every 10.72 at bats... Better than Ruth, Bonds... Aaron.... But who knows? My argument is simple: I understand why Mr. Albright chose to project his stats for Oh the way he did, however, I feel Albright was incorrect. .

For the sake of argument, I tossed in 200 empty at bats that should have been left as walks. I did this so no one can claim I was being pro-Oh.

Mr. Lemmon then provided a comparison of his projection to the outdated projection of mine he was using. Since I'm going to introduce a newer projection, I felt it would be better to insert my comments, and then provide the comparison of our two projections.

First of all, let's be clear about one thing: Mr. Lemmon and I both believe Oh was an absolutely top-notch talent. While I do not believe he was as good as Mr. Lemmon does, I see him as legitimately equal in quality to those men who were selected to Cooperstown the first time they appeared on the ballot. That's a quite select group--the no questions asked Hall of Famers. Another thing we should be quite clear about is that for many reasons, not the least of which is the fact I am trying to convince American baseball fans Oh really was that good, I have tied myself to the evidence, even if I subjectively disagree with it. I feel I must be able to support each and every assertion possible with facts based upon the available evidence in order to have any significant degree of success in making my case. It is my sense that Mr. Lemmon is reacting more as a fan than I am, and does not feel the obligation to the available evidence as I do. That is not to say he is entirely wrong--at least with respect to doubles, I subjectively agree Oh's total is too low.

That said, I beleive some of the starting points for Mr. Lemmon's projection are unsupportable. He starts his analysis of how many homers Oh would have hit by looking at how far he hit them against Japanese pitchers. First of all, I really can't say how accurately the distances presented in Oh and Nagashima: Every Record and reproduced in this file really are. Secondly, this analytical approach assumes without any real study that Oh would have hit balls pitched by major leaguers as far and approximately as often as he did Japanese players. Certainly, the greater speed that major league pitchers possess would increase the distance of a well struck ball, but that same speed combined with the movement major league pitchers have compared to Japanese pitchers would decrease the probability of a well struck ball. There are many other factors which might have an effect on this assumption which is one of the key bases of Mr. Lemmon's projection of how many homers Oh would hit (as well as how many doubles), none of which are investigated to see if Mr. Lemmon's assumption is indeed valid. The piece of evidence I will point to is how players who played in both Japan and the majors did in matched numbers of at bats, and that evidence would seem to indicate Mr. Lemmon's assumption is not accurate.

Mr. Lemmon looked at an older version of my projection and criticized my run and RBI figures. I had already been through that argument, and decided it was best to simply omit projections of those numbers because even after every other variable that can be adjusted for has been accounted for, you'd have to put a player in a specific slot(s) in a specific lineup(s) to deal with how many runs he would score and knock in. Since that is more than I want to deal with in a projection, I'd really prefer to leave such numbers completely out of the picture. If, however, people insist on my including such numbers, I would use a method which takes his projected runs created minus his projected homers and mulitply it by (if we are doing RBI) the following figure:

(RBI in Japan minus HR in Japan) divided by (his runs created in Japan minus his HR in Japan).

We then add back in the projected homers to reach the RBI projection. The runs scored approach is exactly the same except that runs scored are used instead of RBI. I think the numbers generated by this approach are reasonable given the projected stats of the players.

Mr. Lemmon also doubts my projection of Oh's walk total. I think the popular view is almost unanimous that I have given Oh too much credit in this regard. I certainly won't deny the figure I use is eye-popping. However, I can point to two pieces of evidence which indicate that the figure I'm using is too conservative. The first is the one I've already written about: the analysis of matched at bats between players who played in Oh's Central League while Oh was active as well as in the major leagues. When we look at using the actual conversion factor together with the factor expanding playing time, Oh projects to have about 800 more walks than my projection. Mr. Lemmon voices the common theme that "we aren't as walk happy here". However, that does not explain why the above average major league pitchers Oh faced in exhibitions against major leaguers from 1962-1980 (when Mr. Lemmon and I suggest Oh would have played in the majors) yielded 88 walks to him in 414 plate appearances. I'll grant that the games were in smaller parks with Japanese umpires, but if we project that same ratio of walks out to a career of 12,649 plate appearances (the number of plate appearances I use in my projection below), that is 2688 walks, about 20% more than the number I'm using. I realize the walks aren't in the same proportions on a year by year basis as they would be in his career, but I still don't think you can rationalize away a great deal more than that 20% differential from what above average major league pitchers actually did to my projection as to what Oh would have done against average major league pitching.

