Designated HitterOctober 27, 2005
Can Baserunning Be the New Moneyball Approach?
By Jeff Angus

Running the bases has almost always been seen as a side-dish. Even in the Bitgod (Back In The Good Old Days) view, a player who was in the 99+ percentile for baserunning had to have at least one more tool to get to the majors. A fellow like The Panamanian Express was only on the team because the owner, not the manager, insisted. BTW: I think "¿Who is The Panamanian Express?" is the question to the following Jeopardy answer: One of only two players in ML history with a season's worth of games to have created more outs than he had plate appearances.

The knowledge revealed by Sabermetric analysis, combined with the efflorescence of offense since the leagues juiced the ball in 1994, has relentlessly pushed the running game further into the background year-by-year, both for pragmatic reasons and for religious ones.

But this year, there's been a pick-up in interest about baserunning that transcends the easily measurable component of it (stealing). Two successful teams, the Los Angeles Angels of The O.C. and, to a lesser degree, the Chicago White Sox, have declared that part of their success is the ability to squeeze extra bases out of singles and doubles by taking the extra base. After interviewing Mike Scioscia, I wrote a couple of entries at my Management by Baseball weblog, here and here that talk primarily about hitting with runners in scoring position, but some about taking the extra base, as well (explaining how the two are inextricably linked in the Angel system's theory).

¿So what about taking the extra base? What can we learn, what do we know, what can we measure meaningfully?

I have a collection of 2005-through-August numbers from which I'll deliver some findings. The numbers are not as definitive, nor as granular, as the excellent opus Dan Fox of Dan Agonistes' blog has produced. Fox has some great work on his own blog, and a more recent three-part series at The Hardball Times that starts with this one. Fox suggested his numbers didn't match up well with what Scioscia talked about in my interview, and mine don't either. We both think the Angels do some slicing-and-dicing and are analyzing a sub-set of our data.

I don't intend this to usurp or compete with Fox' fine stuff. He has already analyzed what the numbers can deliver in terms of runs and win potential. Consider this more pattern-recognition, looking at the patterns that might provide more insight, a complement to his work, some brain candy delivered to your door to raise questions we should think about answering. I believe the utility of the math is ultimately going to stop short of what we want to know because there are too many environmental variables: individual park factors, outfielders' arms, game situation, base-out status, who the other runners on might be and team strategic theory, to name a few. But just because the math only gets us part of the way there doesn't mean it isn't juicy and valuable. And it doesn't mean we should stop asking questions and looking for answers.

In that spirit, let's set up the workbench with some numbers. These are from the 2005 through the end of August, a fair sample. First, Major League frequencies of opportunities, success and failure in three situations: 1st-to-3rd on a single, 2nd-to-Home on a single, and 1st-to-Home on a double.

Extra Base        1st-3rd        1st-3rd        1st-3rd        1st-3rd
   Opps.            Opps.        Success        % Success    Out tryin'
   12247            6203          1949             31%            60

Runners try for third on a single about a third of the time, and are only caught once for every thirty-two times they try.

                  2nd-Home       2nd-Home        2nd-Home       2nd-Home
                    Opps.        Success         % Success      Out tryin'
                    3846	         2590             67%             129

A lot higher frequency of attempts on 2nd-to-Home (clearly longer throws for the LF and CF), and runners get gunned down at a much higher rate (about half again as frequently), though still not very often.

                  1st-Home       1st-Home        1st-Home       1st-Home
                    Opps.        Success         % Success      Out tryin'
                    1672           763             46%            62

The benefit/cost ratio of already-safe-at-third versus taking a chance on being out at home is dampening attempts. The failure rate goes up sharply again (out about two-thirds more frequently than 2nd-to-home and about two-and-a-half times more frequently than 1st-to-3rd), yet still only 7.5% of the attempts.

DISCUSSION TOPICS: ¿Are outfielders' arms so lackluster, is station-to-station baseball so popular now as a result of higher run production, have teams equipped with analytical systems been able to optimize who goes when?

