AOS

By Roger Weber

Hitters: AOS

OPS accurately predicts a team’s run production and is thus a good indicator of player performance. It also correlates highly with a player’s runs created. It is based on the idea that a player’s responsibility is to get on base and to advance bases and help other players to do the same.

OPS = OBP + SLG

On base percentage measures a player’s ability to get on base. Slugging average represents the average number of bases a player attains per at bat.

Expanded:

OBP = Times on base / Plate appearances

SLG = Total bases / At bats

The two do not follow the same denominator. The reasoning is that walks do not contribute to a player’s performance the same way other hits do because they are not hits and do not create as many RBI. In reality a player on first base advances only to second base on the next player’s single 70% of the time.

Slugging percentage can be easily expanded to represent the average number of bases a player attains per plate appearance.

New SLG = (Total bases + Walks) / (At bats + Walks)

Or

New SLG = Grand total bases / Plate appearances

This new statistic can be incorporated to OPS

OBP = Times on base / Plate appearances

New SLG = Grand total bases / Plate appearances

New OPS = (Times on base + Grand total bases) / Plate appearances

A player’s success at stealing bases also can be incorporated to both OBP and SLG to more accurately reflect how often he remains on base and how many bases he on average attains. A stolen base gains a base, and 87% of attempted steals are of second base. Thus a time caught stealing causes a player most often to lose his position on first base, and the benefit of having been on base at all, in this system meaning two bases. Times caught stealing hold about twice as much negative value as stolen bases gain in positive value, a measure common to other respected statistics like stolen base runs.

New OBP = (Times on base – Times caught stealing) / Plate appearances New SLG = (Grand total bases + Stolen bases – Times caught stealing) / Plate appearances

New OPS = (Times on base + Grand total bases + Stolen bases – (2 x Times caught stealing)) / PA

Or

New OPS = ((H + BB) + (1B + BB + (2 x 2B) + (3 x 3B) + (4 x HR)) + SB – (2 x CS)) / (H + BB)

=

New OPS = ((2 x 1B) + (2 x BB) + (3 x 2B) + (4 x 3B) + (5 x HR) + SB – (2 x CS)) / PA

=

New OPS = ((2 x BB) + H + TB + SB – (2 x CS)) / PA

Removing the division by plate appearances, a player’s production can be summed.

2 x BB

2 x 1B

3 x 2B

4 x 3B

5 x HR

1 x SB

-2 x CS

This is not as intricate or complex as Linear Weights or more complex systems, but because its basis is so sound and accurate, there is no need to make this system more complex.

When divided by plate appearances, the average new OPS is about .870, with a standard deviation of about .1.

When simply summed and not divided, it can be called AOS, and the average total is about 550, with a standard deviation of about 85. It represents a grand representation of the number of bases hit for and the potential to advance bases from teammate production.

Pitchers: Durable ERA

Pitchers are more commonly accurately measured because the statistics used to measure them are based on runs. ERA is a solid statistic, and a pitcher’s durability can be measured using innings pitched.

The two can be combined simply and nicely so as to neither benefit starting pitchers nor relief pitchers overly much.

(100 * (8 – ERA)) + IP

This is useable because it allows pitcher production to be accurately compared with hitter production using the above formula on the same scale. It is based on Bill James’ analysis of which elements of the game are the most important to outcomes.