Who Wins, Part 3—Let's Get Small: Does Small Ball Help Teams Win?
by Mike Carminati
Chris Beatty (Denver): Joe, I hope this isn't too sensitive, but you seem to be disgruntled about the way the game is played today. I notice during your broadcasts you refer often to the way the game used to be and seem let down by modern "changes," such as the overattention paid to the almighty home run, the lack of teams playing "small ball," etc. You compare the present players to the old-timers in an almost distainful [sic] way, as if the present-day players contribute less in some way to the ever evolving world of baseball (with few of your personal favorites excepted, such as Bonds, Griffey, etc.). Can you please respond to this and straighen me out if i'm "off-base?" Thanks a bunch.
Joe Morgan: The game is different today, Chris. IF baseball would have started in 1920 and only cared about the home run. It wouldn't be the great sport it is today. One of the reasons the fans love the game, is because they like to think along with the manager, play the manager, second guess the manager. When you just stand the big guys in the batter's box there is no thinking along with the manager. It's just a home run contest.
I've said it before. I will say it again -- Willie Mays is the greatest player ever. I enjoy a lot of the players today. But the game is different. IF this is the way you like it, great. But in my opinion, there is much more to the game than hitting the ball out of the park.
Next in our ongoing quest for what winning teams do well, I would like to sift through the fossil record and evaluate the case for "Small Ball".
That's sound great, but it's not as straightforward as it sounds. For example, how can you measure how well a team plays hit-and-run? How about moving a runner from first to second by hitting behind him? And what the heck is small ball anyway?
Baseball statistics are just not kept for such things. This makes studying small ball a rather abstract and subjective evaluation. There is some evidence that remains in the statistical record: sacrifice bunts (but not bunt attempts), sacrifice flies, stolen bases and times caught stealing. And how well teams move runners along and play hit-and-run can be evaluated by the number of times they ground into a double play.
Of course, there are some things that aren't measured very well: how often a runner goes from first to third on a single, when a batter fails to move a runner along and instead forces him out, etc. And given that the term "small ball" is a catchall that means different things to different people, how do we even define our weigh our terms.
Well, given the rough terrain, I still decided to forge ahead. First, I defined a new, derived stat that I'll call the "small ball" factor. It will consist of total bunts, sac flies, and stolen bases minus GIDP and caught stealing. The stat is made more problematic because baseball hasn't consistently recorded all of these stats until relatively recently. I'll agree this isn't a perfect solution but it does measure many of the components usually discussed when the topic of "small ball" is mentioned.
Next, we will take this new stat normalized by total plate appearance for a given team and compare it against the team's winning percentage. Does doing these things well spell victory? Let's see.
Here are the correlation coefficients for the small ball percentage for all time and per decade:
That's not very encouraging. There appears to be very little correlation between playing small ball well and winning.
OK, maybe we're shortchanging our new stat by lumping all years together. Maybe we need to adjust for era. Next, we'll adjust the small ball stat by the major-league average for the given year. Here are the results:
That's not much better.
However, it could be that small ball's advantages are not apparent when one looks at a team's entire record. Perhaps small ball's affect only shows up in very circumscribed scenarios. To quote Mel Bernstein in Scarface, "There's an answer to that, too, Tony."
Let's take our adjusted small ball stat and compare it against the team records in various situations that Baseball-Reference.com has conveniently summarized for us. Those scenarios are one-run ballgames in general, low-scoring games (five or fewer total runs), one-run games with low overall scores, and "save" situations (won by three or fewer runs) with low overall scores. I also calculated the expected winning percentage for the above situations based on runs for and against run through the Pythagorean formula and compared those with the small ball stat.
Here are the results:
One-Run Exp Wn%
1-R LS Exp Wn%
Sv LS Exp Wn%
Low Scoring Exp Wn%
Those situations correlate worse to playing small ball than overall team record do. This probably is due to the variability in the smaller sample size, but it sure does not support small ball's correlation to winning in those situations.
So what does this tell us about small ball? I see no correlation between successfully employing strategies generally considered to be part of small ball and winning ballgames. That does not mean that bunting in the ninth inning of a tie ballgame is a bad thing. In-game situations may call for certain strategies. However, excessive use of these strategies apparently does not guarantee success.
Now, maybe my definition for small ball and the small ball factor is flawed. If you can devise a better definition, I’ll run the numbers. Maybe the statistical record is too inherently flawed to support any real conclusions.
However, I think there is some solid evidence that we baseball fans have been sold a bill of goods for years where "small ball" is concerned. And it's time to use what evidence we have to defend ourselves.