Ron Fairly suggested to me (and I tend to agree) that the throwing "tool" isn't accentuated much in the minors or bigs. That would make sense, because if you are a one-tool player, it's the least likely to get you a career outside of pitching, and of any two-tool combo you could Lego together, it's the least advantageous second tool, too. So, he asserts, outfielders are more likely to throw to the cut-off man than attempt the assist. I believe there were more good outfield arms displayed during games in the 1965-1993 period, and I'm not a Bitgod, though we're all capable of little Bitgoddities, so it could be selective memory on my part. But if I was a manager, my logical response to less frequent throwing to the target base would be to ratchet up, in context of course, the frequency of attempts until defenses started to respond. ¿Would you do that?

Some Team Totals

Here's the MLB numbers restated on one crowded line, with a mean average team line following:

       1-3   1-3   1-3  1-3 |  2-H   2-H   2-H  2-H |  1-H  1-H  1-H  1-H
       opps  safe  safe out |  opps  safe  safe out |  opps safe safe out   
MLB    6203  1949   31%  60 |  3846  2590  67%  129 |  2198 1010  46% 74
Mean    207    65   31%   2 |   128    86  67%    4 |    73   34  46%  2

Let's run some individual team totals against that Mean team.

        1-3   1-3   1-3  1-3 |  2-H   2-H   2-H  2-H |  1-H  1-H  1-H  1-H
       opps  safe  safe  out |  opps  safe  safe out |  opps safe safe out   
Mean    207    65   31%    2 |   128    86  67%    4 |    73   34  46%  2

LAA 228 75 33% 3 | 153 100 65% 5 | 62 33 53% 2
CWS 200 72 37% 1 | 114 85 75% 7 | 61 32 53% 4
HOU 160 55 34% 1 | 120 69 58% 5 | 77 35 45% 5
SEA 142 39 27% 1 | 106 73 69% 1 | 65 38 58% 2
NYM 188 65 35% 2 | 95 66 69% 0 | 57 24 42% 2

Keep in mind as Dan Fox already pointed out that these numbers are very context-sensitive (park effects, roster abilities, team strategies, coaching decisions). Seattle has a lot fewer opportunities than average, but they weren't experiencing many baserunners-on situations, and the Astros had about 9% fewer at-bats with runners on-base than the NL average. Compared to the ML average, the Angels have more opportunities, convert at about the same rate for higher gross yield and are thrown out at about the same rate. So they're not better quality, they are higher-input with higher-output and the same quality. Intuitively, one would assume that as you drive up quantity, your added increment would be lower-quality opportunities. That is, everyone is already sending Chone Figgins or Charles Gipson against Jeremy Reed or Judi Dench, so incremental chances would likely come with less-skilled runners or more-skilled outfielders. ¿What do you think?

There's a universal Truth, known to many as Angus' Eleventh Law, that whatever asset becomes debunked as overvalued will become undervalued before it finds its homeostatic set point. If that's in operation here, it may mean that there's a little edge in aggressive baserunning "the market" of MLB teams currently undervalues and, therefore, makes it some opportunity for others willing to pursue it. And if opponents aren't used to seeing such naked aggression, until they do, their immune response will be somewhat impaired. It may be that the alterations teams make to the Scamperball approach disadvantage them in minor ways we can't track through isolating baserunning: perhaps pitcher or infielder concentration, perhaps fielder positioning, or other things. ¿What do you think?

