Brandon Belt’s Turning Point

“I know we were playing Philadelphia, and I want to say we played the Dodgers after that. End of July some time. I kind of realized that, hey, you gotta help the team out somehow. … Sometimes you get lost out there, and you try to start playing to prove yourself, playing for yourself, however you want to say it, but if you go up there and remember that this is a team game, and you’re there for them, you’re going to play better personally in return.” — Brandon Belt

No matter how many player interviews you’ve read, this quote from Brandon Belt fits right in. There’s little to separate it from the post-game interviews that laud camaraderie and perseverance above strategy and nuance. That’s fine — admit too much and you’ve given your competition information. There is one aspect of this quote that might be a little different than most quotes, though. Belt basically provides the exact date on which his 2012 season turned for the better.

Arbitrary endpoints get no love around here. Someone even called seasons arbitrary once. So when you pull out Belt’s first- and second-half splits (He hit .293 in the second half!) you might get some howling. But, as you can see in the quote above, players have moments. They make an adjustment with the hitting coach that sticks and makes a difference. They spot something in their approach at the plate. A nagging injury heals. A personal conflict resolves itself.

Sometimes, the player just needs an epiphany to clear his mind. And there was no doubt among the Giants’ beat writers that Brandon Belt was having a hard time in Philadelphia. Carl Steward, writing for Giants Extra, the Giants blog for the San Jose Mercury News, wrote on July 24th that Belt’s “lack of confidence” was “appalling” and his body language was “pitiful to the point of pathetic.” Belt, in his own blog that was ostensibly responding to those critiques, also mentioned games that he personally lost for the team that week.

The Giants’ first baseman was having a tough time before an off day on July 23rd. Then he got back in the lineup and the “rest is history,” as he put it.

But the how is a little nebulous, maybe. When I pushed Belt about his power numbers, though, he gave us a testable assertion about his second-half changes.

I think I learned how to hit in this park in the second half of last season. You gotta hit your line drives. You have to hit line drive doubles, triples, whatever, because if you hit the ball in the air, it’s going to get knocked down. You don’t worry about trying to hit the ball out, you worry about trying to get your base hits. Try to hit the ball on a line. … Those home runs come when you get comfortable in there. Home runs come in cycles for most ballplayers.

So it’s fairly easy to check the numbers now. The player has given us what is hopefully a non-arbitrary endpoint (July 23rd), and he’s outlined a change in approach (more line drives). His batted ball numbers before and after that date (thanks to BaseballHeatMaps.com) are eye-opening:

Before July 23 After July 23
GB% 43.2% 36.1%
FB% 40.5% 30.6%
LD% 16.2% 33.3%
HR/FB 6.7% 6.8%
HR,FB Distance 279 274

It looks like Brandon Belt knows what he’s talking about. He certainly found a line drive swing in the second half — 33.3% is an elite number. He hit fewer fly balls, the ball didn’t go as far, and he didn’t hit as many home runs as he might have with a more fly-ball heavy approach, but he certainly found the frozen ropes.

brandonbeltbbmix

Now the difficult part, at least for Giants fans: A 33.3% line drive rate is not sustainable, most likely. For batters with more than 1500 plate appearances over the last three years, Joey Votto leads baseball in that statistic with a 25.5% number. And Votto himself defies the findings that suggest that line drive rate is one of the flakiest stats — its year-to-year correlation was the worst among the most common hitting metrics. Even though it was great that Belt found a line drive stroke that worked for him in the second half, he probably also was doing something that he’ll find difficult to replicate exactly the same way again.

But maybe he doesn’t have to. His line drive rate for the year (25.6%) is not as much of an outlier. And all of Belt’s seven home runs came in June and September. Perhaps a year full of this newfound approach will lead to more power outbursts as he gets comfortable in his role as the starting first baseman of the Giants. It’s certainly preferable to the funk he found himself in on July 22nd, 2012.




Print This Post



Graphs: Baseball, Roto, Beer, brats (OK, no graphs for that...yet), repeat. Follow him on Twitter @enosarris.


28 Responses to “Brandon Belt’s Turning Point”

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

    Is the chart mislabeled, or did you run the numbers from June 23rd instead of July?

    Vote -1 Vote +1

  2. LTG says:

    Two questions (both of which could turn out to be very stupid):
    1) What does GB/FB mean in the chart such that the results are 6% and 4%?
    2) Is LD% reliable enough to conclude that there was a (definite, highly likely, probable) change in his quality of contact?