I'm going to spend a little bit more time on the walks projection, because I haven't gotten folks to understand how I arrived at this eye-popping figure which I maintain is conservative if anything. I recognized that what the projections were telling me would not be accepted. As a result, I chose his actual figures of walks in the years I felt he'd play in the majors. The figure I chose was lower in each and every case than any projection I could come up with based on the available evidence! Even so, it is a hard to accept number. Not only have there been hitters like those pointed out by Mr. Lemmon (McGwire, Ruth, Barry Bonds, and Ted Williams) who walked a great deal, but there have been hitters with a great command of the strike zone and far lesser power, such as Eddie Yost and Rickey Henderson. I maintain the evidence suggests Oh had such a superior command of the strike zone, and therefore my projection of his walk total is not too high.

Mr. Lemmon tries to support his home run projection by referring to the number of homers he hit in exhibitions against major leaguers. However, these homers came in significantly smaller ballparks. There is no attempt to account for this, and therefore I think this attempt to support his home run projection is a failure.

Mr. Lemmon assumes Oh would have hit for about the same average in the majors as he did in Japan. This ignores the evidence of matched at bats of Central League players of Oh's time who also played in the majors. I think it is possible Oh would have hit for a better average in the majors than the projection because teams used a Ted Williams style shift against him because he pulled everything. In the smaller Japanese parks, this gave him a rather small area to put balls in play, but he did so well enough to hit .300. The larger parks in the majors would give him a greater increase in percentage of room to work with than hitters who used more of the field. I sincerely doubt it would be enough to boost his average to .300, however.

Mr. Lemmon essentially argues that because some of the best Japanese pitchers could embarrass major leaguers, Japanese pitchers are the equals of major league pitchers. The problem is that while many of the best Japanese pitchers were excellent pitchers on any level, this does not address the quality of average Japanese pitchers compared to average major league pitchers. The matched at bats analysis addresses that issue, and finds that on average, Japanese pitchers allow batting averages approximately 7% higher than their major league counterparts to the same hitters in the same number of at bats. I trust that evidence.

I've indicated I agree with Mr. Lemmon that my approach probably underestimates Oh's doubles total. The main reason is that for those home runs which are not effectively turned into outs by the reduction in batting average, the rest wind up as singles. I believe that for a hitter with Oh's power and the fact he'd have more room to work with in major league outfields when they shifted against him than he did in Japan, I don't think that approach is realistic. Unfortunately, I haven't any hard evidence to back up my subjective sense of the facts on this issue, so I'll revert to the numbers I've got even though I am dissatisfied with them in this respect.

Finally, Mr. Lemmon's effort made me realize that my previous projection had actually cheated Oh in one respect. When I adopted his actual walk figure without any adjustment, I essentially took away the portion of his plate appearances which would occur by multiplying the greater percentage of games by his walk totals. It turns out this deducted 475 plate appearances. For the more formal case, where I want to show I'm trying hard not to oversell Oh's case, it's useful to know this fact but to leave things as they are to allow for some extra rest, possible injuries, etc. However, if we are interested in the most accurate projection possible without concern for such issues of "politicking", I should account for those extra plate appearances. Since I don't want to put them back into the walk column, I put them in the at bats column and adjusted hits, and the extra base hit categories accordingly. Please note his on base percentage will drop slightly from the previous projection because these extra 475 plate appearances include no walks to boost the on base percentage.

Now we'll provide a comparison of my new projection to Mr. Lemmon's effort:
Projection   G AB Runs hits 2B   3B HR RBI BB Avg OBP Slg
Lemmon 2836 10218 1932 3087 519 31 716 2001 1528 .302 .393 .569
Albright 2995   10414   1477   2913   395 44   553  1661   2235   .280   .407   .485

Really, our major differences are on batting average, walks, and home runs. We've discussed those issues already and I will not belabor them. Beyond that, they are either not major differences, arise from the other factors, or, in the case of doubles, the result of my being wedded to evidence which I subjectively feel underestimates Oh's doubles, as Mr. Lemmon argues, though I'm not sure I share his assessment of the degree of the underestimation of this factor for Oh.

