Analysts have made tremendous progress in the effort to quantify individual defensive performance. Since there’s often noise in the data, a gap still exists between the accuracy of advanced defensive and offensive metrics. As technology improves, though, the aforementioned gap will continue to shrink.
Due to the progress made, defense became perhaps the most noteworthy marketplace inefficiency since on-base percentage. The “defense is the new Moneyball” mantra was a common theme in many articles written about the industry change this past offseason (a Google search with those five words generates over 48,000 results). Several teams opted to open up spots for players who are known more for their glove work than offensive production; one example is the Texas Rangers‘ decision to name rookie Elvis Andrus the starter at shortstop, despite his inexperience and offensive limitations.
Eventually, the market will correct itself (if it hasn’t already), but one economical way that franchises have tried to add a few games to the win column recently has been through improving team defense.
Precisely identifying the number of runs a strong defensive squad will eliminate for a team is still nearly impossible, but organizations are getting much better at coming up with accurate estimates. Evaluating team defense, though, has always been a bit easier than judging individual fielders. Bill James created one of the better metrics to judge team defense decades ago. In his 1978 Baseball Abstract, James introduced Defensive Efficiency Ratio (DER), calling it “simply the best team defensive statistic there is.”
Unlike most numbers featured on the back of a baseball card, DER has withstood the test of time with analysts. DER is fairly basic and doesn’t require a degree in mathematics to understand; it simply shows the rate at which batted balls hit into play (excluding home runs) are turned into outs. The frequency at which a defense can convert balls into outs, and not the numbers of errors made, is an effective way to gauge the actual talent level of a team defense; fielding percentage cannot account for balls that players can’t reach. In other words, DER is the inverse of batting average on balls in play (BABIP) for hitters but on a team level.
DER has certain flaws, of course. It doesn’t account for two outside influences, park and pitching factors, which skew the total. Much like hitting or pitching performance, defensive output will either be enhanced or suppressed by a player’s home ballpark. Coors Field, for example, has historically been a difficult place for teams to play defense. Due to a spacious outfield, the Colorado Rockies‘ DER is impacted by their home field.
Pitching staffs that induce above-average fly ball or ground ball percentages impact the rate of outs produced on balls hit into play. Much of this depends upon the correlation of the pitching staff to the fielders supporting it; for instance, a team with three ball hawks effectively patrolling the outfield gaps will produce a higher DER if its pitching staff predominantly features fly ball pitchers, and vice versa.
Former Baseball Prospectus writer James Click successfully improved upon DER by removing park and pitching factors in a series of articles back in 2003. Click developed Park Adjusted Defensive Efficiency (PADE) and Pitching Independent Defensive Efficiency (PIDE), eliminating forces beyond a defense’s control. A main take-away from his findings is understanding the importance of correctly assessing the make-up of a team’s pitching staff and home environment before making any major roster changes for defensive purposes.
Click, currently a baseball operations coordinator with the Tampa Bay Rays, saw the theory put into practice in 2008. Tampa Bay pulled off one of the most impressive defensive turnarounds ever, going from worst to first in DER.
The then-Devil Rays were atrocious in the field in 2007, posting a .656 DER; for those scoring at home, that marked the lowest total since data became available back in 1954. Brendan Harris provided some offensive pop at shortstop but was miscast at the position. Harris split innings there with Josh Wilson and Ben Zobrist, which had devastating effects on a historically bad defensive team.
Elsewhere in the infield, Japanese import Akinori Iwamura didn’t live up to his Gold Glove reputation (he won six in Japan), providing only average defense at third base. The duo of B.J. Upton and Ty Wigginton at the keystone made for some painful moments as well. Upton moved around the diamond before eventually settling in center field. Carl Crawford again was one of the premier defensive left fielders in baseball. Teammate Delmon Young, however, graded out statistically as one of the worst-fielding outfielders in the league. Unfortunately for Tampa Bay, injuries created a situation in which Young was forced to log a considerable amount of innings in center.