    I would also like to have seen his BABIPs on the various batted-ball types in order to deepen the analysis and rule out obvious luck-influence.

    Vote -1 Vote +1

    • Eno Sarris says:

      1) Supposed to be HR/FB. Got it.
      2) There’s just no way a 1/3 LD rate is sustainable. Even on the seasonal level. The best line drive rate in a qualified season since we started carrying the stat is Mark Loretta’s 30.7%. Second is 30.2%. http://www.fangraphs.com/leaders.aspx?pos=all&stats=bat&lg=all&qual=y&type=2&season=2012&month=0&season1=1974&ind=1&team=0&rost=0&age=0&filter=&players=0&sort=6,d

      Add in the fact that it has a poor year-to-year correlation, and the BABIPS (which are tied to line drive rates anyway) are largely irrelevant.

      Vote -1 Vote +1

      • LTG says:

        The question of reliability was not so much about year-to-year correlation nor sustainability, but rather correspondence between what happened on the field and what the numbers say happened. I don’t think your response addressed that.

        And BABIPs in the aggregate are largely a function of LDs, but BABIPs on individual batted-ball types are, obviously, not. If he had a second half jump in BABIP on FBs or GBs that might illuminate the analysis. No?

        Vote -1 Vote +1

        • tynandaly says:

          As a fan who watched him play almost every night, I can say Belt DEFINITELY improved on his quality of contact in the 2nd half of the season. He went from bloopers and easy ground ball outs to hard liners and hard grounders that escaped the infield for base hits. The bigger question is whether he can carry that hard contact this season as well.

          Vote -1 Vote +1

      • chanelclemente says:

        You’re clearly right, but I’ve never seen someone hit with so much top spin on a ball. I wonder what role that plays, if any?

        Vote -1 Vote +1

  3. Mr.MojoRisin says:

    I can’t wait till he’s on all of my teams this year

    Vote -1 Vote +1

  4. Pinstripe Wizard says:

    Belt might be a guy that could sustain a fairly high LD rate. He has never been a big home run guy, including during his time at the Univ. of Texas. He seems to have tried to hit home runs during the first half of last season, and then figured out he wasn’t a home run hitter. Looking at his swing and frame, I would think he could potentially have very good bat control if he wasn’t trying to backspin balls.

    Vote -1 Vote +1

    • Petruchio says:

      >Belt might be a guy that could sustain a fairly high LD rate.

      That sounds like wishful thinking to me. I’ll believe it when he has much more than a half season’s worth of PAs with a high LD%.

      Belt doesn’t need a ton of power to hit balls out of ATT, if he has good bat control. Remember there’s a short porch down the right field line, and pull hitters have taken advantage of it in the past. His best chance for improvement is to cut down on those strikeouts, and let the BABIP fairies do the rest.

      Vote -1 Vote +1

      • He’s only really had about one full season’s worth of plate appearances. Half of that season he sustained a high LD%. My point was that the adjustment might have been something that he needed. I feel like a lot of hitters probably try to hit home runs when they are first called up so that they can make a good impression. If the adjustment sticks and if the second half is more indicative of his true talent, he could be a high LD guy.

        Vote -1 Vote +1

  5. Jason says:

    I think the arbitrary end points conversation has more merit with young players and prospects. With a guy in his sophomore season like Belt, I think there is more to the argument that 2nd half performance may be a better true talent indicator than career performance to date.

    Vote -1 Vote +1

  6. Allan Brownridge says:

    Cool read – thanks.

    Vote -1 Vote +1

  7. Naveed says:

    It’s worth noting that Carl Steward’s writing is remarkably unprofessional, even by the pathetic standards of Bay Area sports writers.

    Vote -1 Vote +1

    • Scott says:

      Yeah, that was awful, lazy writing by Steward. I think he later apologized for it, for what it’s worth.

      Vote -1 Vote +1

    • EricR says:

      Steward was awful for large parts of the season. The media in general harping on Belt’s “body language” became one of the most ridiculous and tiring talking points of last season. It was often combined with talk of how Brett Pill and Hector Sanchez are so good, and they “just LOOK like hitters”. I’ve never heard Krukow or Kuiper talk badly about a player the way they would about Belt. They just constantly threw him under the bus on broadcasts. It’s quite stunning how Belt was able to work through all of that.

      Vote -1 Vote +1

    • sboston says:

      19 xbh in his last 165 abs. .333 last 2 months. 3 homers. Looks like he sacrificed power to raise the batting average. It looks like he has room to grow, but hitting in that park … definite ceiling there.

