When generating my Major League Baseball rankings, I use several formulas that I have in my personal database in an effort to come up with objective measurements for every individual player and, consequently, teams.
For those that wonder about the process, here is a general look at how the rankings are derived.
The basis for the rankings starts with OPS, the combination of on-base and slugging percentage. While this newly-popularized statistic is by no means the complete package in evaluating a player's impact, it is a fair starting point.
With OPS as the basis, next comes addressing some of the shortcomings.
Speed is a part of the game that is ignored by OPS, so I factor in triples and stolen bases (minus caught stealing) to give some credit to those players who alter the game with their baserunning. The inference that I'm making, additionally, is that players who have a lot of triples and/or are effective base stealers are also the ones more likely to go first-to-third or score from second on a base hit.
Another item to deal with is pure power. While slugging percentage doesn't differentiate between the player who accumulates four singles and the player who goes 1-for-4 but get his four bags with one swing, it's not unreasonable to give more value to the player who can change the score of the game in one at-bat because there is more luck involved in hitting singles since balls hit in play can, theoretically, be fielded by the defensive team. The difference isn't huge, but consider it a tie-breaker so that, given the choice between two otherwise very similar players, a team would likely prefer the one who provides more home run power.
The next factor taken into account is RBI. While runs batted in are obviously greatly affected by the number of opportunities, there should be some recognition for the players that are consistently providing that service in the middle of the lineup. Because of the imbalance between teams, it's not a huge portion of the rating but, given otherwise relatively equal players, who wouldn't prefer the guy who drives in more runs per at-bat?
The final offensive measure taken into account for each player's rating is a negative measure for strikeouts, on a per-at-bat basis. With the premise that more good can come from putting the ball in play (eg. ground outs can still advance runners) than striking out, making it preferable to have hitters with lower strikeout rates.
While defensive impact is difficult to measure, it's also not fair to ignore this part of the game entirely, so there a defensive factor included in the overall rating as well. The defensive impact is measured by combining fielding percentage, zone rating and range factor to provide one defensive number. It's not a surefire guarantee to acknowledge the top defensive players at their position, but it's a pretty fair representation.
(Note: Catchers have a slightly different fielding measure that takes fielding percentage and stolen base percentage allowed into account.)
When dealing with pitchers, the measurement begins with earned run average as the basis.
WHIP, or walks plus hits per inning pitched, is taken into account as well since it can help determine if the ERA is merely the fortune of good luck or whether the actual number of baserunners yielded is resulting in the expected number of runs against.
Won-loss differential is the next number included and I use differential rather than winning percentage because, for example, it means that a pitcher who goes 20-10 is due to be rewarded more for his impact than a pitcher who goes 6-2 out of the bullpen.
The final statistic taken into account for pitchers is strikeout/walk ratio. While the strikeouts per nine innings measure is also valuable, my preference in this regard is to lean towards rewarding the efficiency of the pitcher.
Saves are also part of the equation, though not a huge part, but there needs to be some way of crediting closers for the pressure involved in getting the final three outs (or, in rare cases, more) of the game.
Given those individual player rankings, I compile team rankings in the following manner:
First, each team's standard lineup vs. left-handed and right-handed starters is set out, with more weight given to the lineup against righties because, simply, there are more righthanded pitchers.
Each team's starting rotation is also included, with the fifth starter counting for less than the four mainstays because fifth statters generally don't get the same kind of consistent action and, in the premise of which team would win a seven-game series (the theoretical premise of my rankings), the fifth starter isn't likely to have any impact as a starter.
When it comes to evaluating the bullpen, the closer is the most important, but the top three setup men are also getting measured since they are the ones most likely to be involved in "winnable" games.
On a daily basis, the value for each team is adjusted for handicapping purposes based on that day's particular starting pitcher, but for the weekly Power Rankings, the team value is a combination of the values assigned to each team's lineup, rotation and bullpen.
This methodology has been similar for several years, but is always in the process of tweaking and changing every off-season in an effort to better reflect current team values.
Scott Cullen can be reached at firstname.lastname@example.org