The Rays made it a priority to improve at run prevention in the ensuing offseason, leading to the blockbuster six-player trade with the Minnesota Twins. Matt Garza was the key return chip, but the addition of a solid defensive shortstop in Jason Bartlett played a major factor in Tampa Bay’s decision to pull the trigger.
Replacing the Harris/Wilson/Zobrist trio with a solid defender in Bartlett was indeed a huge addition. Despite the worst statistical season of his career at shortstop, due in part to a knee injury that limited his range, the newcomer helped solidify the middle of the infield.
The Rays’ infield play also vastly improved at other spots. One major factor was the arrival of top prospect Evan Longoria, who quickly established himself as an elite defensive third baseman. Longoria represented a clear upgrade over Iwamura, who became more of an asset when he moved to second base. Carlos Pena, meanwhile, continued to play an above-average first base.
The outfield defense, arguably the best in the game, was also an area of strength. Upton flourished in his new home, combining with Crawford to convert more outs on balls hit into the gaps. Early-season trade acquisition Gabe Gross and rookie Fernando Perez also provided plus defense.
Thanks to the improved defensive alignment, the Rays amazingly posted a major-league-leading .710 DER during the ’08 regular season. The 54-point jump from ’07 was the largest one-year improvement in history. Simply put one out of 20 balls that would have been hits in ’07 turned into outs.
Bill James once wrote that “a great deal of what is perceived as being pitching is in fact defense.” The Rays proved that, as many of the same pitchers from ’07 suddenly watched their BABIPs and ERAs drop considerably; returning pitchers produced a 42-point drop in BABIP. Although the team Fielding Independent Pitching also went down 0.52 points (indicating superior pitching and/or luck on home runs and fly balls; 11.1 HR/FB% in ’07 dropped to 9.3 in ’08), the staff ERA fell 1.71 points as its ERA+ increased from 82 to 116.
After watching the Rays (and Rockies before them) improve in the standings with their enhanced defensive play, several teams took a similar approach this offseason. The sample size is small, but one team out of the group (the Rangers being another great example) that has made a notable improvement in DER is the Cincinnati Reds.
According to Sam Grossman, Manager of Baseball Research and Analysis for the Reds, improving team defense was a priority. “We wanted our defense to turn more balls into outs,” Grossman said in a phone call. “Defensive efficiency is a relatively simple stat but essentially measures exactly that. Our improvement on this front shows that we made a concerted effort to improve on defense, and we certainly hope it continues.”
The Reds finished 29th out of 30 teams with .673 DER in ’08. Boosted by the addition of free agent center fielder Willy Taveras and exceptional right field play of Jay Bruce, Cincinnati ranks sixth in the majors with a .709 DER through Tuesday, May 19.
No longer forced to hide Adam Dunn and Ken Griffey Jr. on the corners, the club now features four players who could play center field if needed; Chris Dickerson and Jerry Hairston are also capable of providing above-average outfield defense. For a team that plays in a small ballpark with a pitching staff consisting of fly ball pitchers, upgrading the outfield defense was crucial.
Some teams have taken the concept to an extreme, focusing too much on run prevention at the expense of run scoring; the Oakland Athletics, the first team to exploit the perceived inefficiency, are a good recent example. Every front office must balance how many runs allowed and scored they expect their team to produce, of course, with trade offs necessary on each side. Still, upgrading team defense can go a long way.
Even in its simplest form, DER does a nice job of determining which teams have strong defenses. Although the tool has been accessible for over 30 years, front offices have only recently exploited this relatively cheap way of increasing win totals. It’s unlikely the recent coverage given to team defense will trigger a long-term change in the way rosters are constructed. With the latest examples of teams bettering their records by improving in DER and advancements in quantifying individual fielding, though, the trend should continue as long as upgrading defensively remains relatively cheap compared to hitting and pitching.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
I didn't find the discussion of the Reds as plausible since Corey Patterson was the main defensive CF for the Reds in 2008 and all lack of hitting talent aside, is considered a good fielder... so I'm not sure if Taveras was actually an upgrade. It might just be a sample-size thing, or the pitching staff allowing more groundballs/less home runs or something...
