September 26, 2010

            Today’s games are the end of the second round of the tournament.   There are now 20 players left in the tournament; after today there will be 16.   These are the players remaining, and the schedule:

Baltimore Regional

Cleveland Regional

St. Louis Regional

Los Angeles Regional

Brooks Robinson 78, Doug DeCinces 68

            Trailing 45-39 at halftime, Brooks Robinson used his advantages in Career Length, Defense and Awards to survive a challenge from Doug DeCinces, and move into the third round of the tournament named in his honor.

            Gene Mauch in 1982 said that Doug DeCinces this year “has been the best player I’ve ever seen.”

            There had been several young third basemen in Baltimore who had waited around a year or two for Brooks Robinson to grow old, but he wasn’t in any hurry to accommodate.   When Robinson finally was finished, it was Doug DeCinces who stepped into his position.   Players who replace a legend are notoriously prone to disappoint, but DeCinces was more or less up to the challenge.   A muscular athlete with broad shoulders and a narrow waist, DeCinces hit 37 doubles and 28 homers in 1978.   Those numbers were bigger then than they are now.  The Baltimore Orioles in their history up to 1978 had never had a player with as many as 37 doubles and as many as 28 homers—and the other parts of his game were good, as well.   He was a very good third baseman, not Brooks Robinson, but a good third baseman, would take a walk, and hit for a decent average.

            Baltimore couldn’t reach a long-term deal with him, traded him to the Angels, and he had the year that Mauch so much admired his first year in California.   Gene Mauch, whatever his faults, was not somebody who was going to blow sunshine up your skirt.   He was a straight shooter, to a fault, and he had been around major league baseball longer than anyone could remember, and when he said that DeCinces was the best player he’d ever seen, he said that because he meant it.   DeCinces had hit .304 that year with 42 doubles, 30 homers.   Until 2009 (Kendry Morales) DeCinces was the only Angels player with as many as 42 doubles and as many as 30 homers in a season.   As to his defense. . .if he had played a few more games he would have broken the record for assists for a third baseman.   He had 399 in 153 games; the record is 412.

            There was also a perception, at that time, that DeCinces had had a great year in the clutch.   The record does not easily reveal this to be true; his clutch numbers for the season are not only not good; they’re bad.   20 of his 32 homers were hit with the bases empty.  He hit .334 with the bases empty, .259 with men on base, and hit just .232 in the late innings of close games.   He was third in the MVP voting.

            Here’s a fun fact for you:  DeCinces in his career hit .464 with a slugging percentage over 1.000 against pitchers named “Moore”.   A pitcher named Tommy Moore, he faced only 5 times but had two walks and a triple, making an OPS of 1.600.   Against Mike Moore, he hit .452 (19 for 42) with 4 doubles and 6 home runs.    Against Balor Moore, he was 6-for-11 (.545) with two more homers.   That’s 26-for-56 with 5 doubles, a triple and 8 homers against pitchers named “Moore”.

            DeCinces had back trouble most of his career; you could see it, watching him play; he had a very stiff back, like Canseco or Albert Belle.   He drove in 96 for the Angels in 1986, re-signed with the Angels for ’87 but was released before the season was over.   He didn’t have enough good years to put him at the level of Brooks Robinson, but at his best he was a very comparable player. 

Doug DeCinces—Career Wins and Losses

Graig Nettles 92, Ken Keltner 88

            Graig Nettles used advantages in six areas to grind away at Ken Keltner’s 27-3 advantage in batting average, holding on for a four-point victory to advance into the third round.

            As most of you probably remember, Ken Keltner’s friends in Milwaukee about twenty years ago organized a PR campaign to promote Keltner as a Hall of Fame candidate.   His career in objective terms is a little short of Hall of Fame stature; he probably needed two more years of the quality of his big season in 1948 to bring him up to the level of, let’s say, Ken Boyer.