      Vote -1 Vote +1

    • mooks says:

      i think steward once spent an entire giants post-game blog talking about the ‘upstart A’s’ to get them a little more attention. i mean jeez, i even like the A’s, but i came here to read about the giants seeing as it’s a GIANTS post-game blog. just awful.

      the guy who regularly does that blog, alex pavlolic, is pretty excellent though. and baggarly isn’t so bad himself even after his promotion. the rest, i could do without.

      Vote -1 Vote +1

  8. Nivra says:

    If you bin all his PA into groups of 10 or 20, then calculate LD% for each bin, you would come up with 47 such bins of 10, or 24 such bins of 20. Then if you model each bin as a binomial with an unknown p, you can actually compare your distribution predicted by the pre-July 23rd bins to the distribution predicted by the post-July 23rd bins. Then come up with a likelihood of this happening by chance.

    Or perhaps it’s best to simply do a binomial based t-test of sorts. Assume a binomial distribution centered on the pre-July23rd LD rate mean, then find the probability that distribution would provide data like the post-July 23rd LD rate.

    Vote -1 Vote +1

  9. Nivra says:

    Actually, here you go, the second test is easy to do:

    Pre-July 23rd PA: 258 PA
    Post-July 23rd PA: 214 PA

    Probability that the second half distribution came from a p = .162: 8.4*10^-10

    OK, so let’s try something slightly different.

    Assume a true talent binomial of p = .256 for the season.

    Probability of first half: 0.00022
    Probability of second half: 0.008

    So yah, there’s clearly a very significant effect going on. Pre-July 23rd and Post-July 23rd did not come from the same distribution.

    So how do we predict 2013?
    Method 1) Take his Post-July 23rd rate and regress it to league average. This method assumes that he made a change on july 23rd and it’s a change that he can replicate consistently going forward.
    Method 2) Take his seasonal LD% and regress to league average. This method assumes that he won’t be able to consistently replicate his July 23rd change, and may regress back to his pre-July 23rd form at certain points during 2013.

    Most Giants fans will hope for Method 1, obviously.

    Vote -1 Vote +1

    • Travis L says:

      I started writing this comment to say that LD% doesn’t stabilize that quickly. When I checked the stabilization chart (http://www.fangraphs.com/blogs/index.php/when-samples-become-reliable/), I realized that LD% stabilizes quickly (over 150 PA). So there is hope for Method 1 there.

      That brought up another question — if LD% stabilizes so quickly, why is there such a low Y2Y correlation?

      That led to think this: perhaps some (non-predictive) statistics like BABIP are highly luck-influenced. And other stats, like LD%, actually reflect the underlying skill, *but it’s the underlying skill that significantly fluctuates*.

      Any thoughts?

      Vote -1 Vote +1

      • Steve Staude. says:

        I think it’s worth pointing out that Pizza Cutter’s study that you reference is talking about line drives per plate appearance, not per batted ball (as in LD%). LD% wouldn’t “stabilize” until later, I’m sure.

        Anyway, luck vs. fluctuating skills… it’s an interesting point you make. Still, it’s hard to imagine there’s not a good amount of luck involved in LD%.

        Vote -1 Vote +1

      • bradley says:

        It is like quantum mechanics, both things are happening at the same time, despite their intuitive improbability or impossibility. LD% may fluctuate because it takes significant luck to be that consistent.

        Vote -1 Vote +1

  10. Hurtlockertwo says:

    I never read the Belt quote during the season but he did seem to put it all together in the second half. Belt, Posey and Scutero just tore it up in the second half.

    Vote -1 Vote +1

  11. I clearly remember the Philly series. I’m a huge Belt fan, but he looked lost. I was blogging for Belt to be sent down for a few weeks. Most were calling for more Hactor Sanchez or a trade for a first baseman. Krukow on his morning show was very candid about Belt being lost and how H. Sanchez was a professional hitter. The Giants stuck with Belt despite the criticism and it payed off.

    Vote -1 Vote +1

  12. shthar says:

    How bout an article explaining why he can only homer against left handed pitchers?

    Vote -1 Vote +1

  13. bradley says:

    I could be wrong, but I think Belt is going to be a stud. He will hit for power and average, and could within a year or two be a legit MVP candidate. Posey, Sandoval and Belt will make up a solid core, with Pagan and Pence etc, as nice pieces around these studs.

    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>