Thanks for the comment. As far as individual defensive metrics, you’re right, I did not touch on them as well as I could have. Since my concept/stat was more team defense, I tried to stay away, but the individual metrics could have done a better job of showing what the Rays were able to make such an impressive defensive turnaround. I wanted to put some of this type of info in the piece, but since it was a basics article I did not want to introduce Ultimate Zone Rating or another stat, since I would have had to explain that as well.
Also, if I could submit the piece again, I would focus more on some of the other limitations the stat has. DER is a great team defensive statistic (and I trust it more right now than individual ones), but it cannot distinguish on its own which players are most responsible for the percentage (good or bad).
That is where the advanced individual metrics (UZR, +-, team’s internal metrics) and scouting come into play. DER and its many forms can show a team is slacking, but not necessarily why or more importantly where. And as technology helps improve the advanced data, it will get easier for teams to fully assess (in addition to understanding the make up of the factors discussed in the piece) which positions they can afford to give a lineup spot to a plus fielder even if it means costing a team some runs.
As for the Rays, they obviously felt—with their own internal metrics or the ones available for free and scouts—that shortstop was an area that improving would go along way. With defense, I try to rely on most of the advanced metrics and info available, since there can be some inconsistencies across the board and sample size can play even more of a role than with hitting stats. But for now, I will go with UZR.
Iwamura was not terrible at third base for the Rays in 2007, but his defense there was interesting at times. He produced a -0.1 UZR at the position, costing the Rays a 0.4 fielding runs according to FanGraphs. Most of the other advanced metrics and some of the scouting reports agreed.
As Christina touches on, the Rays did not move him necessarily because they wanted to improve in DER. They moved him to make way for Evan Longoria. He made a nice transition to second base, looking like a season vet around the bag. He produced a 1.6 UZR at the keystone, where his bat played much better; a .323 wOBA is more acceptable up the middle.
Plus, the switch killed two birds in one stone. Longoria was great at third base defensively. He is just a stud, really, and his arrival helped the club improve in two areas on both sides of the ball. Iwamura was also an upgrade over the previous collection of players who took up innings at the position, from Upton and Wiggington to Harris, Jorge Velandia, Wilson and even Jorge Cantu.
As for Young, he was a nightmare in the 242.2 innings he played in center field when Rocco Baldelli went down and the Rays were still using Upton like Chone Figgins. And overall he was a -6.8 defender; he got even worse with the Twins. His big arm drew some wows, but he was a weak defender who frustrated the club with his personality issues and refusal to make adjustments as a hitter. Gross and Perez spent most of the ’08 season in right, and were much better than Young with the gloves.
Overall the outfield was exceptional, according to scouts and the stats. Upton ranked third in the majors among center fielders with a 10.3 UZR. Crawford, a Fielding Bible winner, produced an insane 25.6 UZR, best among left fielders and second in the league at any position.
For the Reds, there are other factors at play other than the addition of Taveras. Dunn and Griffey Jr. were not good on the corners. After Dusty Baker finally realized that Corey Patterson and his sub-60 OPS+ were not good in the leadoff spot (or anywhere), Bruce then spent the majority of innings in center field; Patterson, by the way, was only a tick above league average in UZR the past two years. Bruce was good there as his replacement but not great. And he is much better on a corner, where he becomes a plus. Also, Dickerson and Hairston are obviously better than Dunn.
The sample is still small, but Taveras has been good in UZR after two years with the Rockies where he was in the red. The sample size is still small, but it seems as if the Reds thought he was better than what the UZR figures would suggest.
Shortstop is another position Cincy has improved so far. Paul Janish, Hairston and especially Jeff Keppinger (-13.8) were not great there according to UZR, and the return of injury-prone Alex Gonzalez has helped. Phillips and Votto are also excellent on the right side. But overall that defense is much better. I don’t think they will be this good all year, but they will be improved.
