Defense Efficiency Ratio (or DER, for those in the acronym biz) is a pretty simple statistic. It takes all balls that were batted into play (so, no Home Runs) and calculates how many were turned into outs. It's a simple measure but it reflects a lot of complex stuff, such the quality of the fielders, the gloves, the ballpark configurations, how hard the ball was hit, where it was hit, and probably a few more things I haven't thought of.

Which is why I find the first graph on the right, the League Statistics Over Time graph, so fascinating. It's the story of what has happened to batted balls in play since 1901. (Click to embiggen the graph if you'd like) The story it tells is this: there were wild swings in fielding results during the Deadball Era, followed by a steady period of relatively low DER in the Live Ball Era. In the beginning years of World War II, DER took a jump up and didn't look down again for several decades (that gap between the AL and NL in the '60s is interesting, isn't it?). This jump may have been due to improvements in technology, such as better fielding gloves or park conditions, among other things.

But the thing that really catches my eye (even without the arrow) is the downturn in the early '90s. If you pull up the graph yourself and mouse over the data points, you'll see that the percent of batted balls fielded for outs dropped from 70% in 1992 to 68% 1996. That doesn't sound like a lot, but keep in mind that we're talking over 100,000 balls in play each year. It adds up. What's more, this became a long-term change. DER has crept back up but is only now (2024) back up to 70%.

What happened?

Well, we should first acknowledge that the Colorado Rockies, with their cavernous ballpark, joined the majors in 1993. If you look them up in the Team Timeline Explorer you'll see that their DER is typically two percentage points lower than the major league average. But that's how much DER fell across both leagues during that time, and the Rockies were only one team.

This is where the Multi-Stat Timeline comes in handy. In the second graph, you'll spot two statistics: DER and Slugging Percentage. It's a fascinating graph, isn't it? The two stats seem to dance around each other. When Slugging goes up, the percent of fielded balls goes down. And vice versa. Perhaps balls were hit harder, which made them harder to field. Perhaps the old defensive whizzes who were in the lineup just for their gloves went by the wayside as hitting was emphasized more and more. Perhaps the ball was juiced. Whatever the cause, decline in fielding outcomes is another way the game has changed right under our noses.

Try playing with the graphs yourself to see if you can spot any particular reasons for this intriguing phenomenon.

Everyone knows that 1968 was the Year of the Pitcher, right? That was the year teams scored only 3.4 runs a game, the lowest total outside of the Deadball Era. Bob Gibson set the modern record with a phenomenal 1.12 ERA. Denny McLain won 30 games, etc. etc. Well, what if I told you that there was a recent year in which pitchers dominated just as much as they did in 1968; that there was a year in which their strikeouts, walks and home runs allowed were at virtually the same level as the Year of the Pitcher? See if you can spot it in the League Statistics Over Time graph on the right. (You can click on the graph to embiggen it and get a closer view.)

If you picked 2014, give yourself a prize. In 2014, FIP was actually a bit lower (0.49) than it was in 1968 (0.52). When comparing 1968 to 2014...

• - In 1968, pitchers struck out 5.9 batters per game. In 2014, they struck out 7.6.
• - In 1968, pitchers gave up 0.6 home runs per game. In 2014, they gave up 1.0
• - In 1968, pitchers walked 2.8 batters per game. In 2014, they walked 3.0.

You may ask, why didn't anyone notice this at the time? Well, you see, the FIP statistic that we use here doesn't include a constant to make it equal to the league's ERA. We use "pure" FIP, which only considers the three main components of FIP. So the more recent Year of the Pitcher was overlooked because they just looked at regular old ERA. And ERA masked what was going on. In fact, the major leagues scored 4.1 runs per game in 2014, far above the 3.4 registered in 1968. And why was that?

Have a gander at the second graph, the Multi-Stat Timeline. The blue line is FIP, the orange line is Runs Per Game and the black dotted line with the triangles is DER. As we've already discussed, fewer balls have been turned into outs in recent years and in retrospect you could have called 1968 "the year of the fielder." 71.3% of batted balls in play were turned into outs in 1968, the highest total in baseball history. In 2014, that figure dropped to 68.8%. As we mentioned in the previous article, a lot of factors go into DER, but FIP is a purer stat. It measures the things that pitchers control most. And pitchers in 2014 were as effective as any year since the Deadball Era.

Here is a graph of the number of triples hit per game on the X axis and each team's Slugging Percentage on the Y axis. Each dot represents one team in one year-there are 2,692 dots in all. And, as you can see, this graph is a mess. It's one big blob of dots with a slightly downward trend, meaning that the more triples that are hit, the lower the Slugging Percentage tends to be.

Which makes no sense, right? If you hit more triples, your Slugging Percentage should go up. After all, each triple is worth three bases. But what we have here is a countervailing trend over the years. As baseball time has passed, batters have learned to hit more home runs and fewer triples. As a result, Slugging Percentage has gone up while triples have gone down. It's not that triples don't increase Slugging Percentage, it's that the wider trend has wiped out the natural logic of it all.

If you look at this graph for each era instead of over all baseball history (go ahead and try it), you'll find correlations that make more sense. Here is a list of the correlation between triples and Slugging Percentage in each era:

• -Dead Ball Era: 0.694
• -Live Ball Era: 0.360
• -World War II: 0.516
• -Integration: 0.123
• -Expansion: 0.284
• -Free Agency: 0.126
• -Steroid Era: -0.004
• -Modern Era: 0.036

See that? Through most of baseball history, triples have been positively correlated with Slugging Percentage. It's only in the last few eras, when so few triples have been hit at all, that there is almost no correlation between triples and SLG. And there was never an era in which triples had a negative impact on Slugging (that negative correlation in the Steroid Era is so weak that it can be ignored). So when you investigate the correlation between statistics, be sure to weed out the long-term trends from the short-term effects.