Evolution of baseball player evaluation

Evolution of baseball player evaluation has taken place over many years. Player evaluation is the process by which general managers and other baseball personnel judge the ability of a baseball player to contribute meaningfully to his team.

Baseball has been around for a very long time. As it has matured, survival as a franchise has required greater competitiveness. With so many teams vying for only a few playoff positions, any edge possible to win is sought after by every organization. Putting the best players on the field, within reasonable budget limitations, is the greatest edge a team can have. Winning consistently requires talent, and historically, talent costs money. While that is still true today, teams are finding new ways of evaluating players so that they can field competitive clubs without spending nearly as much as other clubs. Player evaluation, therefore, has evolved over time, changing from an experience-based evaluation gained from scouts watching prospects play games, to one where statistical inferences play a greater part.

Adding statistical inferences to player evaluation has caused a division between some traditionalists and newer thinkers. Many personnel evaluators still do not believe in the information told by the statistics. Most of these evaluators are those scout who have been in the game a very long time and believe they know from experience what makes a great ballplayer. Finding players who can hit, hit for power, run, throw, and field well is enough evidence for these scouts to recommend drafting or signing them. Some general managers, however, believe in the power of numbers. It has been suggested that a strong correlation exists between runs scored and the sum of on-base percentage and slugging percentage. Because scoring runs is very highly correlated with winning baseball games, many general managers and player evaluators now place a premium on patient hitters who draw walks. This is one example of how player evaluation is changing.

The significance of a new view on player evaluation is that it could potentially create the best possible competitive environment in baseball. It will never be possible to limit the spending of major market teams without creating a very low salary cap, which the players would not accept. By giving teams ways to evaluate players, placing a premium on a skill that is not expensive to require, more franchises can compete for the playoffs in a given year. This competitiveness is baseball’s goal, as fan attendance and revenue increase the more competitive the sport is.

The state of baseball competition

Among the chief concerns of Major League Baseball is to maximize profit, since it is a business in the end. To do this, attendance needs to be as high as possible, and the number of television viewers also needs to be high to increase the value of television contracts. It is evident that a major influence on attendance is success or, at least, the perceived chance of success in a particular season. Early in the 2005 season the Washington Nationals won a lot of games and were in first place in their division for some time. During this time, attendance was very high. As the season progressed, however, the Nationals slipped a bit in July and August and as their playoff hopes diminished, so did the attendance figures. It is in baseball’s interest to create more competition in baseball so that attendance figures increase across the board. With more teams making money both the cause – increasing the competition level – and the result – increasing attendance and revenue – will continuously support each other.

In the 1990’s, if a baseball team spent a lot of money it won more often. Major League Baseball’s Blue Ribbon Panel on Baseball Economics conducted a study on this issue that was published in June 2000. Among its contents were some startling facts regarding the likelihood for success of low budget teams. In the five-year period from 1995-1999, no team spending less than the median team payroll amount won a single playoff game. If this trend were to be true all the time, it means that no small market team, and many mid-market teams, would stand no chance to compete against the baseball powers. If you cannot succeed, why put a team on the field in the first place? Major League Baseball has attempted to create policy to address this issue, but teams began to find ways around it as well. Very soon after the period addressed in the report, the Oakland Athletics began to win division championships and some playoff games despite having one of the lowest payrolls in the league. Using a combination of statistical analysis and traditional player evaluation, Oakland general manager Billy Beane was able to acquire inexpensive talent that helped his team win. This is the essence of player evaluation, and Beane’s success has led others to implement similar strategies in an effort to help their teams win.

The old way of evaluating talent

For years, talent evaluation was based on a set of tools a player could possess. These include the ability to hit, hit for power, run, throw, and field. It is certainly good to find players that actually possess all of these skills, but it is rare, and many times scouts look to high school players who compete against talent far below their own, just to find potential hidden gems and lock them up before other teams can find them. Often when these players reach college or the minor leagues, depending on whether or not they are drafted, holes in their game become evident as the overall talent level increases. Many times, power hitters will prove to strike out a lot, or the excellent fielding shortstop will be an average hitter with above average speed but does not add much run scoring ability to the lineup. Speculating on amateur talent can pay off over the long run, as many of the superstars that develop were drafted out of high school. However, with salaries for high draft choices rising continually, the investment became too much for some lower-budget teams. They wanted a safer investment, and began to move away from traditional scouting somewhat.

Using statistics

When a person wants to know what relationships exist between phenomena, statistics can be a very useful tool in finding those relationships. Baseball is no different in this case. Many general managers, looking for an edge within their budget constraints, turned to statistical analysis for that edge. Some of this statistical evaluation was detailed in Michael Lewis’ bestseller "Moneyball." "Moneyball", however, shed this sort of analysis into a pretty limited view. After reading the book, it is not unusual to think that Billy Beane and the Oakland Athletics staff were looking for only players who took a lot of walks. This is only part of the scenario. Statistical analysis has shown a strong correlation exists between runs scored and the sum of on-base percentage and slugging percentage. Because scoring more runs leads to more wins, this leads to the conclusion that the statistical sum (OBP + SLG = OPS) leads to more wins. In "Moneyball", Beane was simply looking for hitters that fit the mold.