Thank you for providing such additional detail in your comment. From your initial entry, I've seen you go into detail. In this article, you tried to cram a number of vague generalizations about player's individual defense (because individual defense is a part of team defense). The problem is, if you couldn't explain DER while staying within the word limit and without glossing over individual defense, then it would've been a better idea to pick a different topic.
The second problem is that, even with the extended comment, I didn't really learn that much new. I'll admit I don't know players exact UZR numbers and the like, but I've read enough online to know why the Rays thought they were improving their defense.
As far as the Reds go, your comment asserts that the Reds couldn't have been sure about Taveras since he wasn't playing good defense in Coors. Taveras's FRAR numbers for two years with the Rockies are +11 and +8 and Patterson's numbers were +14 and +10. It would seem the Reds took a slight hit on defense in hope of an offensive gain.
It seems the Reds improvement has been mostly due to fortune and circumstance and not from any methodical moves the Reds made. The Reds had bad-fielding shortstops and are doing better defensively because Alex Gonzalez got healthy and he gets paid more. Promoting a rookie like Votto or Bruce appears more like they promoted good prospects. I just don't see proof of any sort of Reds master plan that "According to Sam Grossman, Manager of Baseball Research and Analysis for the Reds, improving team defense was a priority".
Again, Patterson would have been out of the picture, anyway. Following his release, Bruce was the center fielder. Adding Taveras let him move to a corner.
It was about more than just one player with the Reds, though. Dunn and Griffey Jr. spent much of the past few years playing below-average defense in the outfield; the Reds were willing to live with that fact due to each player’s offensive output. It seems as if the club’s new approach is to go after more well-rounded players, but the outfield defense right now is obviously much better.
2008:
LF: Dunn
CF: Patterson
RF: Griffey Jr.
2009:
LF: Dickerson
CF: Taveras
RF: Bruce
Dunn and Griffey combined for a -42.4 UZR/150 during their time with the Reds in ’08.
The Reds were linked to Bobby Abreu at one point during the Hot Stove season. Abreu, of course, is a fine offensive player who knows how to work a count and consistently produces solid lines. When the prices fell so dramatically, even a team with payroll constraints like Cincinnati could have landed the free agent outfielder. I can’t speak to what the front office was thinking in regards to Abreu, but they certainly recognized his defensive limitations and probably felt that he was better suited for an American League team so he could DH a lot.
It’s not just the moves they made, but the ones they didn’t make as well.
Also, the Reds traded Jeff Keppinger to the Houston Astros for minor league second baseman Drew Sutton earlier this spring. Keppinger, with a career line of .288/.342/.397 and 90 OPS+, certainly has some value. But a lot of his value comes from his versatility defensively, and he has terrible range. According to FanGraphs, he has been worth -14.8 fielding runs during his major league career. He was a black hole for the Reds at shortstop last year, so getting him out of the mix at the position will help the club upgrade by default. He is a tweener who does not hit well enough to man a corner on a full-time basis, but, according to the numbers and the scouting reports, can’t really play up the middle adequately.
Sutton is 26, but he was one of the only legitimate infield prospects in the Astros’ depleted system. Why Houston would make that trade—since they have no chance of contending—is beyond me. But the Reds saw Keppinger for what he is worth—a complimentary player with defensive limitations in his 30s—and got some value back. Getting rid of him eliminated any change that he would be forced to man shortstop for so many innings if injuries came up again with Gonzalez.
One player who I forgot to mention in my last comment regarding the Reds’ defense is Edwin Encarnacion, who is just a butcher at the hot corner. Since he broke in, he has graded out well below average. His UZR totals from 2006-2009: -9.4, -11.7, -9.8, -3.7. He has been hurt for most of this year, so that has also helped the DER for right now.
So, yes, it’s unlikely that they will convert more than 70 percent of batted balls hit into play into outs. And the sample size is still really small. But they are improved. Losing the key culprits (Griffey, Dunn, Keppinger) for the defensive futility will go along way and Grossman stressed that the team will place more of an emphasis on this part of run prevention going forward.
