A Week of Andrelton Simmons Missing Plays

Let’s talk about slumps. Mike Trout is drawing some attention for his elevated strikeouts, and he’s on pace for 21 fewer batting runs than he produced a year ago. Prince Fielder’s had some trouble adjusting to his move, and he’s on pace for 29 fewer batting runs than he produced a year ago. Robinson Cano up and changed sides of the continent, and he’s on pace for 34 fewer batting runs than he produced a year ago. These are all pretty big statistical declines, and while the players are each too good to give up on after so little time, their numbers are getting noticed. People are impatiently waiting for the players to look like themselves. I should note that Trout’s still been amazing, on account of being Mike Trout, but one can be simultaneously amazing and worse.

You know what nine or ten runs mean. You know the rule of thumb is that’s about what a win is. A win’s a pretty big deal, so those declines above are pretty big deals. And, of course, offense isn’t the only area where numbers fluctuate. Defensive Runs Saved, as shown on FanGraphs, updates daily. A year ago, as a shortstop, Andrelton Simmons was worth 41 runs above average in the field. This year he’s on pace for +4. He’s a full-time player on pace to be worse by 37 runs, and that’s an enormous gap that to my knowledge no one’s discussed.

Certainly, there’s no reason to believe Simmons’ level of talent has meaningfully changed. Oh, he’ll be a mediocre shortstop in time, when he’s 50, but right now he’s 24, and last year he was 23. We’ve known Simmons as the guy who’s broken the advanced defensive metrics, and talent doesn’t disappear overnight. We’ve always cautioned that you need big samples of defensive data to draw conclusions, and a month and a half isn’t enough. Based on 2014 numbers, there isn’t much we can conclude about Simmons’ defensive true talent.

But there’s true talent, and there’s performance. Over long stretches of time, the two are close together, but while true talent changes slowly and predictably, performance can dart around like a fly in a kitchen. One could say that Simmons, to this point, has earned his numbers for the year, even if they don’t quite reflect what he really is. Why should we expect a perfect reflection?

I got to wondering about what Simmons has done. And that took me to his defensive spray charts, and that took me to MLB.tv. The numbers say Simmons has been creating fewer outs in the field. What does that look like? Join me, if you will, on a tour. We’ll cover just the past week, because it’s been an eventful one. Here are six plays that Andrelton Simmons didn’t make.

Play No. 1

Simmons1.gif.opt

Inside Edge actually lists this as an impossible play. That is, they figured 0% of shortstops would turn this into an out, and the Braves announcers said that Simmons didn’t have a chance. What Simmons did do was get close enough to look like he did have a chance, and in the past he’s certainly shown the ability to slide in front of a ball and launch it accurately to first with his shoulder-mounted cannon. In truth, this probably was impossible, but if you close your eyes and imagine, you can see Simmons coming up with magic.

Play No. 2

Simmons2.gif.opt

Almost impossible, according to Inside Edge, but not quite literally impossible. The part you see is that Simmons was unable to barehand the ball cleanly, but even if he had pulled that off, it’s by no means a lock he would’ve been able to throw out Darwin Barney. Barney’s a decent runner, the chopper was slow, and Simmons would’ve had to throw the ball off balance. 1-10% seems about right, erring closer to 1%.

Play No. 3

Simmons3.gif.opt

Darwin Barney again. Super difficult play again. The play Simmons didn’t make, pretty much all shortstops also wouldn’t have made. But the thing about Simmons is he isn’t supposed to be just another adequate shortstop. He’s supposed to make the regular plays, and the extraordinary plays. One gets only so many opportunities to make an extraordinary play.

Play No. 4

Simmons4.gif.opt

Screaming line drive. Nearly caught on the fly, but the ball came out of Simmons’ glove. This was classified as a remote play, but that doesn’t mean Simmons wasn’t upset with himself afterward:

Simmons4_2.gif.opt

Hold yourself to a crazy standard and you’ll be disappointed a lot by things that wouldn’t disappoint other people. This is the third play, incidentally, we’ve seen from the same game. Prepare for a fourth. Could this have been an out? Sure. Almost was. Simmons almost pulled off a Simmons, but instead he pulled off a Normal.

Play No. 5

Simmons5.gif.opt

Here’s the real stinker of the bunch. While it’s never easy to backhand, and while the ball might’ve taken a funny bounce, a ball got through that usually doesn’t, and Simmons was initially charged with an error before the play was apparently changed to a hit. Inside Edge classified this play as basically a coin flip. Definitely not easy, but, again, this isn’t a post about Derek Jeter.

Play No. 6

Simmons6.gif.opt

Simmons came up just shy of stopping this hot-shot grounder from Hunter Pence. Because he didn’t, Pence finished the play on second base, with the rare two-out double that doesn’t score a runner from first. If made, this play would’ve gone on a highlight reel, but instead that’s one fewer highlight for Andrelton Simmons, not that he doesn’t already have enough.

