Report cards (Part 2)

Last time, I put up the framework for several studies based on giving letter grades to team fielding, pitching, and batting components. Now it’s time to take a broad look at one of those time-honored, rarely tested adages: Do pitching and defense win championships?

In distill of the night

There are a couple of questions wrapped up in the adage. It’s relevant to ask whether they win ballgames in the first place, whether one can win without the other, and whether their winning ballgames is in different proportion to their winning championships.

There are 320 playoff teams and 99 world championships through 2005, where my edition of the ESPN Baseball Encyclopedia cuts off. The following chart shows the 320 teams sorted by frequency of grade per component (top to bottom is A to F):

image

So right there we can see a couple pertinent items. While a variety of skill sets can make the playoffs, if your team has a D or F it’s going to be awfully hard to get there. If your available talent forces you to have a weakness, make it hitting, sure; but that’s not the same as saying pitching and defense are the important things.

Now take a look at the frequency of world championship teams by grade:

image

Although the graph by itself says that fielding is the most indicative of championships (after all, 64 A fielders produced 31 winners), that’s misleading due to the slight uptick in A fielders from the days before multiple-tiered playoff systems, due to a game that used to revolve more around its fielders.

The D/F playoff teams are remarkably absent from the world champions, save for offense. It’s probably those weird teams that have propped up the adage far more than is deserved. Examples of champions with weak offenses include the 1906 White Sox (AAD), the 1969 Mets (BAF; the ’73 Mets were also BAF, and they’re the only two F-offense playoff teams), the 1995 Braves (BAD), and the 2005 White Sox (BAD, although their win is muted by the Angels and Astros also profiling as BAD).

What’s probably more accurate to say, then, is that pitching and defense occasionally enable weak offenses to win championships. But even that doesn’t paint a very realistic picture. Don’t even think about a championship if you have D or F pitching. The lone exception is the 1913 Athletics, whose AFA profile is the only F pitching to make the playoffs. The D pitching came closest to winning a championship with the 1982 Brewers (CDA), but the rest of the teams mostly are bad modern entries, like the 1997 Mariners or the 2005 Padres. Realistically, you’re going to need A or B pitching with at least C fielding.

But while D offenses sometimes win, the A offenses are most associated with winning, even more than A pitching. 54 A-hitting champions out of 149 teams is 36 percent odds, better than the 32 percent of A pitching (53 of 165). The difference is slight, but in testing our initial adage it’s relevant; the mashers don’t get into the playoffs as often as the aces, but they’re at least equals once they’re there. The 1932 Yankees (CCA), 1957 Braves (BCA), and 1976 Reds (ACA) are fine examples of teams winning with crushing offenses and little regard to who’s on the mound.

Disjunctivitis

Of course, while we’ve answered some of the question, we haven’t answered whether pitching and defense are linked, e.g. we haven’t seen teams with only one or the other. We’ve seen that pitching and defense can compensate for bad offenses (1995 Braves and friends), but can pitching compensate for bad defense?

As it turns out, not really. Four teams have made the playoffs with F fielding and A pitching: the 1912 Giants (FAA), 1965 Twins (FAA), and 1983/85 Dodgers (FAC/FAB). The 1908 Tigers and 1971 Giants were both FCAs; neither fared well in October. Occasionally the reverse will happen and a team that’s only good with the glove will win—the 1987 Twins (ACD) and the 2003 Marlins (ACB) lead the pack there—but that’s not a model you want to emulate, especially the former.

It wasn’t until 1921 that a team without A fielding won a World Series. Since then, fielding has decreased in importance, but that’s not to say a sloppy team will
win; like pitching, fielding has a minimum level of competence for any championship team. That said, once the minimum’s in place, it’s anybody’s game, and if you can meet those minima while having a team of mashers, you may be in the best configuration to win. If I had my druthers (and let’s face it, who doesn’t want their druthers?), I’d take my chances with a random Joe McCarthy or Sparky Anderson team over the ’95 Braves or ’05 White Sox, and I think you would too.

