Get To Know A Stat: BA/RISP (and what’s wrong with it)

Not THAT RISP!

Not THAT RISP!

“If you don’t get with the times, bro, you better step aside,” Matt Williams

“My favorite stat right now and always has been the stat of hitting with runners in scoring position… [B]atting average and on-base percentage and all of those things are great, but who is doing damage and how can they hit with guys in scoring position? I don’t know if I can help guys with that, but that’s the stat that I’m most concerned with, that I like the most. So do, I care whether a guy’s hitting .250 as opposed to .280? No, I care whether he can drive a run in.”  Matt Williams

Oh Faux Pas!  In the course of two sentences quotes Matt Williams just used the wrong soup spoon at the table of the advanced statistics crowd.  If this were other walks of life, Matt Williams might have essentially just said:

  • That the Star Wars movies are his favorite movies of all time…particularly Episode One… where Vader is a kid.
  • The he is a total foodie, and the Big Mac is his favorite sandwich
  • He loves craft beers.  Especially anything made by Miller or Budweiser.

Comes off a little snobby?  Probably.  But still, he did just walk into a room of hipsters wearing a Nickelback shirt… un-ironically.  So let’s take a look at just what folks find so objectionable about Batting Average with Runners in Scoring Position.

Batting Average with Runners In Scoring Position

So, as you might imagine BA-RISP is exactly what it sounds like.  What is a player (or team’s) batting average with runners in scoring position.  On its face, this is a key stat-I mean what could be more important than getting that key hit with a runner on second or third? If the game is about scoring runs, wouldn’t this be the big one to get ? I mean if you’re a Cardinals fan (and I’m guessing you’re not) right now you’re pumped about the Cards ability to “do it the right way” and “get all the little things done” that bring a run home.  They crushed the BA-RISP rating this year.

But the stat, rightly, has some pretty big detractors as well.  Essentially, the idea is that treating BA/RSP as something other than just good hitting is wrong.  Batting is batting, whether there are guys on base or not.  Teams don’t just hit well when a guy is on based-they just hit well.

To test this out a bit, I think we could look at a team’s batting average and compare it to their batting average with runners in scoring position.  There should be some big swings if this is a skill-that is to say there ought to be some teams that do much worse with BA-RISP (cuz they don’t have enough GRIT) and other teams that crush it (Cuz the ghost of Tony LaRussa haunts their bats or whatever).

So let’s look at some numbers, shall we?  Per usual, I’ve done just enough math to get me in trouble.  I made three charts.  I went through the seasons 2011, 2012 and 2013 and organized a chart showing each team’s Batting Average, Batting Average with Runners in Scoring Position and the difference between the two.  (Note, If you’re feeling adventurous the second column features the same info except with wOBA instead of Batting Average.)  I also highlighted the teams that made the playoffs.

Here is the most recent, 2013.  The Cardinals are clearly an aberration, at 61 points higher than their batting average.  Why?  Because (as we’ll see) this doesn’t really repeat anywhere else over the few years we look at.  Nothing close.  The vast majority of teams, however, are +/- the .010 batting average points over the course of the year when comparing the two.  In fact, 6 of the 12 playoff teams had .010 difference or less in AVG, including two teams that hit worse with runners in scoring position.

2013

2013: Click for Larger

(As an aside, if you’re wondering how the wOBA can look so different remember that doubles, triples, home runs, etc. are weighted heavier than singles.  So if you’re getting a lot of singles with guys in scoring position, it won’t help you as much as getting bigger hits.  By some measures this may be a more fair look at what is going on-but I leave that to the reader.). 

Welll we better look at another year.  Here is 2012:

2012

2012: Click for Larger

Well, I think if you somehow thought that hitting with RISP was key to the making the playoffs you’d have to buckle here.  A full 6 teams made the playoffs hitting worse in BARISP, and another 2 are at less than .010 difference.  wOBA takes an even more dramatic swing.