            This chart compares Ken Keltner’s Win Shares in the old system, published about ten years ago, to his Win Shares value in the new system:

            As you can see, the WSV runs about 5% higher than original Win Shares in a .500 season, and significantly higher in an outstanding season.   As 300 Win Shares would basically put you in the Hall of Fame in the old system, it requires a WSV of about 350 to accomplish the same.   Keltner—obviously a very good player—was a little less than two-thirds of the way there, and would have been about 70% of the way there if we gave him credit for the 1945 season that he missed due to the war effort.   But it speaks well of him that his friends would make the effort.

Ken Keltner—Career Wins and Losses

Chipper Jones 94, Harry Steinfeldt 71

            Chipper Jones used a 27-1 edge in power to leap off to a commanding lead, marching to an easy victory over the infield companion of Tinker, Evers and Chance.

            Steinfeldt in 1906 led the National League in hits (176) and tied for the league lead in RBI (83).   He was second in batting average (.327), and also he led the league in being hit with a pitch (14), which gives him a pretty good on base percentage.   I don’t doubt that he was a very good player that year; I do have some doubt that he deserves to be scored at 31-2, as he was.

            Wins and Losses can be projected from runs scored and runs allowed, as you all know, by the Pythagorean Method.   Suppose there is a league in which the runs scored and allowed average 800 per team:

            But you can get essentially the same results by subtracting 400 from runs scored and runs allowed (if the league average is 800), and assuming that the ratio of the wins and losses will be the ratio of runs scored to runs allowed (over 400):

            So long as the winning percentages are between .333 and .667, there is very little difference between the two methods.    But if the winning percentage goes over .667 or under .333, then we develop serious disparities between the two methods:

            The Win Shares method relies on the accuracy of the alternative win projection system, which ordinarily works fine because almost all teams have winning percentages between .333 and .667.   However, there are a few teams in history—like the 1906 Cubs—who have gotten out of range, and then you get significant discrepancies between expected and actual winning percentages, and then you get unreliable evaluations of won and lost contributions.

            Just as I was writing that, I suddenly saw—five years too late—what I could have done to avoid the problem.   Oh, well.  Maybe I’ll live long enough to do Win Shares 3.

Harry Steinfeldt—Career Wins and Losses

Penquin Advances

            Ron Cey used a big advantage in power to jump-start a wire-to-wire victory over 1920s glove wizard Willie Kamm:

            The four players eliminated today (DeCinces, Steinfeldt, Keltner and Kamm) are all of similar quality.   Ranked by Win Shares Value, they range from 247 (Steinfeldt) to 220 (Kamm)—about a 10% difference, top to bottom.   All four players were outstanding defensive players, all with Defensive Winning Percentages between .642 (DeCinces) and .690 (Kamm).

Willie Kamm—Career Wins and Losses

Brooks Robinson Tournament Gold Gloves

            The system of Win Shares and Loss Shares, unfortunately, is not designed to say who should win a Gold Glove.    The system does not assign defensive wins and losses per position; it assigns credit for defensive play at each position, assigns responsibility for defensive innings and also for making outs, and translates of these into one defensive won-lost record.   It doesn’t create a won-lost record for a player as a third baseman; it creates a defensive won-lost record for him as a complete player.

            In cases such as when a player plays multiple positions or is used as a defensive substitute, this is a real limitation.    As long as a player is a regular player and plays only third base or plays 95% of his innings at third base, it’s just a theoretical observation.    As the players we are concerned about here are regulars and are, for the most part, dedicated third basemen, we can pretty much assume that their fielding wins and losses reflect their performance at the hot corner.

            In some seasons, like 1960, our field of 66 players includes all or almost all of the good defensive third basemen in the majors.   Representing 1960 here we have Ken and Clete Boyer, Brooks Robinson, Frank Malzone, Don Hoak, Ron Santo and Jim Davenport—literally all of the good defensive third basemen in the majors that year.   (The other regulars were Eddie Mathews, who was not a glove man, Gene Freeze, who was terrible fielder, Eddie Yost, who was old and who was never a great fielder, and Reno Bertoia, who was just a utility guy playing quasi-regularly for a bad team.)     While we don’t have that kind of coverage of all of baseball history, the players in this tournament do represent about one-third of all the regular third basemen in major league history (one-third of the seasons), and a higher percentage than that of the best defensive third basemen.