What’s been learned? First of all, I’ll note that none of these runners actually ended up scoring. Everybody got stranded, so Simmons didn’t meaningfully cost his team much of anything. And all but one of these plays would’ve been incredibly difficult, so we can’t even classify this as a defensive slump, really. And this brings us to the expression, that defense never slumps. When people say that, they’re thinking about the routine plays, the high-percentage plays. That defense doesn’t slump, barring a case of the yips. But there’s easy defense and challenging defense, and while the challenging defense is more infrequent, it’s also volatile, because the percentages shift so drastically. The idea behind the expression is that defensive performance is stable. But it likely isn’t toward the extremes, because the extremes involve full-out dives and barehands and throws while falling away. It takes little time for routine defensive ability to stabilize. It takes a lot more time for extreme defensive ability to stabilize.

For Simmons to do what he’s done in the past, statistically, he’s had to be amazing at both the routine and the exceptional. Probably, based on the numbers, he was a little over his head in the exceptional department, and in a sense that’s what regression is. His numbers are down now because he’s made fewer of the insane plays, but the reality of Andrelton Simmons might lie in between, where maybe in the future he makes one or two of the plays above. We know he can do things most shortstops can’t. Andrelton Simmons has the same defensive talent as ever. All the numbers are are estimations, and it can be a challenge to estimate the rare.




Print This Post



Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.


46 Responses to “A Week of Andrelton Simmons Missing Plays”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Fergie348 says:

    Brandon Crawford makes at least half of those plays.

    -53 Vote -1 Vote +1

    • Not a Homer says:

      *Brandon Crawford makes at least half of those plays if he had the quickness and arm strength of Andrelton Simmons.

      +12 Vote -1 Vote +1

      • Fergie348 says:

        I’ll give you quickness but no one playing shortstop today has a more plus arm than Crawford.

        -34 Vote -1 Vote +1

        • AK7007 says:

          You aren’t making much of a case for not being a homer.

          On the other hand – did anybody click on that link I posted? You can close your eyes and pretend that Crawford made the plays Simmons missed, because odds are he has made plays just like them in the past. And missed plays like them as well. And missed easier ones. But somehow made harder ones. Baseball is so quirky sometimes. Nobody reminds me better than Jeff though.

          Vote -1 Vote +1

        • Bronnt says:

          I suspect you know nothing about Andrelton Simmons. His arm is amazing. He was drafted as a pitcher because he was clocking 98 regularly.

          +16 Vote -1 Vote +1

        • Dan Ugglas Forearm says:

          The main reason Simmons is what he is on defense is his arm. He’s never had an 80-grade glove, but his arm allows him to play a bit deeper in the hole than most shortstops, and that allows him to get to more balls than most shortstops, knowing that he can make up for it on the throw. Without an arm like his, I doubt his range ratings would be anywhere near as high. Arm=Simmons. Simmons=Arm.

          Vote -1 Vote +1

    • Bip says:

      I think the absolute strongest statement you can make is Brandon Crawford *could* make *some* of those plays, *sometimes*. But you can also say that about Simmons, so what are you really saying?

      The Giants have shifted very aggressively this season. In their series against the Dodgers, Crawford appeared to be everywhere at times, but that was because he was. He may have made some of these plays because he was better positioned for them. I don’t know if we should necessarily credit him for that though.

      Vote -1 Vote +1

  2. John says:

    We’re probably looking at DRS too early.

    Tulo leads with 11 but Jhonny Peralta is 4th among shortstops with 5. That screams of a stat that needs a bigger sample size.

    Vote -1 Vote +1

    • larry says:

      from what ive read about DRS, we will still be looking at it too early when the season ends.

      Vote -1 Vote +1

      • jim S. says:

        The real question is: who and how many ARE actually looking? John Dewan at BIS (the folks who produce DRS) won’t actually say how many evaluators look at each play. In Phoenix in March I had occasion to ask him and he replied with a smile, “Fifty.” Sounded a bit, er, defensive to me.

        Vote -1 Vote +1

        • bstar says:

          jim, I saw someone from BIS on Clubhouse Confidential this winter and he said one guy looks at every play of the game and then 3 or 4 guys review his work. He mentioned a total of 10-12 man-hours on each game.

          Vote -1 Vote +1

    • Arc says:

      What do you mean “probably”? It is known by everyone and written in this very article that it’s too early.

      Vote -1 Vote +1

    • Or you could realize that Jhonny Peralta is a pretty decent shortstop, and has been for a while according to advanced defensive metrics. We’re looking at DRS too early only in the same sense that if we were to talk about offense we’d be looking at wOBA too early. They have roughly the same variation.

