Atlanta Braves Pitching EFF - What is the EFF Statistic?

The following is a statistic developed by Randy Cox of Huntsville, AL and has been produced here by permission. Any use of this information outside of AllBraves.com must be granted permission. Any questions regarding this stat may be directed to

The "Eff" pitching statistic is an alternate measure of efficiency in a pitcher's performance. Unlike the traditional ERA statistic, "Eff" gives no regard to the number of runs surrendered. Rather, it is based on the number of "bases allowed" per inning, offset by the number of strikeouts per inning.

The rationale for the "Eff" is based partly on the difficulty involved in dealing with runs scored by runners who reach base on one pitcher, and actually score after a pitching change has occured. Since, for purposes of the ERA, a run is simply charged to the pitcher responsible for the runner reaching base in the first place, it frees the responsibility from the pitcher who may well have allowed the runner to advance from 3rd, 2nd, or even from 1st base to score the run. This causes the ERA to be a often misleading measure of effectiveness of relief pitchers, particularly those who commonly enter a game while an inning is in progress, and typically inherits base runners.

The Eff is an equally revealing measure of all pitchers, whether a starter, long/middle reliever, situational, or a closer. Is premise is based upon the fact that the action to be concerned with is allowing batters to reach, and advancing them once they've reached, since obviously, no one scores until he's touched 1st, 2nd, and 3rd bases first. The number of bases allowed per inning pitched is determined by:

( (HITS_ALLOWED-HRs)*(STANDARD_BASES_PER_NON_HR_HIT) + (HRs*4) + (BB,HBP,WP,BK) ) / INNINGS

TO EXPLAIN:

* (HITS_ALLOWED-HRs)*(STANDARD_BASES_PER_NON_HR_HIT)

Every "hit" is at least a single, or one base. But there is a big difference
in allowing 10 hits, all singles, and allowing 5 singles, 2 doubles, 1 triple, and
2 home runs. So first, we look at all "non-homer hits", or (HITS_ALLOWED minus HRs).

It isn't enough to just say 1 base per all non-homer hits, since some are for extra
bases. Since doubles and triples allowed is not typically available, we use a factor
that represents average-bases-per-non-homer-hit, which is derived from the current
Braves' team hitting stats. It typically ranges from 1.25 - 1.30 bases per such hit.

* (Hrs*4) Obviously, that's 4 bases per home run.

* (BB,HBP,WP,BK) Walks, hit batsmen, wild pitches, and balks all allow 1 base.

So for intance Mike Hampton's numbers on May 17 were:

Implementing the calculation,we have:

( (42 - 4) * 1.278) + (4 * 4) + (14 + 0 + 1 + 0) ) / 55

which results in 1.45 bases allowed per inning...quite effective!

This number is refined to the final EFF, which adjusts the Bases/IP downward according to
the pitcher's average strikeouts per inning (usually less than 1), divided by some number
which is yet to be determined as to what is best for this purpose. The larger the divisor,
the smaller the number subtracted from the Bases/IP, and the LESS difference an impressive
strikeouts/IP makes. So, it becomes subjective at this point.

My thinking is this...
As a pitcher, there is no more effective performance than to strike out a batter.

Consider 2 things...

1) If a pitcher has a low strikeout ratio, it means the batters are putting more balls
in play, and "making thing happen". Even if "out" balls are being hit, they are
subject to errors, and while not the pitchers "fault", a strikeout would not have
allowed for the possibility.

2) Non-strikeout type outs often advance baserunners...sacrifice bunts, sacrifice flies, infield grounders that either score runs, or move a runner from 2nd to 3rd, or even 1st to 2nd. Although there's no way to measure these advancements from the stats available, we can reasonably know that a lot more of it is going on with pitchers whose outs are registered by a high ratio of balls hit into play. Such pitchers are indicated by a low strikeout ratio. So, we deal with it by rewarding pitchers according to their strikeout ratio, where a high number represents less balls being hit into play, potentially moving runners.

Thus, the reward of a high strikeout/IP seems needful. Of course, it needs to be understood that using this moves the calculation from the more "concrete" number of bases allowed per inning, to a number that is simply being adjusted downward in a way that does not translate into a clear counting of base runners and their advancement.

AND FINALLY....

Each pitcher's EFF is then divided by the STAFF EFF (which is the same calculation, simply
using the staff totals). This causes each EFF to be translated to the Relative EFF, where
all values below 1.00 mean "Move effective than the staff average" and all values above
1.00 mean "Less effective than the staff average". The table is then sorted by this value,
and the accompanying chart reveals the current pitching staff relative effectiveness.

Incidentally, all pitchers who have contributed to 2005 stats who are no longer on the
roster are placed at the bottom of the table. (mlb.com no longer includes them, but we
have to include their numbers for season staff totals)

TO COME LATER.....

The "SSI"...the Situation Severity Index...to be utilized in a new statistic that is
formed and added to ONLY when a pitcher enters in mid-inning. This will measure a
reliever's effectiveness in dealing with the situation he inherits...and how he comes
out of it.