A Follow Up Note From Davis O. Barker
Interesting thoughts ..... ones I would imagine anyone who has done any "original" research has pondered ...... led me to believe that anyone who does the same stats (including the official league statistician) would most likely have different totals on something ...... there is too much margin of error to assume otherwise.
I have redone stats, as has Willie Runquist, and one does expect discrepancies, but there are certain stats that have to correct; e.g., if a team plays 154 games, then there must be 154 games started for the team’s pitchers. Inning pitched is the same type of stat. Where discrepancies creep in are generally on errors, assists, extra base hits, etc. I’ve found errors on at bats, and on hits, but if a league has batters getting 11,138 hits in a season, then you have the right to expect that pitchers gave up 11,138 hits.
One of the things I encounter often is discrepancies in box scores of one game in various papers ..... I often work in regional works, like the Dallas, Houston, and Shreveport papers, and local papers as well. It is a fairly common occurrence that a box will vary from paper to paper .... I talked to a retired sports reporter and a newer one about the issue ..... both said in the pre-computerized days it was usually about space and personnel.
The variances I have found are mostly in fielding, where some discretion is needed on the part of the scorer, and errors— I have found— are more often than not pure typos that come from the typesetter rather than the reporter.
The personnel that transcribed this type of stuff was usually lowest on the totem pole and often were not even baseball-oriented or even fans (have encountered the same when contacting a local library asking for a copy of a particular box score on a particular day) ....... regardless, it was always about space— if one column was too long, they would drop a line from a box - most often a late inning defensive sub or a pinch hitter ..... they are usually caught by the researcher because of the totals, but who the missing player is not usually identified ...... interestingly they often occur at or near the bottom of the page - when they suddenly realize that they have used their allotted inches ......
For the Coast League & California League I have found this only vary rarely. There are many papers who pride themselves on producing a good box score.
I found this especially true of the 1930s ..... why, I don't know ...... probably financial - fewer pages. Another thing I found was that until after WWII box scores were not standardized ..... big problem with locals in the 'Teens and early '20s was that local reporters (especially in towns without much past experience in OB), assigned to the game, may not have been baseball fans themselves ...... I can pretty much ID them because these papers normally ran only line scores .... and articles are one big baseball cliché of phrases.
If someone wishes to follow my tracks, they are welcome. The 1925 Texas Association stats took over a year of regular work and almost 30,000 miles on my truck to complete. Finding over 400 box scores in obscure, small-town papers was an ordeal. I was not long into the project when I realized why no one had ever done them.
Last summer I did the two O. B. seasons for the San Joaquin Valley League of 1910 & 1911, and in the hundred-mile radius of the towns I put 2,500 mile on my car, so 30,000 miles in Texas doesn’t surprise me. For the 1903 and 1906 PCL seasons, I traveled all the way up the coast to Seattle (via Fresno) to get all the box scores.
The bad thing here is that I always seem to follow some of the work of Vern Luse, whom I never met but greatly admire. Texas is a big place, and way behind in the age of microfilm. It is fairly easy to go on-line today and locate a paper ..... in Vern's day he was working hit-and-miss ..... bothers me when I find different stats .... knowing that I located more missing box scores, not because I am a better/more dedicated researcher, but because of technology and the fact that the trails have been blazed ahead of me by people like Mr. Luse. I hope that came out right and you understand what I am trying to say. I seem to meet a lot of people in this pursuit who seem to believe they are the only ones working ..... they are the only ones capable of doing legitimate research ..... or that their work is definitive and above question ...... I'm all for accuracy, but there is so much more out there that needs to be done that it is hard to get involved general scrutiny ..... one of my great fears is spending time, money, and sweat on something that has already been done ...... that is why some form of informational clearinghouse is vital ..... thanks to you, there is some form of gathering point.
The accuracy I’ve been talking about is one of properly compiling averages. Different sources obviously will lead to slightly different averages, but if you are adding up the same source material twice, and get a different set of numbers, it might be incumbent to go back and correct your numbers. And as to Charlie Chech— innings pitched cannot be off that much without an error: Counting a couple of games twice; counting 8 innings as 9 over the course of a season; missing reading the name of a pitcher; or any combination of the above.
As to Vern Luse, the leagues he compiled were leagues where most of the box scores appeared in Sporting Life, and supplement with other papers in those leagues. What Davis and I do are travel to towns where you pull up to the front door of libraries. In those libraries you sometimes find nuggets— like a second newspaper that only lasted a year— but covered baseball like The Sporting News! Minor league baseball research is a lot like detective work.
Lots of thoughts on these issues ....... things that have always bothered me as I compiled yearly stats ..... anyway, thanks for the blog!!!!!!!!!!
OK, Davis, now what does the “O” stand for?