Conclusion

So while it’s unproven that pitching and defense win championships—indeed, the best link is with offense—bad pitching and defense can sabotage championships in a way that bad hitting doesn’t. But there’s far more potential to rack up wins with an offense, and if your pitching and defense are serviceable, then the offense is the key to success. You can pull out miracle seasons with bad offense, but why count on that when you could just hit your way to glory?

References & Resources
Same as last time, with a check on the connotation and etymology of druthers at the very end. It turns out the word is only as old as the National Association, being condensed from a corruption of “I’d rather” (namely, “I’d ruther”) that somehow got turned into a noun.


5 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
MikeS
14 years ago

I think your analysis is potentially flawed.  Based on your previous post, there is much less spread in talent for defense than for Pitching or hitting.  Teh defference between A and F for fielding can be as little as 3 wins, but for pitching and hitting is at least 12 wins.  This will naturally skew the results towards hitting and pitching.  A team that has +6 fielding wins is no better than a team with +1.5 fielding wins in this analysis.

I believe this to be a common flaw in many SABRmetric analyses.  Defense is the hardest thing to measure and so it is downplayed.  When I make this claim, somebody quotes UZR or some other similarly poor defensive metric to show that there really is not much spread between the best and worst defenders which is just circular reasoning.

My big question which I have never had satisfactorily answered is simple:  why isn’t the spread in defensive talent the same as offensive or pitching talent?  It would be interesting to assume that the talent and value difference between the best and worst defender is the same as the best and worst hitter and see what becomes of defensive statistics then.

Brandon Isleib
14 years ago

There’s never been a 6-win fielding team, nor has there been a team at minus 3 wins either.  By my sorting there are 208 A fielders, 464 Bs, 760 Cs, 406 Ds, and 214 Fs.  The pitching and hitting distributions are flatter than the fielders; in other words, there aren’t too many other places I could have assigned cutoffs and gotten even numbers.

The results are a little bit skewed toward fielding insofar as it has less impact on overall wins but is counted equally into GPA.  That impact will show up more in future articles rather than this one, but it’s there all the same.

The main bugaboo in answering your question is separating defense from pitching.  The parts that we know for certain to attribute to the defense are limited at this point.  Still, like pitching, there’s clearly some minimum of defense that’s necessary to be a true contender; you have to be at least league-average to succeed.

Ken
14 years ago

MikeS – The spread in defensive talent is not the issue – it is the spread in defensive value that matters. Defense has less of an effect on the game simply because there are many, many plays where good defense or bad defense are irrelevant – obviously strikeouts, walks and HR, but even most balls-in-play are unaffected by defensive ability.

MikeS
14 years ago

@Brandon Isleib, @Ken.

I’m not saying you guys aren’t right.  I’m just saying that it has never been adequately proven to me that defense is not important.  Defense may be unimportant, but it is possible that the reason it is thought to be unimportant is because we do not have good ways to measure it.  When Brandon says “there has never been a 6 win fielding team” what he is really saying is that by his model, it has never happened.  If the model is faulty, the results are unreliable.  “Wins” as a statistic is all modeling.  3 hits in 10 at bats is a .300 avereage.  If one hit is a homer it’s a .600 slugging percentage.  With no walks it’s a .900 OPS but deciding how many wins that equals is only modeling and models may not be perfect.  Everybody agrees that fielding is the hardest thing to measure and so it follows that it is the hardest thing to design models for.  All I’m saying is that just because you can’t measure something accurately does not mean that thing is less important than the things that are easy to model.  All it means is that you don’t truly know how important it is.  Yet the SABR community often ignores that.  The answer is better fielding metrics.  Easy to ask for, hard to obtain.  As in all sciences, if we do not question our methods and take everything as correct, we will never improve.

Ian
14 years ago

Might the ESPN Baseball Encyclopedia’s fielding ratings be called into question? It gives the 2003 Marlins an A in fielding, when their team UZR, Total Zone and DER were all negative. They did, however, lead the league in fielding percentage. Seems like it might be flawed.