Mr. Lemmon responded to my comments with the following:

I feel if Negro League players can be put in the HOF alongside Cuban players, and other people who are of questionable caliber or missing verifiable statistics, Oh certainly belongs there too.

I am a fan, without a doubt, but not one who screams at the TV when an ump misses a call, nor one who attends games in full team gear or body paint in the coldest of weather. In fact I did not enter this as a debate but as a quick viewpoint. It's usually the fan who feels the need to find more things to be right about before they hang it up. My wife is a pregnant member of Mensa. So trust me, I know what it's like to be the middle of a discussion versus a debate or an argument!

I fully understand the difference in Nippon vs US leagues, bat manufacturers, balls, parks, player strength and culture. This is why I spoke generally on figures and didn't dwell on trying to make things seem 'bottom lined.'

U.S. players who move to Japan and excel there and yet not so much here can attribute this more to than merely being superior human beings or playing against lessor opponents... This would be a debate all in itself!

Oh was walked more often against U.S. All-Star players because he was pitched 'around' more often than challenged and he wasn't falling for it by swinging. I actually spoke to a few people who pitched to him. Not sure if you knew I owned a magazine that had 100,000 copies distributed each month and am a nutritionist to many pro athletes. You'd be surprised how many doors such a thing opens.

I had fun doing this. I really didn't expect it to be presented as an argument though. I am more of a Jim Albright fan than anything else, to be honest. I am sure I could take the time you offered, sit back, rethink the details about what I know, do research on other things, or even speak with people I know from Japan, players who once played there themselves, etc., but that takes the joy out of it for me.

Other than that, hey man, this is great. I know I didn't mention this earlier but one of the first things I did years ago when I discovered the net was find your articles. I suppose reading A Zen Way Of Baseball brought me back, and I am happy it did.

Elect Oh!

As the posting author, I will allow myself the indulgence of the final word. I meant this as a debate and not an argument, and I hope it is not taken that way. Mr. Lemmon and I have our differences over the methods to follow and the resulting conclusions, but I have tried to stick to facts in my responses. I certainly do not feel Mr. Lemmon in any way disrespected my work nor that he was nasty in any way. I hope that feeling is mutual. I appreciate Mr. Lemmon's saying he is a fan of my work--too often all I hear is a deafening silence.

On a couple of specifics in Mr. Lemmon's response: I don't want to ever use the mistaken selections to Cooperstown to justify anyone else's selection--that only does futher damage to the Hall, in my opinion. Fortunately, it's fair to say neither Mr. Lemmon nor I feel inducting Oh would be anything approaching a mistake.

I can only say that the consensus of opinion is that NPB is not quite of major league quality (although close), and my research confirms that opinion. Further, since my research and methods in converting batting records have worked as well as they have for Ichiro, Taguchi, Shinjo, and the two Matsuis, I take such as confirmation of the general accuracy of said research and methods.

I can believe Oh was pitched around in the exhibitions to some degree because Nagashima, as good as he was, was a significantly lesser hitter than Oh. I would certainly accept that this could drop Oh's walk totals to where I project. However, I have a hard time accepting that this factor requires an additional discount of the exhibition results by about another 25-30%. Nagashima hit like Ron Santo or Ken Boyer, two fine players of his own time, and was quite used to having Oh walked in front of him (it almost certainly happened over 1200 times in their careers). Also, I have to believe the major league pitchers were more anxious to challenge Oh in these exhibitions to see how good he was, to give the fans a show, and to show that they, the major leaguers, were superior than they would have been after Oh demonstrated in the majors just how good he really was. I believe they would have found it necessary to pitch him quite carefully if he had been playing against them in the majors as well. After the fact memories are always subject to being subtly altered in retelling, and while I give the retelling of memories significant consideration, I rarely swallow such recounted memories whole.

I'd like to thank Mr. Lemmon not only for the correspondence and his work, but for making me think through my own efforts in a different way than I have in the past. I appreciate it because I think it helps me refine my own work, which I know is hardly free from error. Also, it gives me a chance to further explain my own work, for which I am grateful.

HomeGuru's Baseball Book StoreLink to UsBraintrust & Mailing ListsEmail the GuruContact InfoBaseball Analysis Home