June 7, 2009

Prospectus Idol Entry

Everything I Know About Baseball I learned In Sandlot

by Brian Cartwright

Printer-friendly

I got started on a life of number crunching playing in and then keeping the records for the summer league in my hometown of Johnstown, Pa. Since 1946, our league has been the host for the All-American Amateur Baseball Association National Tournament (AAABA, pronounced Triple-A-B-A). Each summer, nearly 300 of the country's best players 20 and under compete for a week in a 16 team double elimination tournament in front of scouts from most every major league team, in what can be a pivotal step in their dreams of fame and fortune as professional baseball players. When I was 18 and playing in the Johnstown league, I walked into the City's Recreation Office and asked if I could get copies of past season's batting and pitching statistics. It turned out their statistician had quit, and I ended up being offered the job. As a result, within a few years I found myself in a rather unique situation, armed with both a collection of Bill James' Baseball Abstracts and a stack of score sheets containing detailed play by play ready to be analyzed. Many of the core principals that I rely on today were learned watching those high school and college players on the summer sandlots.

Many of those who came to Johnstown did go on to star in the big leagues. Before I started working as statistician and head scorer in 1978, those who played in the AAABA included Al Kaline, Joe Torre and Reggie Jackson. An average of two or three players each year could be counted on to eventually play in the majors. In my years up until 1990, I saw Orel Hershiser, Chris Sabo, Shawon Dunston, Walt Weiss, Chris Hoiles, Jim Abbott, Todd Jones, Denny Neagle, Jim Leyritz, a fifteen year old third baseman named John Smoltz, and many others who were less well remembered in the big leagues. Each season, I would wonder how the best of the players I watched all summer in Johnstown would project as professionals. I had the league and pro records of Pete Vuckovich and Gene Pentz, and later Shawn Hillegas would make it to the majors as well. Not having a sufficient sample of amateur players, I bought a few packs of note cards and a copy of 'Who's Who in Baseball' in an attempt to compare the minor and major league stats of at least a few hundred players. The paperwork turned out to be too daunting, and would have to wait twenty years before I could run the numbers through Microsoft Access to create my 'Oliver' projections.

The Johnstown league has played on a variety of fields. The flagship is the city owned Point Stadium, a former minor league park seating over 10,000 built in 1926 and rebuilt in 2006. Fit into a rectangular city block, the dimensions are reminiscent of Fenway Park. Only 262 feet down the left field line in the original stadium, a 30 foot screen extending into left center keeps many fly balls in play, while right center extends to well over 400 feet. Other local fields ranged from one with a 320 foot fence to all fields to others with no fences. The Point had a 60 foot backstop, while most fields had no foul territory. I learned that foul outs decreased all batted ball types across the board. With short fences, the outfielders play closer to the infield, allowing fewer singles by catching more short flies. However, playing up allows balls to get through the gaps quicker, resulting in a higher percentage of extra base hits, but virtually no triples. Conversely, with no or deep fences, the outfielders play further back and have more singles fall in front of them, but have more time to cut off balls in the gap. Those balls that do get by the outfielders are much more likely to be triples. In developing a table top dice simulation similar to Strat-o-Matic or Pursue the Pennant, I tabulated the batted balls in each fielding zone at each park to find the percentages of outs, singles, doubles, triples or homeruns. This let me understand that once the batter has put the ball in play, the pitchers job is done, with the fielder and park determining the outcome for that type of batted ball (such as 'line drive to right center'.) Expanding on Bill James' 'Defense Efficiency Rating' for teams, fielders could be judged by the outs, hits and total bases allowed given the mix of batted balls hit to them.

Johnstown is one of two dozen cities in the eastern U.S. that have AAABA affiliated leagues, including Brooklyn, Buffalo, Philadelphia, Altoona, Pittsburgh, Cleveland, Cincinnati, Chicago, Dayton, Youngstown, Chicago, Birmingham, Atlanta, and the 'Big Four' powerhouses of New Orleans, Detroit, Washington and Baltimore. For many years, including the time I played, Johnstown was a city run 'recreational' league, intended to give area youths a chance to play each summer. Recruitment was limited to those who lived within 25 miles of City Hall, and as a result the league champion never had much of a chance going against the champions of the more 'elite' leagues. Washington, for example, defines its 'local' unlimited recruitment area as Virginia, Maryland and the District of Columbia, with three additional players allowed from anywhere else in the world. While the NCAA still allowed the practice, the University of Michigan and Michigan State each kept their underclassmen together as teams in the Detroit league. One, this taught me that if the talent level of the batters and pitchers in a league were roughly equal, regardless of what that level was, the statistics would be in an equilibrium, with batting averages near .270. Secondly, that the ratio of the talent pool to the available rosters spots defined the level of competition in a league. For example, for many years pitchers in the Johnstown league likely collected some horrendous Pitcher Abuse Points. In 1978, the league's best pitcher started 16 seven inning games and completed 15 of them in a 40 game schedule, then blew out his arm in the first game of the playoffs. Starting the next season, cautious managers sent their starters out once a week, and the offense skyrocketed once the balance between batting and pitching had been altered. In the major leagues in the 1990's, expansion, expanded bullpens and a decrease in workload for front line starting pitchers contributed (along with new ballparks and likely changes in the baseball) to a similar increase in offensive totals. Front offices responded to the lack of depth in their pitching by searching out new talent pools - more aggressive scouting of Latin America and expansion into Japan, Korea, Taiwan and Australia.

One summer I had my own experience with international baseball. Johnstown's reputation in hosting the AAABA earned it the host site of the 1984 'World Friendship Games', featuring age 16 to 18 players from eight countries. In the AAABA, there was variation in the quality of the players from one city to another, but they all played the same style of baseball. Observing the international teams, I identified the 'American' style of the United States, Canada, Australia and the Netherlands, the 'Latin' style of Panama and Colombia, and the 'Asian' style of Taiwan and South Korea. For me, this also developed an understanding of game theory. Taiwan, the eventual champion, played extreme 'small ball', line drive contact hitters who didn't go for power but made frequent use of bunts and steals. The batters showed discipline in rarely swinging at pitches out of the strike zone. In response, the Taiwanese pitchers could not afford to throw many balls, and learned how to pitch to contact. The Panamanians were at the other extreme. As illustrated by former Pirates Manny Sanguillen and Rennie Stennett, their batters practiced 'grip it and rip it', willing to swing at virtually any pitch. In this environment, there was no need for the Panamanian pitchers to learn control. They likely got the best results on pitches thrown out of the zone. In games played against their own in Taiwan or in Panama, these 'talents' likely cancelled each other out, achieving statistical equilibrium. Not so when they played each other. The Panamanian pitchers, who against their own batters didn't need to throw strikes, proved incapable of throwing strikes when they did have a need, walking 15 Taiwanese batters.

After marrying and relocating to the Virginia suburbs of Washington, D.C. in 1985, I spent three summers doing stats for the Clark Griffith league. One of the AAABA's 'Big Four' franchises, I had the pleasure of watching Rick Reed, Willie Blair, Chad Ogea, Pete Schourek and others in my short time there.

My time in amateur baseball, which included the power to instruct the scorers in their recording of play by play, gave me a tremendous opportunity to observe and develop the principles upon which I base my analysis today.