 
Baseball Primer Newsblog — The Best News Links from the Baseball Newsstand Saturday, October 06, 2012Holmes: Percentage of runners driven in is an important stat that should get more attentionWhat happens when a professional hitter violates the RBI machine code? Get Carter (Joe Carter that is)!
Repoz
Posted: October 06, 2012 at 07:29 AM  30 comment(s)
Login to Bookmark
Tags: history, sabermetrics 
BookmarksYou must be logged in to view your Bookmarks. Hot TopicsNewsblog: VIDEO: Brewers, Pirates brawl after Carlos Gomez triple
(85  2:59am, Apr 21) Last: Barnaby Jones Newsblog: Bryce Harper benched for 'lack of hustle' despite quad injury (103  2:21am, Apr 21) Last: theboyqueen Newsblog: Ivan Nova’s season in jeopardy after tearing elbow ligament (5  1:33am, Apr 21) Last: Dr. Vaux Newsblog: Doug Glanville: I Was Racially Profiled in My Own Driveway (410  1:25am, Apr 21) Last: Rob_Wood Newsblog: mets.com: Through hitting system, Mets aim to build winner (19  1:19am, Apr 21) Last: bobm Newsblog: Tim Federowicz has rough night behind the plate  True Blue LA (14  1:18am, Apr 21) Last: Robert in Manhattan Beach Newsblog: OT: The NHL is finally back thread, part 2 (162  1:12am, Apr 21) Last: Robert in Manhattan Beach Newsblog: OT: NBA Monthly Thread  April 2014 (384  1:09am, Apr 21) Last: Los Angeles El Hombre of Anaheim Newsblog: Zander Hollander, Sports Trivia Shepherd, Dies at 91 (11  12:21am, Apr 21) Last: puck Newsblog: Daniel Bryan's 'YES!' chant has spread to the Pirates' dugout (132  12:15am, Apr 21) Last: Monty Predicts a PadresMariners WS in 2016 Newsblog: Chase Utley is the hottest hitter in baseball and has a shot at .400 (72  11:39pm, Apr 20) Last: Lassus Newsblog: OMNICHATTER for April 20, 2014 (94  11:35pm, Apr 20) Last: Dale Sams Newsblog: OT: The Soccer Thread March, 2014 (952  11:01pm, Apr 20) Last: Jose Can Still Seabiscuit Newsblog: Pirates Acquire Ike Davis From Mets (54  9:15pm, Apr 20) Last: Canker Soriano Newsblog: Minuteman News Center: Giandurco: This means WAR (95  9:05pm, Apr 20) Last: BDC 



Page rendered in 0.4552 seconds 
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
1. depletion Posted: October 06, 2012 at 08:58 AM (#4256625)Calculating RBI% as the article states, basically penalizing the batter who draws a base on balls, is just absurd.
Maybe RBU%/(1  OBP) would be useful, runners driven in per out made with runners on. And yes, you do get an out for a sac fly.
agree that you HAVE to subtract IBB  you can't drive someone in if you have no opportunity. and agree that sac flies and squeezes should also count.
am not understanding why so many of all yall are like all  beeeg deeel  about driving in runners. i mean, not making outs is wonderful. no disagreement here. but still, increasing the runs scored by your own team is important, right???
Yes, normalizing counting stats by number of opportunities produces "junk stats."
Like BA, OBP, SLG, HR/PA, K/PA, BB/PA, TTO %, SB% ... all junk.
If you're at all interested in normalizing opportunities, the place to start is with Tom Ruane's article on rbi (and distribution of baserunners) at retrosheet.
without adjusting for offensive performance. in other words, of course Cabrera drives in a high percentage of runners, he's a terrific hitter. the stat tells me nothing.
It's not perfect but its useful. I'd rather use RBI% than just RBI. Are there flaws? Sure, but just because something isn't perfect doesn't mean it doesn't have value. Until park effects, weather effects, opposition quality, time if day. Umpire quality, individual pitch quality and The other 800 little details that go into every single moment if a players life can be quantified all stats are going to be somewhat flawed.
EDIT: and the side bonus of using SLG rather than some form of rbi is that it doesn't treat a walk with a runner on base as an implicit failure.
Without those inclusions, this is as dumb of a stat as productive outs, save percentage(that doesn't include holds), or that silly runs stat that is runs+rbihr.
Sure, but so what? I would rather ride a horse than a donkey to work. But even better is to drive my car.
The problem with the stat speak articles is that they make assumptions upon assumptions and produce a number that doesn't remotely register as intuitive.