Look at 2011

2011

2011: Click for Larger

With this year-start looking at particular teams and comparing them to other years.  The Rays in 2011 made the playoffs with -.020 swing, but missed them in 2012 with a plus .003 swing and made them again in 2013-this time up .010.  The Yankees follow a similar plan. Even the vaunted Cardinals have a negative year (and still make the playoffs) one out of the three years.  It’s across the board. Teams that make the playoffs or don’t, traditionally good or not, are not showing some sort of pattern to hit above (or even below) their normal average.

Well what the hell?

Now look-I’ve hid the ball on a few things here to help make my point.  Clearly things like Pitching and Defense factor into winning games, it’s not all just hitting. But the only reference to wins here is in regards to teams that make the playoffs.  Ultimately, I don’t think that this really disrupts the point I want to make here.  Specifically:

  • That right there is why folks who don’t like BA-RISP don’t like it.  It ignores all the regular hitting that goes into getting into scoring position.  Much like the RBI, all the glory is going to the guy holding the bat, and everyone forgets about the guy who hit the triple to give him a chance.  Better hitting teams hit better in all situations…and also give themselves more hitting chances (by extending innings).
  • The third problem to stem from the same place is Sample Size.  Those 1846 PAs for the 2013 Red Sox with Runners in Scoring Position represent less than a third of the 6382 the team took. As sample size shrinks, the variation increases.  Any player can hit .500 with only two at bats, but with a 1000 it gets to be impossible.  Over 30 different hitters at any given time, 1846 PAs just isn’t that many-and it certainly isn’t nearly 6400.
  • Which links to the last point.  In SABR Bible The Book there is a chapter devoted to how, even after a full season, a true .330 wOBA (you can pretend its AVG if you like) hitter might end up +/- .016 points and still be a .330 hitter (the difference being, basically, luck or what have you).  That’s the first standard deviation.  I haven’t done the math myself, but I can’t imagine the SD in this case would be much different. That is to say if we consider any difference between AVG and BA-RISP within the +/- .016 (or even .010 from what I see) to be just dumb luck….well then for the most part there is no such thing as BA-RISP.  Or, at least, it’s not any different than hitting in general.

You want a team that hits well, not just a team that hits with RISP well.  If you have a team that hits well they will hit well with RISP (+/- a negligible amount).

BA-RISP has its uses.  It can tell you why a team scored many runs (because they made hits with guys in scoring position).  It just isn’t going to tell you what will happen next. You can’t “up” your BA-RISP as is evidenced by the fact that basically all the teams above fluctuate around their regular Bating Average.

If you had to guess how a team would hit with runners in scoring position, you are going to be closer than most if you just stick with the team average.  If you want a team to score more runs you need players that hit better, period, not guys who hit better with runners in scoring position.  Here is one last chart to help drive that point home.  Same 3 years, same teams, this time organized by wOBA top to bottom.  Highlighted teams made the playoffs.  This big Blue Line divides the top 15 teams from the bottom 15 teams in each year.

wOBA

All but 3 teams were in the top 15 in hitting by wOBA over the last 3 years.  The other three teams were all .003 or less away from the top 15.  I think that speaks for itself.  Matt Williams favorite stat might be BA-RISP, but he should save himself the trouble of sorting on Fangraphs and just stick to AVG/wOBA which are right there on the front page.

3 thoughts on “Get To Know A Stat: BA/RISP (and what’s wrong with it)

  1. The way I see it, batting average is, like winning percentage, a binomial distribution. A hit is a success, the average is the probability.

    Now, assuming 5550 ABs for a team for a season, the standard deviation for the average is between .0055 and .0060 (around 32 hits), depending on the average.

    Based on 2013, the average number of ABs w/RISP was about a quarter of the total. Making the sample size that much smaller increases the standard deviation doubles to .012 (16 hits or so).

    Looking at 2013, only 8 teams had a BA-RISP that was more than one standard deviation (.012) away from the overall BA average. And only one team was outside of two standard deviations, which is typically the confidence interval for determining if a sample average (BA-RISP) is different from the population average (BA).

    So, to me, it means the 2013 Cardinals were an abomination, I mean outlier.

    • Hahah!

      Well good! I’m glad that my gut was right that it was probably somewhere between .010 to .016 for a standard deviation. .012 Seems legit, and as you say, most teams are in that first SD.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s