Some Individuals' Totals

Here are some individual players' numbers we can chew on. I consolidate all three situations (1-to-3, 2-to-Home and 1-to-Home) because breaking up individuals' segmented opportunities into three very small piles of data means a single additional opportunity or out radically changes the outcome. The ML average was 34% of opportunities converted into safe advances. Here are the baserunners who had the highest number of opportunities:

                      Opps   Safe   Safe    Out
Johnny Damon   Bos     66     38     58%     1
Manny Ramirez  Bos     63     21     33%     1
Bobby Abreu    Phi     63     32     51%     2
Derrek Lee     ChC     61     31     51%     1
Brian Giles    SD      61     28     46%     1
Miguel Tejada  Bal     59     31     53%     1
Derek Jeter    NYY     59     27     46%     2
Alex Rodriguez NYY     59     25     42%     2
Miguel Cabrera Fla     58     20     34%     0
Jason Bay      Pit     58     32     55%     0
Ichiro Suzuki  Sea     58     27     47%     2
Mark Teixeira  Tex     57     30     53%     1

What earns you a place on the list is a (a) high OBP after subtracting homers and, concurrently, (b) players coming up after you who are likely to hit singles and doubles. Very, very context-sensitive, and one has to think that Damon and Ramirez are the poster-kids for this particular Jimmy Fund. There's a decent spread of high- and normal success percentage runners in this table. There is no bell curve distribution for percentage attempted (no surprise...there's almost nothing in nature that manifests as a bell curve outside the minds of the early-20th century researchers who invented the bell curve). The 90th percentile rank for safe% is 60% and the 10th percentile rank for safe is 29%.

Here are the top runners by percentage of opportunities converted into safe advances (minimum 25 opportunities).

                 Opps   Safe   Safe    Out
Figgins   LAA     50     34     68%     3
Miles     Col     25     17     68%     0
Hardy     Mil     25     17     68%     0
Sullivan  Col     28     19     68%     1
Hairston  ChC     31     21     68%     0
A Boone   Cle     43     29     67%     2
Beltran   NYM     45     30     67%     1
Barmes    Col     30     20     67%     0
Crisp     Cle     44     29     66%     0
Podsednik CWS     32     21     66%     0

Not too many catchers or over 40s on this list. Two words that should strike fear into right-fielders and recreational hoops players everywhere: Aaron Boone.

Boone looks as out of place on this list as Ted Nugent would at a Friends meeting. Here are the trailers by percentage of opportunities coverted into safe advances. The local commuter trains (minimum 25 opportunities):

                 Opps   Safe   Safe    Out
Ortiz      Bos     49     6     12%     1
B Molina   LAA     32     4     13%     0
Berkman    Hou     30     4     13%     2
Thome      Phi     33     5     15%     3
T Martinez NYY     25     4     16%     2
Ward       Pit     36     7     19%     2
Zaun       Tor     45     9     20%     0
V Martinez Cle     44     9     20%     1

Too many catchers on this list. What ever happened to speedburners like Choo-Choo Coleman? I'll nominate Jim Thome as the least useful baserunner with his low percentage and extra outs. Who would you nominate?

There are a lot of questions to be answered. I haven't begun to ask them all. I ask you to view this as a first cut at a foundation for discussion. Of things we might find out based on the data we have access to, what are the most important ones we can discover; ergo, what are the next logical questions?

* * * * *

Jeff Angus is a management consultant specializing in knowledge management and change management, and the stats columnist for The Seattle Times during the baseball season. He writes the Management by Baseball weblog. His current book is "Management by Baseball -- A Pocket Reader," and he has a book coming out in May from HarperCollins, called Management by Baseball: The Official Rules for Winning Management in Any Field.

[Additional reader comments and retorts at Baseball Primer.]

Comments

Finaly....The Chicago White Sox win The World Series.

www.2005whitesoxchamps.com

OK, so who is the other player besides The Panamanian Express to create more outs than plate appearances? Without checking, my guess is another sprinter who was hired by Charles Finley -- Herb Washington.

This may sound silly, but I have two sons, both district champion cross country runners that play softball for fun. Everyone goes crazy when the see them run. They are famous in the Seattle Jewish community for their speed on the team. They have never played league baseball because they went to small religious schools that didn't have teams. One is 18 and soon to graduate from high school the other is 20 at the U of W.
Would you have any interest in meeting them.
Do you know of any running scholarships or professional training available.
Thank you!
Chana Uslan
p.s. please keep my comments annonymous

Yes, Phil, Herb Washington is the other player in answer to the question Jeff posed.