SABRMatt's study on "Gappiness"
Draw the 1st triangle so that it catches .500 SLGrs within it


Dave Fleming had suggested that teams with gaps --- > have a very hard time winning the World Series.  This made a fair bit of sense to me.  SABRMatt followed up by getting us some data on it:


Do gappy clubs that make the post-season do worse than clubs that are not gappy?  For each playoff team, I did the following:

I took the position splits for each position on the diamond including DH and PH but excluding P on offense and, for every position that was below average in OPS+ compared to the league split for that position, I gave points based on the formula (PA / 600) * delOPS+ (the difference between 100 and the split OPS+)

I took SPs 1 through 4 by innings pitched and then all of the remaining starters with at least 5 starts lumped together as well as the entire bullpen lumped together and found their OPS+ allowed and used the same formula as above (except that now delOPS+ is split OPS+ minus 100)

I did this for every playoff team from 1990 through 2014

I then broke the playoff teams up by "loser's bracket" (all teams eliminated before the WS), "runner up" (WS runner up) and champion.

Using my scoring system the average loser's bracket playoff team had a gap index of 87.4, most heavily concentrated on the bottoms of rotations, the bullpens and the skill positions (SS/CF/C/3B/2B)...and oddly...DH (seems AL clubs are having a devil of a time finding good DHs)

The average gap index score for runners up is 69.6

The average gap index for champions is 68.8

Implication: in tournament baseball, gappy clubs are easier to eliminate than non-gappy clubs (that is statistically significant, that difference, I tested with student t-test), in aggregate, but once you hit the WS, the difference is negligible...possibly implying that teams with gaps that can be "played into losses" by better clubs get knocked out quickly, but teams with gaps that are less of a factor (like 2014 SFG who had gaps at the bottom of their rotation, but didn't use much of that bottom), once they hit the WS, are not impacted.

Some points of caution:

This uses regular season aggregate scoring, and obviously, contending teams make moves to try to plug their gaps before the post-season. So the team you face in the playoffs may not be as gappy as their score indicates (but I suspect this is just as true for all contending clubs on average).

Also, post-season games are played different than regular season games and bullpen gaps that get cancelled by top performances in the regular season averages might wind being more important or less important depending on the series and opponent.

Food for thought.

- See more at:


:: golfclap ::

The teams in the World Series showed 70 "points" of gappiness and the teams that didn't make it showed almost 90.  That is one whale of a reflected phenomenon for a study of this nature.  It strongly suggests that "gaps" are a major handicap in making the World Series.


Exec Sum

For those who didn't want to plow through Matt's methodology, it seems to look like this:

1B 600 110 0
2B 600 110 0
3B 600 90 +10
SS 540 90 +9 (scaled to PA)
3B 480 70 +24
C 600 95 +5
OF's, DH   if all plus 0
SP1 ? 58 allowed 0
SP2     0
SP3   104 +4 (scaled?)
SP4   112 +12
All SP's w 5 GS   120 20
Bullpen   84 0
Total     +75


So, you might say that a World Series team (1990-2014) had seven "units" worth of a position performing at 90, not 100.   A non-WS team had nine such "units."

It's a Jamesian first cut; Matt is the first man to draw a triangle around this problem, such that the answer lies inside the triangle.  Drawing the sides of the triangle closer together, that is the job of anyone who wanders into the cubicle next.

As Matt says, such a simple formula isn't going to put "PAID" to the issue.  A team might (and does) put "hard RBI" men into its lineup, for Sept-Oct, that were nowhere near the ballclub in June.  Many other such complicating variables arise.

But!  This is a large amount of loose data, and it polarizes just the way we predicted it would.


Using his method for last year's Mariners, I count:

Pos PA OPS+ Black Hole Points
C 609 82 +18
1B 659 88 +13
2B 688 141 na
3B 670 116 na
SS 605 93 +7
LF 650 84 +17
CF 712 56 +52
RF 650 96 +4
DH 633 56 +46
SP1     0
SP2     0
SP3     0
SP4     0
SP5:  ER, BM, TW   80? +15 ?
RP     0
TOTAL     +157

Which is twice what it should have been.  Maybe I'm missing something.

Compass heading:  The issue we are trying to get to, in plain English, is "Hard RBI up and down the lineup, with elite pitching 1-2-3 and to close the game."

Thanks mucho,




I use OPS+ allowed for pitchers and scale to the same 600 PA against (BFP)...our SP5 cluster was actually only worth about +4 due to scaling and the presence of pitchers like Paxton in the sample. But Elias was worth +2 also (a little below average OPS+ allowed).
In any event...that gap score for Seattle 2014 sounds about right to me...they had a sucking chest wound playing CF and inexplicably getting 700+ plate appearances. And they had no DH and their no DH kept hitting clean-up. :)
El no esta bueno.
Take out those gaps and replace with, say, Billy Butler in a DOWN year and Austin Jackson in a NORMAL year...and you have a gap score of about 65. :)


This from Dave Fleming today:

What would be your definition of "great"? I'm not being a smartass, just genuinely curious. Mine would be 6-7 seasons of an OPS of 150+. Maybe yours would be higher. Regardless, such a steep cutoff would definitely create a group of elite players.
Asked by: rwarn17588
Answered: 11/4/2014

 It's an interesting question. My notion of 'great' is this: if you had an annual conversation about the best player in the game, was this guy a part of that conversation for a while? I'd actually be hesitant to tie it to a number: I prefer to keep it subjective. Right and Miggy and McCutchen and Cano are the guys I think of, with Posey and Giancarlo knocking on the door. Kershaw and Felix and Wain-o are the big names among the pitchers, with Sale and Price and Strasburg and Cueto tapping on the glass.

Okay, so there is one team in the majors that has 2 of the 7-9 "great" players in the game.  These 2 are at exactly the spots I like filled on a roto team:  #3 hitter at SS/2B/C, and Aircraft Carrier SP.
Their other semi-franchise player (whose 2014 season would have been #5 in Derek Jeter's career) is at 3B, another glove spot.
They've got Zduriencik stocking the farm.
No excuses for the period 2015-17, hoser.

Add comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd><p><br>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.


  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.