(All stats are current as of the end of April 24th.)
During the offseason, Jason Heyward and Troy Tulowitzki were two of the highest-profile players on the trade block. Heyward was ultimately dealt as the Braves gear up for the future and the Cardinals look to fortify RF after the passing of Oscar Taveras. Tulowitzki was not dealt, as the Rockies hope that they can make an improbable run to the playoffs. Both players could be looking for new homes within the next year, as Heyward hits free agency (barring an extension) and Tulowitzki would be a very tempting target at the trade deadline or in free agency.
However, both players have started the season slowly. While Tulowitzki has a 103 wRC+ (which is pretty darn good for a SS), that figure is far below his 2014 results (171 wRC+) and his career figure (125 wRC+). Much of the blame can be placed on his .197 ISO, which is far below both his 2014 and career ISO. Tulowitzki has been able to counteract the drop in power somewhat due to a .370 BABIP that is far above any BABIP he has recorded over a full season. Heyward’s drop has been even more severe, as he is the owner of a B.J. Upton-esque 64 wRC+. While much of that should be attributed to a paltry .235 BABIP, some blame also can be ascribed to a poor batted ball distribution. However, it is too early to say that either player won’t see these trends reverse as the season progresses.
On the other hand, both players are suffering a precipitous and concerning decline in their plate discipline. Tulowitzki’s K rate has shot up from between 15 and 16 percent to almost 24 percent. Likewise, his walk rate has fallen to a paltry 1.6 percent as he has drawn one walk over the season. That shift is being driven by an increase in his swings on pitches out of zone, which has grown to 35 percent from 27 percent in 2014 according to Pitch F/X data:
In addition, Tulowitzki is making less contact as he swings, as his contact rate is below 80 percent – a percentage he has never had at the end of the season. He is also swinging and missing more and is over the league average for the first time since his disastrous cup of coffee in 2006. Tulowitzki’s also seen 8 percent more pitches in the zone (a higher figure than ever before), which indicates that pitchers are not as afraid of him as they once were. All of this comes directly after he had hip surgery, which suggests that he may not be fully recovered yet or that the injury may have eroded his skills slightly.
Heyward also has seen his plate discipline deteriorate but not to the same level that Tulowitzki has. First the good news: his strikeout rate, while slightly elevated from his totals in the past few years, is still in line with his career norms. However, the rest of his plate discipline numbers are worse than his career numbers. As noted by Bernie Miklasz, Heyward only has one walk, is swinging at far more pitches out of zone than ever before, and is seeing fewer pitches in the zone than ever before. Miklasz also notes that Heyward is pounding groundballs – he is currently putting 62 percent of his balls in play on the ground. This is far above his career averages (as shown in the chart below) and is a sign that chasing more pitches is not helping him generate power.
In addition to the points that Miklasz made, Heyward is also swinging far less at pitches in the zone. This season, he has swung at 58 percent of pitches in the zone, the lowest percentage since his rookie year. These diverging trends have allowed Heyward to set a personal record: for every pitch that Heyward swings at out of the strike zone, he only swings at 1.04 pitches in the strike zone.* This is far below his career ratio of 1.69.
Now, as loyal FanGraphs members (only the truly committed read the Community board!), I can hear your refrain of “Small Sample Size.” And I certainly agree that it is too early to completely believe in the magnitude of these changes. It is extremely unlikely that both players will walk less than 2 percent of the time this year. However, I believe that the magnitude and consistency of the changes is a clear sign that both players are suffering due to the erosion of their plate-discipline skills. Both players have reached the stabilization point for strikeout rate, are halfway to the stabilization point for walk rate, and Heyward is quickly approaching the stabilization point for groundball rate. In addition, per pitch metrics like O-Swing and Z-Swing stabilize quickly, with swing rate stabilizing at 50 PAs. While those stabilization points only denote the point at which the data is half noise and half signal, the changes are consistent enough across multiple measures of plate discipline that its extremely hard to argue that it could **all** be a fluke. While both of these players are plus defenders and have the power to still be plus hitters with poor plate discipline, their value will suffer unless they can find a way to turn around their plate discipline.
* This statistic can be calculated using the following formula: (Zone%*Z-Swing%)/((1-Zone%)*O-Swing%).
Sorry for the relatively long delay – sometimes life gets in the way of our best laid plans. In case you want a refresher, here is part 1: http://www.fangraphs.com/community/testing-the-eye-test-part-1/
In part 1, I found that, counter to my expectations, range correlated most strongly with FSR data of all the UZR components (UZR itself had a stronger correlation with FSR). I expected the strongest-correlated component to be errors, which was actually one of the least-correlated components. However, I wanted to go a little bit farther and look at the difference between correlations between the UZR components and FSR and the correlations between the UZR components and UZR itself to get a sense of what the fans weight more than UZR does. As a reminder, here is the data set I compiled for this analysis:
“I pulled the defensive stats of every player who qualified (minimum of 900 innings) at a position from 2009-2014 (FSR data is only available for those six seasons on FanGraphs). I then disregarded catchers, as UZR does not cover the position. Likewise, pitchers are left out because they are not covered by UZR or FSR. That left me with 761 player seasons across the other seven positions.”
Without further ado, here are the correlations between UZR and its components:
Position |# |ARM |DPR |RngR |ErrR
1B |118 |N/A |0.207 |0.930 |0.326
2B |117 |N/A |0.275 |0.907 |0.465
3B |107 |N/A |0.166 |0.948 |0.386
SS |130|N/A |0.459 |0.866 |0.384
LF | 71 |0.584 |N/A |0.895 |0.196
CF |115 |0.357 |N/A |0.935 |0.069
RF |103|0.310 |N/A |0.906 |0.061
I always had a suspicion that range was the most important component of UZR but these results are insane. It turns out range is far and away the most important component of UZR. Interestingly, the weakest correlation for range is at SS, perhaps because shortstops without proper range are moved to another position. ARM, although only calculated for outfielders (a real shame as Andrelton Simmons deserves credit for being able to make this throw), has the second-strongest correlation but lags range by a large amount. Like the FSR correlation, it is surprising that LF has a stronger ARM correlation than CF or RF. DPR narrowly edges out errors, although the correlation for errors is far stronger when you only consider infielders. Now, to get a sense of the difference, here’s the two sets of correlation subtracted from each other (positive numbers mean the correlation with UZR is higher and negative numbers mean the correlation with FSR is higher):
Position | # | ARM | DPR | RngR | ErrR
1B | 118 | N/A | -0.005 | 0.644 | 0.006
2B | 117 | N/A | 0.116 | 0.437 | -0.082
3B | 107 | N/A | 0.011 | 0.315 | 0.125
SS | 130 | N/A | 0.095 | 0.437 | 0.041
LF | 71 | 0.074 | N/A | 0.369 | 0.010
CF | 115 | 0.120 | N/A | 0.441 | -0.002
RF | 103 | 0.096 | N/A | 0.365 | -0.006
There are two different ways to look at this: one is that FSR has nearly the same correlation as UZR in most categories. That’s good! It lends a lot of credibility to FSR to know that you can predict FSR nearly as well as UZR with ErrR or DPR. On the other hand, look at the huge difference in the range column. It appears that the fans are severely underestimating the importance of having great range (or have different ideas of how to evaluate range). That’s a problem! As we just saw, range is the most important component of UZR for every position. It is also not terribly surprising as I hypothesized at the beginning of this series that the fans are underestimating the importance of range in favor of flashier tools. This also explains a lot of the discussion about Derek Jeter’s defensive ability (or lack thereof).
This sums up the research portion of this series. I think all of this does lend a lot of credence to FSR: it does reflect that range is the most important component of defense and it does a good job of properly ranking the importance of the other components. In addition, the correlation between FSR and UZR is fairly strong but not so strong that the two systems are interchangeable. However, when considering FSR, be sure to mentally adjust when a player has particularly good (or bad) range.
In part 3, I will examine some of the player seasons that produced the most disparity in the two rankings.
