Skip to content

Outing the Productive Out: Has ESPN Unlocked the Secret of Postseason Success?

2003/12/02

Did anyone else see this quote from Buster Olney’s November 25 piece on ESPN.com?

There have been 142 postseason series since 1969. In 129, one team or another has had an advantage in Plate Appearances — and in 63.6 percent of those 129 series, the team with the advantage in Plate Appearances has prevailed. Factor in the 13 series in which opposing teams have tied in Plate Appearances, and it can be said that teams with a deficit in PAs have won just 33.1 percent of postseason series.

Prior to this season, no team since the 1997 Florida Marlins had won the World Series with a deficit in this statistic.

OK, I’m fibbing a bit. This isn’t actually a quote. I’ve replaced the term “Productive Outs” in Olney’s original column with “Plate Appearances” and recalculated his numbers based on PAs instead of POs.*

“Why?”, you might reasonably ask. Because Olney’s article makes the claim that teams that play “small ball” are proportionately more successful in the playoffs than teams that emphasize walks and slugging (read: the A’s) because their offense is more “diversified”. The way Olney and Elias Sports Bureau purport to measure this is the painfully contrived statistic “Productive Outs”:

This is the Productive Out, as defined and developed by ESPN The Magazine and the Elias Sports Bureau: when a fly ball, grounder or bunt advances a runner with nobody out; when a pitcher bunts to advance a runner with one out (maximizing the effectiveness of the pitcher’s at-bat), or when a grounder or fly ball scores a run with one out.

What makes this ridiculous is that by Olney’s own methods, Productive Outs are neither the best indicator of postseason success nor even a particularly good one. Unsurprisingly, runs scored was the best indicator of the series winner in my study; in 112 of 142 series, the winner held an advantage in runs scored. It stands to reason then that a team holding an advantage in any statistic that positively correlates to run production will have better odds of being the series winner. In Olney’s article, Steve Hirdt of Elias acknowledges this with regard to home runs.

However the point of my Folger’s Crystals-like substitution of Plate Appearances for Productive Outs is that even a “run-neutral” stat such as PAs shows a stronger correlation than POs with winning postseason series.

Check out the following table:

Advantage 

held by:

 

Series Winner

Series Loser

Neither

Runs

78.9%
17.6%
3.5%

Home Runs

61.3%
26.1%
12.7%

Plate Appearances*

57.7%
33.1%
9.2%

Productive Outs

57.0%
34.5%
8.5%

Walks

56.3%
35.9%
7.7%

Productive Outs should enjoy an advantage over Plate Appearances because certain run-scoring outs (fly balls and ground balls with one out) are included by definition, while unproductive outs are carved away. Plate appearances as a statistic says nothing about what happened during the PA.

Why then are the percentages so close for the two stats? I think the answer is that both are a byproduct of another statistic that positively correlates to run production – namely, runners on base.

As Olney concedes, an offense can’t generate a Productive Out without baserunners to move over. The more baserunners an offense generates, the more Plate Appearances it accumulates, the more runs it scores and the more series it wins. If this is the case, then On-Base Percentage should show a strong correlation to winning postseason series.

Let’s run the numbers again:

Advantage 

held by:

 

Series Winner

Series Loser

Neither

OPS (OBP + SLG)

73.2%
26.8%
0.0%

On-Base Percentage

72.5%
26.8%
0.7%

Slugging Percentage

71.8%
26.8%
1.4%

Productive Outs

57.0%
34.5%
8.5%

OBP, Slugging and OPS all show a much stronger relationship than Productive Outs to postseason series winners. 73.2% of postseason series winners enjoyed an advantage over their opponent in OPS. The A’s emphasis on OBP, not simply walks as Olney states, is redeemed by the 103 of 142 series winners that posted higher OBPs than the losers.

The lesson to be learned here is not that Productive Outs play no role in winning post-season series. They most certainly do, especially in late-inning tied or one-run games.

The lesson to be learned is that if you go to all the trouble of inventing a new stat to support your argument, be sure it says what you meant it to before breathlessly unveiling it to a skeptical public.

*Actually, total PAs by team for each postseason series were not shown in the otherwise magnificent Baseball-Reference.com database, so I substituted AB+W. For purposes of narrative clarity, I have called these “Plate Appearances” above. Sorry to mislead you.

Advertisements

Comments are closed.

%d bloggers like this: