Monday, November 16, 2009

Portents of Defensive Doom

Projections are a weird beast. The grinding mathematical efficiency of the whole thing leaves a bad taste in many people's mouths. The wide range of offensive projections run the gamut and give hope to those searching for it, provided they search long enough. Jeff Zimmerman of Beyond the Box Score uses age regression and past performance (explained here) to project UZR for every player with at least 63 games played at a given position over the past 4 years. Guess what? It doesn't look good for our Blue Jays.

I've created three groups: Current Blue Jays, free agent Blue Jays, and potential Blue Jays as targeted by myself or the greater rumour mill.

NamePositionProjected UZRNumber of Games at Position in Last 4 YearsAge on Opening Day
Adam LindLF-2.020826
Travis SniderLF-0.76922
Vernon WellsCF-8.055332
Aaron Hill2B1.948328
Lyle Overbay1B-0.355033
Edwin Encarnacion3B-9.047627
Jose BautistaCF-4.26829
Joe Inglett2B-2.312229
Marco ScutaroSS-0.331134
John McDonaldSS1.429035
Rocco BaldelliCF1.711928
Brandon Phillips2B6.258928
Carlos GonzalezCF1.011224
Dexter FowlerCF-7.613624
Matt MurtonLF0.819028
Felipe LopezSS-8.727929
Adrian Beltre3B7.655231
Brett GardnerCF3.012126

Before we get into a big fight, let's get this out of the way.
  • No, UZR is not perfect. What it does, especially with 3 years of data, is paint a pretty good picture of a player in the field.
  • No, projections aren't perfect either. While the varieties of offensive or pitching projections swing wildly, I think defense is the type of thing that can project well. For most of these dudes, this is who they are.
  • A lot of these guys, especially the young guys are close to average. This is because, well, that's how it gets to be the average. Young players require more games to truly show how crappy they can be, so they tend to get the benefit of the doubt.

Sigh, not happeningSafe to say this bad defensive team isn't about to get much better. Some of these stiff corncobs will benefit from a healthy slathering of Butterfield, but let's not expect too much. EE displayed a strong arm at third base, but he is who he is. If anything, his rocket arm is detrimental to his chances of improvement as a fielder (his strong arm overcame other deficiencies, allowing bad habits to creep in at an early age.)

Had Jose Bautista done ANYTHING at the plate in the parts of 2009 that "mattered", I'd be tempted to give him the everyday rightfielder's job in 2009 2010. But he didn't so fuck him. Would a guy like Matt Murton be much better or worse? Probably not, but he'd be cheaper. Is he a non-tender candidate? Likely. The arbitration system is broken and I don't expect the Jays to Do The Right Thing by their journeyman.

As for the "targets", not much to choose from there. Rocco's best days in the field are behind him, Matt Murton is pretty much average all around. Brandon Philips is a fine second baseman but the Jays already have one of those. The numbers suggest both men are good where they are. Former Jay Felipe Lopez had an outstanding year with the bat though his defense is pretty much awful. Two of the young outfielders that would come to town at a terrible, terrible price are speedy glovemen with limited offensive upside (Gardner & Grand Theft Auto) while Fowler is a toolsy stud that is charitably described as raw (read: black dude!) with great potential.

Additionally, the Jays added a couple quad-A nawspects this month. Either of these guys could become another Marco Scutaro or Johnny Mac. Mike McCoy played all over the diamond with the bulk of his time coming at short. His defense was worth +7 runs according to TotalZone. He also pitched an inning in 2009, something I haven't seen a Blue Jays position player do in a long, long time. Jarrett Hoffpauir plays second base almost exclusively and doesn't appear to do it particularly well. So we can pretty much count him out.

I'm of two minds when it comes to shoring up the Jays potential defense. It would be foolish to overemphasize the impact of marginally better fielders. The difference between a good fielder and an average one (or an average fielder and a bad one) is rather slight. A large percentage of plays made are pretty routine, whomever you choose to send out there at the big league level has a pretty good shot of making a clean play on it.

That said, we must all recognize one thing: if Roy Halladay goes, so go a whole lot of strikeouts and a whole lot of groundballs. If Shaun Marcum is to step in and reclaim his place in the rotation, many flyballs come with him. Dustin McGowan has a GB/FB above 1, but he certainly puts his share of runners on base. Cecil, Romero, Rzepczynski, Purcey, Janssen&mdash none of them are as efficient as HALF of Roy Halladay. Defense will matter for this club now and in the future, to support the burgeoning pitching staff and give a team in need of every possible break slight edge.


  1. I think the proper vernacular for black dudes are "beasts".

  2. So, with +7 defence, isn't McCoy a good SS option, considering his high OBP and his speed?

    You didn't say much about him. Why?

    I scouted and signed him myself.

    I am very handsome. Ask Andrew Stoeten. He likes me, as you should REALLY like Mike McCoy, our starting SS in 2010.


  4. Why would anybody want Brett Gardner?

  5. Great post - it inspired me to finally read about how UZR is calculated.

    I like UZR but one of the limitations it seems to have is that excellent defenders 'steal' balls from those around them. If a ball is hit into the hole for example, both the 3rd baseman and the shortstop are responsible for it. So if an excellent defender like Rolen makes more plays in that zone, it (as far as I understand) would inflate Scutaro's numbers because he is not penalized for not getting that ball. Whereas if Encarnacion misses something Rolen catches, Scutaro is penalized.

    UZR ratings seem to be dependent somewhat on the fielders around you. To what extent I'm not sure. Could that (partly) explain why Scutaro's UZR plummeted towards the end of the year, and was always really good with Rolen? Is Wells bringing down those around him?

    Would replacing the really poor defenders on your list, or changing their positions, improve everyone's numbers? Not disagreeing necessarily but I wonder if it is wise to overestimate the impact of really good or (in the Jays case at CF and 3B) well below average fielders.

  6. I think that's a legitimate point George, but I think the reason UZR needs 3 years to present the entire picture helps weed that kind of stuff out. Some people believe Chase Utley's numbers are inflated because he plays such an extreme shift, take that for what it's worth.

    I don't think Scutaro's numbers fell off because of Rolen leaving, I think it was fatigue. He was never a crazy range guy, he just "made all the plays." As the season wore on, he seemed to tire physically and mentally, causing the error runs portion of his UZR to take a dive.

    @Viner: I didn't say much about McCoy because I don't know much about him. If those numbers (in just one year, in 80 games, at age 27) hold true then sweet, sign me up.

    Brett Gardner? I'll take a speedy flycatcher that walks in 10% of his at bats. .337 wOBA last year? Sold.

    Also, I edited a bunch of typos in the post. I rush when I'm at work lest I get fired.

  7. By the way, Gardner would be part of a LARGE package in exchange for Paul Beeston's immortal soul.

  8. Plus, Wilner, Brett Gardner is whiter than the driven snow. That alone makes him valuable.



Send forth the witticisms from on high