Abstracts From The Abstracts
The First in a 12-Part Series: 1977 Baseball Abstract
I own all of the Baseball Abstracts by Bill James. My 1977-1979 and 1981 books are re-prints and the 1980, 1982-1988 are first editions. I ordered the 1977-1981 books at the same time, and it just happened that the 1980 edition that was sent to me was an original.
The 1977-1981 Baseball Abstracts are rather crude. The pages of the first three editions were stapled using a plain card stock cover and back page. The 1980 and 1981 books were bound using a textured card stock cover and back page. All five were typed and copied with several noticeable strikeovers, white outs, and handwritten corrections throughout the pages.
As noted on the inside of the back cover, "The 1977 Baseball Abstract was written and compiled by:
Your suggestions and comments concerning the publication are invited."
Below the Baseball Abstract title on the front cover, it reads "FEATURING 18 CATEGORIES OF STATISTICAL INFORMATION THAT YOU JUST CAN'T FIND ANYWHERE ELSE".
You turn the baby-blue cover page and, boom, the first page is the beginning of the National League monthly hitting records in alphabetical order, starting with Rob Andrews of the Houston Astros. James writes:
The first section of this book gives the complete month-by-month playing records of all major leaguers who appeared in 100 or more games in 1976. These records, compiled from the daily box scores carried in the newspapers, provide some interesting insights into the course of the campaign. For example, was Robin Yount, the 21-year-old Milwaukee shortstop, really strong enough to lead the league in games played? Check his monthly batting averages. The records of rookies are particularly interesting. You can "see" Hector Cruz learning to hit major league pitching, watch the league catch up with Jason Thompson. Wonder how Bill Madlock won the batting title? Look him up.
Flipping to page 18 out of the 68-page book, I look up Yount's stats and his monthly batting averages are listed at .341 for April, .272 May, .269 June, .276 July, .220 August, and .206 September. Yount tied John Mayberry and Rusty Staub for the A.L. lead in games played with 161. The curiosity in me made me look up Mayberry's and Staub's records. Mayberry, who turned 28 that September, "hit" just .175 in the final month of the season. The 32-year-old Staub hit a monthly low of .268 in September, lowering his batting average to 11th in the league at .299.
Underneath Yount, who is the last player listed, James asks the following trivia question:
The record for RBI's by two teammates in one season is 347, by Gehrig (184) and Ruth (163) in 1931. But who holds the season record for runs scored by two members of the same team?
ANSWER (typed upside down at the bottom of the page): The record for runs scored by teammates is also held by Gehrig and Ruth--in the same season! The dynamic duo also scored 312 runs during the 1931 season. Incredibly enough, the Yankees did not win the pennant, finishing 13 1/2 games behind Connie Mack's A's.
James lists the National and American League monthly leaders in the ten basic stats (G, AB, R, H, 2B, 3B, HR, RBI, SB, AVG) on the following page and the monthly team batting records on pages 20-24. He asks a second trivia question at the bottom of page 24.
Who was the only pitcher ever to lead his league in most strikeouts per nine innings and fewest walks per nine innings in the same year?
ANSWER: Walter Johnson, which is really to be expected. The trick is that he did it in the worst season of his career--1920. While Johnson had some eye-catching SO-W ratios like 313-76, 303-76, 203-56, and 243-38, the only season in which he--or anyone--led the league in both ratios was 1920, when he was 8-10 with 78 SO and 27 BB in 144 innings.
James provides the monthly won-lost records of pitchers on pages 25-31. The Cincinnati Reds, who were 102-60 in 1976, had seven pitchers with 11 or more wins, led by Gary Nolan with 15. Randy Jones, the N.L. Cy Young Award winner, was 18-4 at the end of July before sputtering home with a 4-10 record in August and September.
The next section of the book is on stolen bases. James breaks out "the world's only stolen-bases-against statistics" for catchers and pitchers. "Lacking access to the official records of the games," James "counted stolen bases against pitchers, and catchers, starts." He states "the distortion this creates is minor for catchers, who finish most of their starts, but fairly significant for pitchers."
In pointing out the importance of the catcher, James notes "there were more bases stolen against the Dodgers in the 54 games that Steve Yeager didn't start than in the 108 that he did." Yeager led the N.L. in stolen bases (51) divided by games starts (108) and was followed by Johnny Bench, Gary Carter, and Bob Boone. Yeager, Bench, Carter, and Boone were generally considered the four-best defensive catchers in the league during that era. In fact, one of the latter three catchers was awarded the N.L. Gold Glove every year from 1968-1982.
Page 34, which discusses pitcher run support, attendance by pitchers, and fast and slow pitchers based on average game times, isn't listed because the text almost runs off at the bottom (with the last line on the page handwritten). Steve Carlton, who led the majors in run support in 1976 with 6.17 per start, actually was responsible for provoking this study by James when Lefty went 27-10 in 1972 for a team that was 32-87 "when he wasn't on the hill." Bert Blyleven, for what it is worth (wink, wink), had the second-lowest run support in the A.L. at 2.75 runs per game vs. a league average of 4.01.
