Historical hitting by position

Maybe the most important tenet in sabermetrics is that numbers don’t mean anything without context. Thirty home runs can be a lot or merely average depending on whether you play for the Mariners or the Rockies. A 3.00 ERA makes a Cy Young winner today, but would get you sent back to the minor leagues a century ago. An .800 OPS means vastly different things for a shortstop and a first baseman.

We’re going to focus on that last distinction today. Intuitively, we know that catchers and first basemen should not be evaluated on the same standard. However, it is generally difficult to realize just how large the disparity between the average catcher and the average first baseman is.

For example, Brian McCann, who has posted a .769 OPS this season, has been criticized for his struggles all season, while Kevin Youkilis, who has an .879 OPS, has garnered praise for his breakout season. But if you look at the numbers, Youkilis is having an average season at first base and McCann is having an average year at catcher.

Undeniably, some of that sentiment steams from the fact that McCann was significantly better last season while Youkilis was worse, but if you asked any informed fan (even yourself, if you were being honest) who was having the better season, I don’t think you would find one person who wouldn’t say Youkilis. The right answer, of course, is neither.

The difference in hitting between a catcher and a first baseman is huge—two wins a season. It is equivalent to the difference between Carlos Beltran and Ray Durham. I didn’t even realize Durham was still playing these days.

So today, we’re going to take a look at how offense has evolved at each position over the years. We’ll see how second and third base switched their positions on the defensive spectrum, how shortstop has become a significantly more offensive position in the 21st century, and how the difference in production between positions has decreased in recent years, among many other things. And hopefully, by the end of this article, we will all have gained an appreciation for how a player’s position can make him into a great hitter or a merely average one.

Let’s start by taking a look at how players at each position performed last season, denoted in runs versus average per 162 games:

Catchers: -8
First basemen: +13
Second basemen: -5
Third basemen: +6
Shortstops: -5
Left fielders: +8
Center fielders: +2
Right fielders: +7
Designated hitters: +13

If we line the positions up with the strongest-hitting on the left and the weakest on the right, our results will perfectly match Bill James’ defensive spectrum:

DH, 1B, LF, RF, 3B, CF, 2B, SS, C

This is a pretty important realization because what we’re seeing here is that offensive and defensive ability are, on a whole, correlated perfectly negatively. The tougher a player’s position, the less he is required to hit and vice-versa.

But those requirements are ever-changing. We can find that by looking at a graph of offensive production by shortstops and catchers since 1871. All numbers in this article are runs above average per 162 games, and in this graph, shortstops are denoted by the red line and catchers by the blue. I’ve used a three-year moving average in all these graphs to slightly smooth out the lines:

Let’s work backwards here. We see that the average shortstop is a significantly better hitter today than he was 30 years ago; in 1977, an average shortstop was 20 runs below average. There’s actually a huge drop in shortstop offense in the early ’70s, though I’m not quite sure why. Perhaps as strikeouts went down, teams began to view shortstop defense as more important.

You can also see that shortstops were actually roughly average through much of the deadball era. Again, I can only offer a conjecture as to why, but I think Bill James actually explained this a long time ago. In the New Historical Baseball Abstract, James writes that one mark of a poor quality of play is that players to the right of the defensive spectrum are the best hitters.

For example, on many high school baseball teams, the best hitter is also the team’s ace. In the same way, great athletes like Honus Wagner were often both great hitters and great fielders. While defense at shortstop was incredibly important during the deadball era, fielding ability was better correlated with hitting ability than it has been since.

While shortstop hitting has improved over the years, catchers these days are actually worse at the plate than they were a decade ago. I think, with catchers, we can identify roughly (very roughly) three different eras in terms of hitting: 1883-1913, 1914-1986, and 1987-present.

In the early days of baseball, catchers were absolutely awful hitters—on average, -14 runs per 162 games. That’s to be expected, as bunts were very common during these years and so a catcher’s defense was very important. Also, as pitchers began to throw overhand and batter-friendly rules disappeared from the game, catching pitches itself became more difficult and thus restricted the position to relatively good fielders.

As more players became accustomed to catching and bunting became less prevalent, catcher offense slowly improved. The average catcher between 1914 and 1986 was -5 runs per 162 games—still below-average but not that bad.

Since then, catcher hitting has once more fallen, though the reasons for that are far less clear. In the past 20 years, the average catcher has been -9 runs below average. Any theories as to why are much appreciated.

Next, let’s take a quick look at second and third basemen. Much has already been written about hitting at these two positions, so I don’t have much to add. Here, the red line denotes third basemen and the blue line, second basemen:

The graph is quite messy prior to the mid-1930s, when third basemen overtake second basemen as the better hitters for good. Again, the “why” here has been provided by Bill James: Baseball evolved in the 1920s and ’30s, with a greater emphasis placed on the double plays.

As double plays increased, so did the importance of a good fielder at second. Whereas earlier many good fielders had gone to third instead of second, now, fielders who could turn the double play were asked to play second.

What’s interesting, though, is that the gap between the two positions has been closing for three decades now. Perhaps infield defense has become less important with the proliferation of strikeouts and fly balls and better hitters are stationed at second base these days.

Let’s look at the only infield position we have yet to examine: first base, and take a look at designated hitters as well:

First basemen have been consistently very good hitters, but just how good they have been has varied. In the mid-1890s, they were actually close to average, while in the 1930s, the average first baseman was 20 runs better than the average hitter (thank you, Jimmie Foxx and Lou Gehrig). But since World War II, the play at first base has stabilized at about 10-15 runs above average. That’s a huge difference, about 35-55 points of OPS.

Interestingly enough, designated hitters have consistently ranked below first basemen in terms of overall hitting, though the gap these days is very small. One possible reason for that gap is that players hit worse at DH than they do at first base, as the authors of The Book: Playing the Percentages in Baseball have found. Another is that good hitters generally become designated hitters only as they get older and completely immobile in the field, meaning that the best hitters play first base.

Anyway, there isn’t much to say about first basemen and designated hitters. What’s most important is that we realize just how good the average player at these positions is and mentally adjust for that when evaluating these plodding sluggers. Let’s move on to outfielders. This graph is going to be a bit messy; the blue line represents left fielders, the red line denotes center fielders, and the green line depicts right fielders:

You can see that in the post-deadball era, center fielders have hit significantly worse than the corner outfield slots, likely because of a large increase in fly balls in the current era, which center fielders could leverage more than left or right fielders.

Despite the need for stronger arms in right field, the two corner outfield slots have hit about the same over the past century; left fielders do possess a strangely large advantage around 1900. We also see that corner outfielders are almost as good with the bat as first basemen. A guy like Ichiro is worth about a win more in center field than he was in right.

One interesting thing about all these graphs is that it seems that all these advantages have been decreasing in recent years. A look at the spread in positional hitting confirms that fact:

This is a graph that cannot be easily explained. There is a confluence of factors that have been working in both directions, though the decrease in spread has been pretty constant, if extraordinarily slow and fluctuating, since the 60s. That suggests that modern day positional adjustments are not as important as they once were.

But without accounting for a player’s position, we cannot really evaluate his performance with very much certainty.