By comparison, the Reds have visibly spent a lot of effort to improve their pitching staff in recent years by acquiring Arroyo, Harang, Cordero and Volquez and developing Cueto and Bailey. Thus, there is evidence of the Reds changing their organizational philosophy in regards to pitching.
Ryan Freel is no longer on the Reds. He was traded to Baltimore for Ramon Castro, before being shipped again to Chicago for Joey Gathright and his jumping-over-car abilities earlier this month. At the time of the Keppinger trade, he had already been dealt to the Orioles.
Plus, Freel and Gonzalez always seem to be hurt. Keppinger would not have been a bad guy to keep for depth if he could have provided even average defense at shortstop.
Jonny Gomes was non-tendered a contract by the Rays. Gomes then signed a minor league deal with the Reds over the winter, so the franchise has almost no commitment to him. He did not make the team originally, but was called up last week. Obviously—as that picture of him with his hands up looking for the ball after he lost in the lights at Tropicana Field shows beautifully—he is not much of a glove man. He is a nice guy to have in the organization, but it’s doubtful that he is going to make any real impact with his new team.
As for calling up a minor league first baseman, have you looked at the Reds’ organizational depth chart? Kevin Barker is the starter for the Triple-A Louisville Bats. Barker can hit a bit but is a 14-year minor league veteran who was drafted back in 1996. Formerly a top prospect in the Brewers’ system, he would not be worth making a corresponding roster move with Votto battling his illness recently.
Logan Parker, 24, is more of a prospect than him. Parker has a career .819 minor league OPS, but he has fewer than 150 plate appearances above the Florida State League. Yonder Alonso is the key first base prospect, of course, and should rise quickly, giving the Reds a difficult decision on what to do at first base when he’s ready; you draft on talent, not need, so it is a nice problem to have but one of those guys will have to move.
Votto is back now, and hopefully he’ll be able to put his difficult week behind him and won’t have any lingering effects from his inner ear infection.
As far as the pitching, every player you mentioned was acquired/drafted before or during the Wayne Krisvky regime. Arroyo was traded for Wily Mo Pena by Krivsky, Volquez for Hamilton under Krivsky, ect.... Krivsky then signed Cordero and gave the extensions to Harang and Arroyo back in the winter of 2006. Harang, I believe, was acquired by Jim Bodwn (I would have to look it up) back for Jose Guillen around 2003. After Bowden was fired, Dan O’Brien was at the helm when a lot of their pitching prospects, including Bailey, were picked in the draft. The Reds certainly have good pitching and will need to have guys who miss bats and trigger ground balls to succeed while pitching at Great American Ball Park, but improving the defense will go along way with helping the development of the pitchers. Based on my conversation with Grossman, I think that now more than ever they are trying to support their promising young pitching staff through upgrading on defense. It’s not as clear-cut as when the Rays traded their right fielder and others to upgrade defensively at shortstop, but I think we’ll see more of a focus on improving defensively from them in the future.
The Reds didn't take the added savings from Griffey and Dunn and funnel it into a stud defender. They still picked up an all-hit/no field hitter in Gomes. They didn't sign Rafael Furcal or Orlando Cabrera. They didn't even swing a trade for a Matt Joyce type, much less a cheap defensive option like a Ronny Cedeno or Felix Pie.
If the Reds are emphasizing defense then they shouldn't care about Kevin Barker's bat... they should be able to find someone somewhere in their organization, on the waiver wire, or any of the unsigned free agents who can play first base that's not a catcher. Heck, move Encarnacion over there if, as I understand it, his main problems has been throwing. They've done it before ala Scott Hatteberg. My point is, finding a league-average-defensive-first basemen should be the easiest player to find and the Reds didn't. They didn't give any minor league vets a try or pluck someone off the waiver wire or trade for someone cheap... they moved a catcher over. Votto 's still experiencing symptoms... so all the more reason for the Reds to find a backup.