If I'm going to look at rbi opportunities in a stat, let me make the decisions on how important it is to judge it and how it rates(as the reader) you come up with the stat that does what it should do. Instead of trying to put a win component into it, how about focus on the base component. The RBI percentage stat that I want is batters advanced, vs potential bases they could have advanced. Note: that isn't really rbi's so in the rbi world, it would be the runners driven (and how many bases they advanced) vs potential bases advanced.
Example. Man on first you hit a double he scores, that would be 3 bases advanced for the rbi, versus a potential of 7 bases advanced(the three that the man on first could advance and the four you could have advanced with a homerun) That would be the rbi component of the stat. (Actually to be honest, that is somewhat doable with retrosheet and a spreadsheet, it's just time consuming) And of course the another version of the stat would be bases advanced percentage which in this case, it wouldn't care about whether you delivered the rbi or not, but instead rates the total number of bases advance (in the example I gave it would be 5 bases out of a potential 7)
The problem is that the people doing this type of research, get this far and continue going on, where they then plug in the base out states into the equation and make then effectively are comparing it to league average hitter and it really does make it tough to trust the numbers after that.
Reading the first Tom Ruane article I could find, his assumption is to remove walks from the equation, which is absolutely 100% wrong. The complaint by traditionalist is that players who walk frequently are not expanding their strike zone for the team. You need to include walks in the equation or you are not talking the same language. (I can see removing hbp and intentional walks...and unfortunately there is no way to remove unintentional intentional walks)
As I posted recently in another thread:
This also seems a more precise split than using RISP, with a runner on 1B in about 50% of PA with RISP
Because one wants to know who drives in baserunners more consistently, regardless of who drives in more batterrunners on home runs.
Ex. Career stats for selected onetime Cards' 1B since 1974:
That is just a way to penalize homerun hitters. A run driven in, whether he is on third base or at the plate is worth one run regardless. When you start to parse out driving yourself in, you are intentionally making a stat that is weakened as a analytical tool.
Hamilton ranks #1 in percent of runners driven in from 1st base, but #15 (tied) in percent of runners driven in from 2nd base and #48 in percent of runners driven in from 3rd base.
Miguel Cabrera ranks #8 (tied) in percent of runners driven in from 1st base, but #12 (tied) in percent of runners driven in from 2nd base and #14 in percent of runners driven in from 3rd base.
Tables below are data from the 2012 BP report at the link noted in [15] above, sorted by 2012 percentage of runners driven in from each base, preceded by the number of runners driven in from each base:
2012 percentage of runners driven in from 1st base:
2012 percentage of runners driven in from 2nd base:
2012 percentage of runners driven in from 3rd base:
Source: http://www.baseballprospectus.com/sortable/index.php?cid=1253206
That's simply not true.
In fact, your proposal is analytically weak because you advocate including HR in the numeratorsince 1 HR = 1 RBI even with the bases emptybut you seem to ignore including batter PA in the denominator even though 1 PA = 1 (batter)runner on and eligible to be (self)driven in on a home run.
I don't ignore batter pa in the denominator except when the pa results in an intentional walk or a hit by pitch. All Pa's should count in which the player had a large choice in figuring into the result. That includes walks, (choosing not to swing) sacrifices etc.
My example was effectively number of bases advanced divided by potential number of bases advanced...or number of bases advanced towards a run divided by potential number of bases...in either equation I would include the batter as 4 potential bases.
Their formula completely ignores the batter as a potential run, in fact they actively penalize the batter for driving himself in, by removing those from the equation.
Counting HR and all PA raises certain high HR/PA hitters on the list and hurts certain low HR/PA. Isn't that correlated with SLG? Does the stat then add much informational value over SLG?
If, as in [20], you are also interested in who also drives in baserunners excluding batterrunners (esp if you are in the playforonerun situation), it would seem to me that removing HR and all PA tells you something about the RBI "productivity" of lower SLG hitters in with nonbatterrunner baserunners on as compared to high SLG hitters.
Also from BP for 2012 (except ABI% defined below):
White Sox driving in runners from 3B (R3_BI%), 2012:
OBI% is baserunners only driven in
ABI% is all RBI including HR as a percentage of batter PA plus baserunners
6 Alfonso Soriano 29 24.60%
MVP! MVP!
You will notice the %age of runners driven in from 3rd list is not exactly murderer's row ... that's pretty much nothing but making contact.
Johnny Ray drove in more runners because he was AWESOME.
The fact that he was my favorite player as a kid is purely a coincidence.
What species critter be this? I'm usually pretty good with players who start, never heard of this guy.
http://www.baseballreference.com/players/p/pachejo01.shtml
From BR PI event finder:
All of MLB: 19164 Plate Appearances in 2012, 3rd Occupied
Batters:
Pitchers:
You must be Registered and Logged In to post comments.
<< Back to main