      Vote -1 Vote +1

      • DNA+ says:

        The variation is not the same though. For offensive outcomes the variation is in the sampling of real outcomes (e.g., we never mistakenly record an out as a home run, etc.). For the defensive measures, much of the variation is simply in the qualitative judgements of the play by the scorers. It is variation that isn’t real, but rather introduced into the system by our inability to accurately and precisely measure the outcomes.

        Vote -1 Vote +1

        • Although by the same token, our recordings of singles, doubles, triples, etc. is just our “best guess” at how hard a ball was hit, and how deserving a player is of credit for that hit.

          We give a player full credit for a single past Derek Jeter that wouldn’t have gotten past Andrelton Simmons, even though everything else about the hit was the same. We need to be careful about how sure we are that we’re actually measuring real things. Sure, defensive metrics are more qualitative, but even the quantitative nature of offensive statistics are full of arbitrary qualitative assumptions.

          Vote -1 Vote +1

        • Although by the same token, our recordings of singles, doubles, triples, etc. is just our “best guess” at how hard a ball was hit, and how deserving a player is of credit for that hit.

          We give a player full credit for a single past Derek Jeter that wouldn’t have gotten past Andrelton Simmons, even though everything else about the hit was the same. We need to be careful about how sure we are that we’re actually measuring real things. Sure, defensive metrics are more qualitative, but even the quantitative nature of offensive statistics are full of arbitrary qualitative assumptions.

          Vote -1 Vote +1

  3. Brian L says:

    I prefer to think of this as opportunities taking a long time to stabilize, as opposed to performance or defensive ability stabilizing. In a way it’s just nomenclature, but I think too often people equate DRS or UZR directly to the player’s performance, i.e. a decrease in UZR meaning the player is playing worse defense, rather than recognizing how dependent these metrics are on the number and “makeability” of defensive plays provided to the defender.

    Jeff’s post shows this perfectly. Had 4 of those 6 balls been a matter of inches closer to Simmons, he could have “performed” the exact same and made 4 of the 6 plays. He could be performing at the exact same level this year as last year, just not seeing the same quantity of barely makeable opportunities that allow him to run up historically high metrics. Over a larger sample, the range of defensive opportunities becomes more normally distributed and the player’s performance is represented better by the numbers.

    Vote -1 Vote +1

    • King George II says:

      That’s often the same thing with offense. If a few fly balls traveled 10 ft further or a few extra hits stayed fair, you’d see a dramatic increase. This may just be semantics, but I view both as performance-based and as different than true talent-based.

      Vote -1 Vote +1

      • Brian L says:

        The distinction I see is that on offense you have a meaningful degree of control over your opportunities. You can swing or not swing, and you have some degree of control over the type of contact you make. On defense, you have no control over where the ball is hit in relation to where you start.

        Vote -1 Vote +1

        • Neil S says:

          But you *do* have some control over where you’re standing to begin with.

          I think the better analogy to hitting would be that you have limited control over who the fielders are – some days you’re hitting a grounder in the hole against Andrelton Simmons, some days you’re hitting hitting it into that same hole against Derek Jeter.

          Vote -1 Vote +1

    • bstar says:

      Well said, Bryan, I agree totally with your perspective. In fact, I think differing opportunities to make great plays is one of the strongest underlying factors that drive these metrics.

      But that’s the reason one-year sample sizes are too small, let alone a six-week one. Three years? I think we can reasonably expect that to have evened out.

      As for Simmons, I have little doubt he is an all-time great defensively. I also have little doubt Jeff has hit the nail on the head: he is just barely missing making these great plays.

      Yesterday, Simmons fielded a ball deep in the hole, a really tough play. But he slightly airmailed the throw and forced Freddie Freeman off the bag so he didn’t get the runner. Last year that guy was out 19 out of 20 times.

      His performance has been down slightly, no doubt.

      Vote -1 Vote +1

  4. dirtbag says:

    What would Willie Bloomquist have done?

    Vote -1 Vote +1

    • Samuel L Farnsworth says:

      WWWBD? I can tell this meta joke is too complicated for you; it’s like a Linda Ronstadt song — blew by you.

      Vote -1 Vote +1

  5. Izzy says:

    Jeff Sullivan is the best.

    Honest question: Inside edge has Andrelton Simmons as only not making four plays this year (one impossible, two remote, and one unlikely). His routine, likely, and even plays are all at one hundred percent. Why then, are there six plays here that he didn’t make?

    Vote -1 Vote +1

  6. Costanza says:

    I refuse to allow Trout to be mentioned w/o mentioning Cabrera. Miguel Cabrera is on pace for about 40 fewer batting runs than last year.

    Vote -1 Vote +1

  7. filihok says:

    Unsurprisingly, a fantastic article from Jeff Sullivan.