            I thought it would be fun, then, to look back year by year, and see who was the best defensive player in our tournament in each season in history.  We have at least one player in the tournament for every year since 1881.  The weak spot in our coverage is the years 1910-1922; all we have in that era is Heinie Zimmerman, some old guys who were past their prime, and a very young Jimmie Dykes.    Other than that, though, we generally have two or three outstanding defensive players in every season.   These are the best. . .the Gold Glove candidates from the Brooks Robinson Invitational Tournament:

            Denny was the only candidate 1881-1883; Nash joined him in 1884.   Denny was the better third baseman 1884-1886; Nash moved past him in 1887.

            Denny was still around in this era; Nash was just ahead of him defensively. 

            Collins’ defensive value of 18.88 in 1899 was the highest figure in the tournament, so let’s look behind that one a moment.

            Collins replaced Nash as the third baseman of the Boston Braves in 1896.  In 1899 the Braves won 95 games (95-57), and they did it with pitching and defense.   Despite playing in a hitter’s park (Park Factor of 107) they were third in the league in runs scored, but first in (fewest) runs allowed. 

            That appears to have been more defense than pitching.

• Their pitchers actually led the league in (most) home runs allowed.
• Their pitchers walked more than the league average number of batters.
• The team allowed 31 more earned runs than the first-place Dodgers did—but 44 fewer unearned runs.

When we start handing out credit to the fielders, then, there is a lot of credit to be given on this team.  Remember, in 1899 there were only about two strikeouts per team per game, and there weren’t many home runs.   Defense was much more important then than it is now, and also, third base was a more critical defensive position in 1899 than was second.   Teams bunted several times a game.   A third baseman’s ability to field a bunt was a critical skill.   There were many fewer double plays than there are now, which made second base less critical.  

      In this context, in which defense was so critical, Collins led the league in assists by 18.   His fielding percentage, .943, did not lead the league, but was a full 40 points better than the league norm (.903).   All factors considered, it is perhaps the most impressive defensive season at third base in the history of the game.

Collins remained the dominant defensive third baseman among this group until 1904, being replaced by Steinfeldt, sort of by default; we are entering the era now when we really are only dealing with one regular. 

Zimmerman—not a great defensive player—is winning these by default.

Kamm edged Dykes in 1926 by a margin of 11.14 to 11.13—the thinnest margin in the study. 

      In 1930 all four active contestant (Dykes, Kamm, Lindstrom and Whitney) were between 4.45 and 4.69.  

Elliott won the MVP Award in 1947.   In 1945 and 1946 he ranked first in the small group although he was a half-time third baseman and a half-time outfielder.  Keltner had the two best defensive years of the decade, but was up and down on defense as he was at the bat. 

            Boone was actually playing shortstop in 1951.   In 1955 Hoak finished first ahead of several regulars although he (Hoak) was a half-time player.

Now we are reaching the era of players that some of us remember.   Robinson’s 10.38 in 1960 was the highest figure by a player in the tournament since 1932, but Boyer passed him in 1961.  In 1961 Clete Boyer nosed out his brother, 11.82 to 11.05, Robinson third among nine players, and the same players finished 1-2-3 in 1962.   In ’63 it was Santo-Clete-Brooks.   In ’64 Boyer was first, Brooks second as they both won MVP awards.   The 1967-68 contests were both close between Santo and Robinson, each winning one.

      So then we have an answer to the question:  when did Nettles pass Brooks?   1970.   Brooks was second among the dozen or so candidates every year through 1974, and then Cey moved into second in 1975.   There were also several outstanding third basemen in the 1970s who are not in our study—Schmidt, Aurelio Rodriguez, Darrell Evans.   Buddy Bell won the first of his six Gold Gloves in 1979.

From 1984 to 1991, whenever Wallach didn’t win he was almost always a close second. 

Matt Williams also had a couple of years as a close second.