the man driving ‘em in deserves the glory he has gotten all along

He deserves some glory, to be sure. He does not deserve the glory he has gotten all along, which is kind of the point.

Correct me if I'm wrong, but I think the main reasons that saberdudes deride the obsession with RBIs are because they're heavily influenced by the rest of a given player's team and are prone to huge fluctuations on a year to year basis, therefore offering little when trying to quantify and/or project the total value of an individual player.

Those reasons are good enough for me to not give the "big RBI man" heaps of praise.

Leadoff hitter hits a double, next guy hits a single but the runner holds at 3rd. Next guy hits a lazy fly to RF scoring the runner.

Somehow the guy who hit the lazy fly gets kudos for the RBI and the other two get largely ignored by the common fan (they dont' pay much attention to R's for some reason)

RBI make very little sense.

I didn't RTFA, but


I'm surprised that this guy doesn't advocate getting rid of batting average and awarding the "batting title" to whomever gets the most hits - regardless of the number of plate appearances.

I think of RBI like I do of wins. They might not be a good indication of a player's ability in a single season but generally, the guys with the most RBI over a significant period of time are the best power hitters.

Leadoff hitter hits a double, next guy hits a single but the runner holds at 3rd. Next guy hits a lazy fly to RF scoring the runner.

Somehow the guy who hit the lazy fly gets kudos for the RBI and the other two get largely ignored by the common fan (they dont' pay much attention to R's for some reason)

RBI make very little sense.

Makes a lot of sense to me. The double and single can be rendered useless without the lazy fly ball.

The old Elias Baseball Analysts had some questionable stuff in them from time to time, but I think they got a bit of an undeserved bad rap, especially considering some of the different ways they presented data.

One of the stats they used to have were the percentage of runs a player drove in from each base (IE, he drove in the guy from third 60% of the time, from 2nd 30%, etc.) What this was neat for was to show just how many more baserunners were on base for a guy like Joe Carter compared to other better sluggers who drove in fewer runs. IIRC, one year a guy drove in 20 fewer runs than Carter, but drove in a higher percentage of runners from each base. He just had fewer guys to drive in. And that's one of the basic problems with the RBI as a raw counting stat: it's like comparing a guy who hit 30 homers in 300 at bats to a guy who hit 35 in 670 and concluding the second guy has more power.

RBIs aren't worthless, they're just more or less worthless once you add inofrmation that is simply more valuable.

I don't get the argument that just because a statistic generally correlates well with quality, that it is useful. You could make up a statistic on the fly that tells you nothing but correlates with quality.

I could invent a stat right now called Pancake Flops, and have the formula be: (3*2B + HR^1.3 - outs)/GIDP. Chances are, the leaders in Pancake Flops will be good players, but that doesn't make the statistic at all useful.

One of the stats they used to have were the percentage of runs a player drove in from each base (IE, he drove in the guy from third 60% of the time, from 2nd 30%, etc.) What this was neat for was to show just how many more baserunners were on base for a guy like Joe Carter compared to other better sluggers who drove in fewer runs. IIRC, one year a guy drove in 20 fewer runs than Carter, but drove in a higher percentage of runners from each base. He just had fewer guys to drive in.

The offensive equivalent of the Danny Jackson-Walt Terrell comparison.

Selected Pancake Flop leaders for 2007:

David Ortiz -4.1
Carlos Lee -7.2
Chipper Jones -7.4
Magglio Ordonez -8.1
Matt Holliday -8.3
Albert Pujols -8.9
AROD -9.8
Jorge Posada -9.9
Vladimir Guerrero -10.7
Ryan Braun -11.3
Brandon Phillips -11.9
Miguel Cabrera -12.1
Paul Konerko -12.3
Nick Markakis -13.0
Aaron Rowand -13.1
Ryan Howard -13.3
Mike Lowell -13.5
Derrek Lee -14.8
Jimmy Rollins -14.9
Derek Jeter -15.2
Todd Helton -15.3
David Wright -15.6
Brad Hawpe -15.6
Adam Dunn -16.3
Garrett Atkins -16.8
Prince Fielder -16.9
Ken Griffey -17.3
Justin Morneau -18.1
Carlos Pena -18.1
Dan Uggla -19.5
Alfonso Soriano -21.2
Jim Thome -22.5
Lance Berkman -22.6
Hanley Ramirez -22.7
Chase Utley -23.2
Mark Teixeira -24.0
Mark Teixeira -24.0
Eric Byrnes -27.0
Carlos Beltran -28.5
Jimmy Rollins -29.4
Alexis Rios -30.4
Pat Burrell -36.0
Brian Roberts -38.1
Adrian Gonzalez -41.5
Adrian Gonzalez -46.0
Ichiro Suzuki -54.1
Chris Young -58.3
Jose Reyes -64.0
Curtis Granderson -86.0
Grady Sizemore -100.9

Grounding into more DP is a positive Pancake Flop!

AROD -9.8

Only fifth in the majors? He even chokes at Pancake Flopping.