    Not looking forward to Sullivan’s slump

    Vote -1 Vote +1

    • bookbook says:

      Yes. But will the slump reflect a variation in the quirky realities of baseball or inherent performance variability from the writer’s mind? What is Sullivan’s true WOBA (writer of bull***[oney] average), and how long does it take to stabilize?

      Vote -1 Vote +1

  8. cnote66 says:

    Then Simmons goes and makes a dynamite play tonight!

    Vote -1 Vote +1

  9. BaseballGuy says:

    A very good shortstop who isn’t THAT much better than everyone else. Last year was a bit of a fluke in that he happened to get a ton of balls right at the edge of his considerable range. This year, so far, they’re an inch or two farther (or closer, and are therefore categorized as easier chances). I guess this is just a long way of saying that defensive metrics need such a huge sample size that they’re borderline unusable.

    -9 Vote -1 Vote +1

    • Steven says:

      They are borderline unusable if you are using them incorrectly, just like every other stat. Maybe they are slightly less reliable than other offensive stats, but not inordinately so. If you are using DRS this early to say what the true talent level of defenders are, then you are just as wrong as if you were using wRC+ to decide which hitters are the best (e.g. Seth Smith and Adam LaRoche are not two of the best 7 hitters). It’s not like other stats are reliable true talent indicators either (e.g. Marlon Byrd is not a true talent 136 wRC+ even if he legitimately hit that last year).

      Vote -1 Vote +1

  10. Brian Snyder says:

    Im curious what happens when Simmons makes a play on a ball that would be deemed “Impossible” if he didn’t make the play. Do we then have to go look at all the impossible plays not made by other fielders and show them as “almost impossible”?

    Vote -1 Vote +1

  11. Greg says:

    Great article. Thank you.

    Vote -1 Vote +1

  12. coreyerb says:

    Errors are kind of useless but I wonder the last time the official scorer in Atlanta gave Simmons an error. #5 could clearly have been an error, and likely would be if it were on the road, but gets changed in Atlanta. Seems like every play like that is ruled in his favor.

    Vote -1 Vote +1

  13. john says:

    I’ve been to a number of his games this year and he is definitely stepped back a level. I’ve seen him make really bad errors (official scorers in Atlanta tend to call them hits) and he hasn’t been making spectacular plays.

    On the first point I believe it might be small sample size and routine errors will end soon enough. But on the spectacular plays, I wonder whether the incessant shifting is hurting him as it makes you make plays from different angles.

    Vote -1 Vote +1

  14. Bip says:

    I should note that Trout’s still been amazing, … but one can be simultaneously amazing and worse.

    Yeah, this really only applies to Mike Trout.

    Vote -1 Vote +1

  15. Tyler says:

    I’d love to see a study on SS positioning. One of the reasons Simmons is so good is because his arm allows him to play deep in the whole, which allows him to get to more balls. Of course, I don’t actually know that he plays deeper than most shortstops, I’m just basing that off of what I see on the television, which is of course not much.

    Vote -1 Vote +1

  16. Phantom Stranger says:

    I watch about 90% of Simmons’ defensive innings. A couple of things about the falling numbers this year. Chris Johnson is being much more aggressive this season in cutting off easy grounders to SS. Last year he was hugging the baseline and let Simmons range closer to third. The Braves or someone is positioning Simmons closer to second a lot more this year. I don’t know if that reflects poor shift research by the Braves or their pitchers simply not pitching to the projected shifts. The second base play for the Braves this year has been historically awful, they’ve cost Simmons more outs than can be expected with mediocre doubleplay turns.

    Simmons has made a few more mistakes this year, both in throwing and handling the ball. He did miss a little time with injury, I don’t think he’s playing at 100% capacity in the field all the time. None of this is real cause for concern, he still has the best defensive tools in the game and the defensive stats are a mere blip at this point.

    Vote -1 Vote +1

  17. DD says:

    Sullivan: #Andrelted another great article

    Vote -1 Vote +1

  18. Martin says:

    Another factor for his DRS to decline could be that the Braves now have a group of pitchers that get more fly balls than ground balls. This could also help explain why Heyward has an insane 11.2 DRS or 2.8 more than Mike Trout. Heyward had a similar experience last year when his DRS dropped, but its up again this year.

    Vote -1 Vote +1

  19. Martin says:

    sooo…. I posted the wrong stats. Rookie mistake looking at the DRS leaderboard.

    Heyward has 16 DRS in 2014. Trout has 6. Heyward had 20 in 2012.

    Vote -1 Vote +1

  20. What would be interesting to see is what Simmon’s advanced fielding numbers looked like in 2013 broke up into quarter-seasons, which is roughly where we are at now. Then we could see the volatility of the statistic, and get a better sense for how this quarter-season fit in with his other ones.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>