And yes, every pitcher I mentioned was before Jocketty... my point was that it was evident the Reds had changed their philosophy due to the types and bulks of moves. I just don't see that evidence in this "defense-first" approach.
It's all nice if Mr. Grossman says he is emphasizing defense, but I just don't see it yet.
I know you're just using this as an example. But do you think the Reds should have signed Orlando Cabrera, who was attached to a draft pick? Furcal did not want to leave L.A., and it's doubtful he would want to play in Cincy. Plus, the Reds did have the 14-M tied into Gonzalez, so having that much money tied into SS might not have been the best allocation of resources.
I get what you're saying, though, wanting direct evidence. But the changes are more subtle. It perhaps would have been a wise decision for them to invest in a cheap defensive upgrade, though. You're right there. Someone like Cezar Izturis, who granted got $6-M from Baltimore over two years.
I see where you're coming from, Richard, with the pitching, and there is not that direct evidence yet. It will be interesting to see how this plays out for them.
Just like anything, there are various degrees of how good an option is. Assuming the Reds value defense above all, signing a Cabrera is a decent option but has its problems, though he also could be capable on the offensive side of the ball. A better option if unable to trade Gonzalez is to pick up a cheap all-glove reserve, preferably one who can also act as a defensive replacement at third base.
Then again, if you can't find a all-glove, no-hit shortstop and a backup first basemen, the real problem is not your major league roster, but your minor league system.
I believe you have to look at the total contribution, offense plus defense plus baserunning. Tampa Bay made huge improvements because they went from wrost to best defense at ss while keeping the offense there the same, and then with Longoria upgraded both the glove and bat at 3b. Dunn for Dickerson? I'd say Dunn still contributes a better total package. Sacrificing the bat for the defense goes back to the 'old school'.
With the Rays, I would say that Harris (.341 wOBA) was better offensively in 2007 than Bartlett was in 2008 (.311 wOBA). The defensive difference was was huge, though, given how the TB roster was built. I thought it was interesting that he produced a 2.1 UZR in '08, down from 11.9 and 4.2 the previous two years. He was clearly a major upgrade, anyway, and there may be some noise, but that stood out to me. His knee injury limited his range, he said, and he is grading out better this year in a limited sample size. With a .690 OPS, he did not deserve to be the Rays' Team MVP, but perhaps the local chapter of the BBWAA did so as a symbol for the team defensive turnaround.
However, Harris was worth 2.1 WAR in '07 at shortstop for the Rays, with Bartlett putting up a 2.7 mark in '08. Do you think the difference was worth more than the 0.6 worth of WAR, though, given the make-up of the Rays' roster and how important the defensive upgrade was for their run prevention? With how much the run prevention pushed them to the top, given how Friedman built the club, I don't think the Rays would have been nearly as good if Harris provided the exact level of production. (On a side note, is there any SS in the AL who is playing better than Bartlett right now? He has seven homers and a .453 wOBA. The .411 BABIP is going to come way down, but where did he and Ben Zobrist find the power?)
I would say you need to understand the context for a lot of trade-offs and understand the make-up of the roster at hand.
Keep in mind, as well, that the science of studying defense is still a bit foggy. I think it was the 2009 BP Annual that said Bartlett was just an average defender at shortstop (though still vastly superior to Harris). I haven't studied UZR yet but I don't think there's been a defensive metric yet that has been universally viewed as a highly accurate tool.
Another problem I had with Tyler's entry is that the opening sentence leads us to believe this is going to be about individual defense, but winds up discussing team defense.
Overall, it ended up being vague, whether it was the qualities of Delmon Young's defense or why moving Iwamura showed a benefit.
I also thought some of the critical comments either needed a brief justification or a red pencil. If you're going to say the A's went too far, especially in a Basics article, then you need to back it up.
...But maybe I'm just grumpy about the gratuitous use of impact-as-a-verb in paragraphs 6 and 7.