As long as I can remember, I’ve been a fan of good defense. Growing up my favorite player was Andy Van Slyke, and as a Braves fan I’ve had the privilege of rooting for defensive wizards such as Greg Maddux, Andruw Jones, and now Andrelton Simmons. Advanced defensive statistics are one of the things that drew me into sabermetrics and I spend entirely too much time obsessing over pitch framing.
Foremost among the new wave of statistics is UZR, Ultimate Zone Rating, which is the metric that is used to calculate the defensive portion of fWAR. In addition, Fangraphs also carries DRS and FSR, or Fans Scouting Report. While UZR is my preferred metric, I’ve always been intrigued by FSR. After all, I pride myself on my knowledge of the defensive ability of players on my favorite team and it makes sense to me that there is a wide population that has a pretty good idea of the quality of Chirs Johnson’s defense (namely, that it sucks but improved a lot in 2014).
I decided to take a look at the correlation between a player’s FSR and the components of his UZR (ARM, DPR, RngR, and ErrR, as well as total UZR). For this exercise, I pulled the defensive stats of every player who qualified (minimum of 900 innings) at a position from 2009-2014 (FSR data is only available for those 6 seasons on Fangraphs). I then disregarded catchers, as UZR does not cover the position. Likewise, pitchers are left out because they are not covered by UZR or FSR. That left me with 761 player seasons across the other seven positions. Here’s the correlations between FSR and UZR and its components for those seven positions:
Position |# |ARM |DPR |RngR |ErrR |UZR
1B |118 |N/A |0.213 |0.285 |0.320 |0.396
2B |117 |N/A |0.159 |0.470 |0.547 |0.637
3B |107 |N/A |0.154 |0.632 |0.261 |0.673
SS |130|N/A |0.363 |0.428 |0.344 |0.592
LF | 71 |0.510 |N/A |0.526 |0.186 |0.664
CF |115 |0.237 |N/A |0.493 |0.071 |0.548
RF |103|0.214 |N/A |0.541 |0.067 |0.613
There’s a lot to look at there, but first let me draw your attention to one fact: UZR has a higher correlation for every position than any one of its components at the same position. That’s a big plus for FSR, as it shows the fans don’t get so caught up in one area of a position to ignore how it fits into the whole. It also runs counter to my expectations, as I expected the fans to strongly favor players who avoided making errors (as it seems the voters of the Gold Gloves do). Instead, the component that averages the strongest correlation is range, with ARM (which is only calculated for outfielders) a distant second. Errors only beat out double play runs, which is an indication of how informed fans have moved from using errors as the primary way to evaluate defense. Indeed, errors had a strongest correlation of any component at only two positions: 1B and 2B. Further, errors had an extremely weak correlation with FSR in the outfield, with CF and RF featuring almost no relationship at all.
I was also struck by how strong the correlation between FSR and UZR was at every position. With the exception of 1B, every position’s correlation between the two metrics was above .5, with four of the seven positions above .6. The correlation between FSR and UZR was strongest at 3B, with LF a close runner up. 3B also features the strongest correlation between FSR and a component of UZR – in this case, RngR – and the smallest gap between UZR and one of its components. This finding surprised me, as I typically picture range as a CF tracking down a fly ball hit far over his head. Indeed, the average correlation between RngR and FSR is higher in the OF (0.520) than in the IF (0.454) despite the strength of the correlation at 3B.
I was also surprised to see the strongest correlation between ARM and FSR in LF, not RF which is typically known as the haven for strong arms. I have two theories to explain this incongruity: the first is that this simply is a small sample quirk. The other is that the selection bias for RF creates a situation where the distribution between the strongest and weakest arms is simply too small to make a significant difference in the data. Indeed, the range between the highest ARM in RF (Jeff Francoeur’s 9.7 in 2010) and lowest (Curtis Granderson’s -7.4 in 2014) was approximately 3 runs smaller than the difference in LF between Yoenis Cespedes’ 2014 (12.4) and Ryan Braun’s 2010 (-7.9).
Overall, this shows the strength of FSR. While its certainly not the same as UZR, the correlations are strongest between total UZR and FSR, and the components with the strongest correlations appear to generally be appropriate for the position. In Part 2, I will examine which components are over or under-emphasized by FSR.