One criticism of "Moneyball" is that readers can come away believing that Beane was the first to use statistics to evaluate players and become a more competitive baseball team. In fact, statistics have been used by managers and general managers for some time. In the 1940’s, Branch Rickey’s Dodgers hired the first known statistician, Allan Roth, to help make personnel decisions. Among those decisions was to trade the solid .306 hitting Dixie Walker who was out of the league just two seasons after the trade, as predicted by Roth’s numbers. In addition, Jackie Robinson was moved to the cleanup spot in the lineup in 1949 despite a career .296 average because he hit .350 with runners in scoring position. That year he batted .342, drove in 124 runs, and was named National League MVP. The Oakland A’s, even before Billy Beane’s arrival, used statistics to create an American League juggernaut in the late 1980s and early 1990s, winning three pennants in a period of five years. Their personnel decisions were driven by on-base percentage, and they acquired a number of frequent walkers to help build overall team success.

Beginning in the 1950s, a number of mathematicians of various backgrounds with an interest in baseball began conducting their own research into more intricate baseball statistics. Many new hybrid statistics were concocted during and after this time. Some of these statistics include OPS (on-base-plus-slugging percentage, OBP + SLG), total average (which also takes into account hit-by-pitch, stolen bases, caught stealing, and times grounded into double play), batter’s run average (OBP * SLG), scoring index ( ((1B + BB + HBP + E)*(TB + SB))/(plate appearances)^2 ), and runs created ( (H + BB)*TB / (AB + BB) ). Each of these statistics has good correlation with the number of runs produced by a team over the course of a season and, interestingly, some statistics correlate more strongly with team runs scored during certain seasons or decades than do other statistics.

OPS is the most widely discussed new wave baseball statistic. It is easy to calculate and intuitive. It makes sense to the casual baseball fan that getting on base and moving runners over together lead to the team scoring runs. Because this is a widely known and verified piece of information, it is incorporated into many teams’ decision-making processes when acquiring or evaluating new talent. Even without good metrics for evaluating pitchers or fielders, at least in as intuitive and direct a manner, general managers are now equipped with a new tool for making personnel decisions. This approach reflects a significant change in the overall philosophy of baseball personnel managers. There have always been a few willing to go outside the box, but now more and more managers are realizing the potentially valuable use of statistics.

Cautions about statistics

Everyone has heard the phrase “Lies, damned lies, and statistics.” This phrase is an accurate assessment in the potentially deceptive power of statistical inference. It is very possible, depending on what data are used for the analysis and the approach taken when conducting the analysis, to create a seemingly significant relationship between a large variety of statistical observations. In addition, those who do not fully understand the relationship described in a statistical model are prone to misunderstanding and using the information incorrectly. As the game of baseball evolves, more statistics will be created in the attempt to fully understand the game. While this represents a positive philosophy shift – it will likely level out the competitive balance of baseball some – this type of information being misused by a person not familiar with both the data used to derive the relationships and the process used to do so could potentially devastate a franchise, leaving it crippled for years. As this fascinating philosophy change among the management of the game evolves, it will be interesting to see how the information is used.

Potential future additions to knowledge

Much effort has been placed on determining the benefit a hitter gives a team based on statistics. However, less is available for pitchers and fielders. One statistic, WHIP (walks + hits per inning pitched) is a reasonable predictor of pitching success. It does not, however, give much insight into whether a particular pitcher gives up singles or home runs, even though the two types of pitchers may have identical WHIP statistics. OBP allowed would probably make a good predictor of pitching success, but it is not currently widely used. Also, not much has been done to incorporate the value of good fielding into the overall value of a player. It is common old-time baseball knowledge that good defense in the middle of the field (catcher, second base, shortstop, and center field) is paramount to good pitching success and overall team success. Teams can win without this solid middle of the field, but have to compensate by scoring more. More research into the value of fielding and pitching in relation to hitting will hold some value in determining the most efficient way for a baseball team to spend its limited cash resources.

References

*Devlin, Keith (2004). Devlin’s Angle. Retrieved from [http://www.maa.org/devlin/devlin_09_04.html]
*Levin, R. C., Mitchell, G. J., Volcker, P. A., & Will, G.F. (2000). "The Report of the Independent Members of the Commissioner’s Blue Ribbon Panel on Baseball Economics". New York, NY: Major League Baseball.
*Lewis, Michael (2003). "Moneyball". New York, NY: W.W. Norton & Co., Inc.
*Schwarz, Alan (2003). "The Numbers Game: Baseball’s Lifelong Fascination with Statistics". New York, NY: Thomas Dunne Books.

See also

*Baseball statistics
*Sabermetrics
*Bill James
*Value over replacement player

External links

*Society for American Baseball Research [http://www.sabr.org/]
*Link to MLB Blue Ribbon Report [http://mlb.mlb.com/mlb/downloads/blue_ribbon.pdf]