This article is exactly right. RBI are historically overrated and anybody that has digested Bill James knows it. That said, I'm tired of seeing them trivialized by some statheads.

Only fifth in the majors? He even chokes at Pancake Flopping.

Otoh, the Sox are going to have to give Ortiz another award for being "The Greatest Pancake Flopper in Sox History."

Barry Bonds had a Pancake Flopping of -11 in his age 42 season. Clear proof of steroids.

I look at RBIs as being a "fun stat". I think they're great in isolation looking at things a player did. I guess I don't see anything wrong with using it with "down ballot MVP votes", which I assume is the votes to fill out a ballot. My only problem is using it project future performance (which the author rightly says is bad) or using it to compare past performance with other players.

But talkin' about Player X getting Y RBIs in Z amount of time? Awesome!

RDF. Posts 10-15 illustrate why I love this site.

Chances are, the leaders in Pancake Flops will be good players, but that doesn't make the statistic at all useful.

Obviously you were wrong. The leaders are all good players because the stat is VERY useful.

But talkin' about Player X getting Y RBIs in Z amount of time? Awesome!

Even better is Player X getting Y RBI's in Z amount of opportunities. But then you probably need to weight them, and then all of a sudden you're cooking with LWTS.

I always thought it was weird that the MSM talked about batting average with RISP, when really talking about the percentage of runners driven in with RISP would have been just as natural and about as easy to calculate.

Even better is Player X getting Y RBI's in Z amount of opportunities. But then you probably need to weight them, and then all of a sudden you're cooking with LWTS.

Bill James talked about Victory Important RBIs. I started playing with them in my head one day and came up with something similar to WPA. Too bad the Mills Brothers were 30odd years ahead of me. SOSH, refresh my memory. What was the Terrell-Jackson comparison?

I always thought it was weird that the MSM talked about batting average with RISP, when really talking about the percentage of runners driven in with RISP would have been just as natural and about as easy to calculate.

I agree. I find it odd that LOB is put in a box score, but doesn't seem to be tracked to the same degree over the season.

The old Elias Baseball Analysts had some questionable stuff in them from time to time, but I think they got a bit of an undeserved bad rap, especially considering some of the different ways they presented data.

One of the stats they used to have were the percentage of runs a player drove in from each base

and I wish I had saved my old copies to confirm this, but what struck me at the time was how little difference there was between players (and even between teams) in percentage of runners driven in from scoring position

the way to score runs is to get a shitload of runners on in the first place

Bill James talked about Victory Important RBIs. I started playing with them in my head one day and came up with something similar to WPA. Too bad the Mills Brothers were 30odd years ahead of me. SOSH, refresh my memory. What was the Terrell-Jackson comparison?

I don't have the Abstract, Jon, so I'll do this from memory. Anyone with access to the 1988 Abstract is free to correct any lapses.

In 1987, Danny Jackson went 9-18 with a 4.02 ERA. Terrell went 17-10 with a 4.05 ERA. The previous season was even more dramatic, as Jackson was 11-12 with a 3.20 ERA to Terrell's 15-12 with a 4.56 ERA. To combat any "knows how to win, or just pitched well enough to lose" type arguments, James looked at each player's won-loss record for each level of run support. In almost every category, Jackson had a better winning percentage than Terrell. In other words, if each pitcher got 4 runs, Jackson likely performed better. It's just that Terrell had so many more opportunities at the high end (6-plus runs or whatever), while Jackson had so many more at the other end (1 or shutout).

What was the Terrell-Jackson comparison?

In 1986-1987, Danny Jackson made 61 starts, Walt Terrell 68. Per Retrosheet, Jackson's per-start breakdown at various levels of run support:

2 or fewer: 39 starts, 238 1/3 IP, 6.11 IP/start, 4.00 ERA, 1.48 RS/9, 4-28-7 W-L-ND
3 to 5 runs: 16 starts, 115 IP, 7.19 IP/start, 2.97 ERA, 4.65 RS/9, 11-2-3 W-L-ND
6 or more runs: 6 starts, 44 IP, 7.33 IP/start, 3.07 ERA, 9.00 RS/9, 5-0-1 W-L-ND

and Terrell's breakdown:

2 or fewer: 22 starts, 125 1/3 IP, 5.70 IP/start, 5.82 ERA, 1.54 RS/9, 3-17-2 W-L-ND
3 to 5 runs: 24 starts, 173 2/3 IP, 7.24 IP/start, 3.89 ERA, 4.60 RS/9, 11-5-8 W-L-ND
6 or more runs: 22 starts, 161 IP, 7.32 IP/start, 3.58 ERA, 10.03 RS/9, 18-0-4 W-L-ND

RS/9 is run support per nine batting innings, calculated while the pitcher was still in the game - thus, a visiting pitcher who pitches six innings is credited with 7 batting innings, since he's still technically "in the game" in the top of the 7th, and with the runs scored by his team through seven innings.

Jackson's low run support, and Terrell's high support, is just astonishing for those two seasons. That's about as extreme a comparison as you can get.

-- MWE