With respect to attendance, James concludes that Tom Seaver, despite comments by a New York sportswriter which triggered the study, Mark Fidrych and Jones were the only "pitchers who could definitely be said to be 'draws' in 1976." James also offers "the first actual data I've ever seen on how fast a pitcher works" and determines that "the fastest pitcher in the majors, Jim Kaat, was a staggering 42 minutes faster than the average game started by the slowest, Nelson Briles."
James then offers the finest writing in his initial book when he delves into fielding statistics over the next 8 1/2 pages.
...a fielder's visible fielding range, which is his ability to move to the ball after it is hit, is vastly less important than his invisible fielding range, which is a matter of adjusting his position a step or two before the ball is hit.
So if we can't tell who the good fielders are accurately from the record books, and we can't tell accurately from watching, how can we tell? We could tell, quite simply, the same way that we tell who the good hitters are--by counting things.
...There are a limited number of places to which a baseball can be safely hit. Suppose that we divide the baseball field into 16 sections. An infield hit could just as easily be recorded as a "single 1"...A single that goes between the first baseman and the bag could be recorded as a "single 2" (or double 2 or triple 2, as the case may be). The 3 hole would be between the first baseman and the second baseman, the 4 hole between second and short, the 5 hole between the shortstop and third baseman, the 6 hole between third and the baseline.
...It's a very simple system--any scorer would have it memorized in three games. And I sincerely believe that at the end of only one season, the impact of such a system upon the accurate evaluation of talent would be enormous, and enormously to be desired. Shortstops would have as much chance to become superstars as outfielders ("Did you see the season that Mark Belanger had last year? Only 22 hits in the 5 hole in 157 games! What a year...")
James also suggests we separate "fielding" in the sense of "catching" the ball from "throwing."
Take, for example, the Royals' third baseman George Brett. Brett in 1976 handled 501 chances, which is a lot, but made 26 errors, which is also a lot. Looking at that record 20 years from now, almost any fan would consider that Brett had good range in the field, but rather bad hands. In point of fact George has excellent hands--as good as anybody's. What he also has is a strong but inaccurate arm. A lot of his throws to first base don't go to the first baseman. Of those 26 errors, I'd bet 20 were on throws. Why not record the fact? Why not separate fielding from throwing. Lumping the two together makes no more sense than lumping together balks and wild pitches.
...The same holds for outfielders. Why don't we count how many bases are advanced on fly balls to the outfield? Mickey Rivers gets a lot of bad press for his arm, but how much does it really cost him? How many times last year did a runner tag up and score form third on a fly ball hit to Mickey Rivers? How about Cesar Geronimo? How many did he throw out trying? Or even, how many were in a position to try and didn't?
...But because fielding statistics are inadequate, we don't know. So why do we go on using a set of fielding statistics that has been outdated for decades? I'm not arguing for my system but for a system. If we think about it, we can well devise a simple system of fielding statistics that would be complete and informative. It is worth the effort.
In the next section, James lists every major league pitcher who pitched at least 100 innings in 1976, followed by the number of home runs he gave up per 1000 batters. James states "some pitchers will surrender home runs 15 times as often as others," making the case that a pitcher's tendency to allow HR "is a major part of their pitching record, but mostly an ignored one."
James proceeds to turn career records into seasonal notations, the first time I can recall seeing statistics expressed per 162 games. The following nine active players in 1976 had averaged 30 or more HR per 162 games:
Hank Aaron 37
Dick Allen 33
Johnny Bench 30
Reggie Jackson 33
Dave Kingman 36
Willie McCovey 35
Frank Robinson 34
Mike Schmidt 35
Willie Stargell 34
For reasons unknown to me, James spends six pages on a baseball simulation game using a deck of cards. I believe this effort died on the vine as I don't recall reading anything more on this subject in future Abstracts.
In the last section of the book, James discusses "another almost-new statistic, this one named power percentage" (or what is now more commonly referred as isolated power). "Power percentage is that part of slugging percentage which is accounted for by the extra bases; in simple terms, slugging percentage minus batting average." James informs us that "slugging percentage is a summary statistic, while power percentage is a descriptive statistic that applies to power alone."
Knowing that a player has a .400 slugging average tells you nothing about what kind of player he is, neither what kind of an average hitter or how much power. He could be a .350 hitter with no power or a .230 hitter with great power.
Put in its proper context, the 1977 Baseball Abstract is simply a classic.
By the way, did Hector Cruz ever learn how to hit major league pitching? The journeyman outfielder-third baseman, who hit 13 HR in 1976 and only 26 more the rest of his career, retired in 1982 at the age of 29 with a career batting average of .225.
Next up: 1978 Baseball Abstract.
[Additional reader comments and retorts at Baseball Primer.]