原文連結
http://www.fangraphs.com/blogs/all-star-break-pitch-framing-update/
FG今天的這篇文章把上半季的接捕能力訴諸數據來討論
不過因為用的數據不同，所以限制和探討的方式也就不同
但跟之前那種質化的探討，這篇有不同的趣味
因為時間的關係，加上這一篇重點在數據，所以就不細細翻譯了。
不過重點應該都有翻到，有錯請指正
If this were any other stat, it wouldn’t be worth a post. If this were
instead “All-Star Break Home-Run Update,” it’d be a waste of your time,
because you could simply just look up the stat on the FanGraphs leaderboards.
They’re right up there! But, at the moment, FanGraphs doesn’t house and
update any pitch-framing statistics, and while that could change in time, that
’s the way things are today, meaning this post could have some substance.
Most people can’t look this stuff up on their own, so I’m here to provide
for you while crossing something off my weekly quota. Everybody wins.
FanGraphs is pretty selective for the intellectually curious. And, of course,
baseball fans, and intellectually curious baseball fans have generally been
interested in pitch-framing research. It’s just another thing that players
can be good or bad at, so fans want to know where their catchers rank. Right
now, we have a little break in regular-season action, so it seemed like a
good time to post the latest numbers, through the middle of July. It was
either this or a .gif post about Yasiel Puig and no there weren’t any other
options. I’ll give a quick explanation, before the data.
You can skip this part if you’ve read this part before. This is about a stat
I’ve called Diff/1000 because whatever. It’s the difference, per 1,000
pitches, between actual strikes and expected strikes, based on
plate-discipline statistics. Those stats are available here on FanGraphs, and
you can calculate these numbers yourselves, whenever you like. This isn’t as
rigorous as a PITCHf/x analysis, but what this has going for it is
simplicity, and it’s all approximately correct. There’s good agreement
between the numbers generated by this method and the numbers generated by
more detailed methods. So I prefer this, because I can calculate the numbers
I want in a minute, and I don’t have my own PITCHf/x database. Anyway,
Diff/1000. A positive number means more strikes than you’d expect. A
negative number means fewer strikes, and so a smaller strike zone. The
average team throws several thousand pitches in a regular season, which I
probably didn’t need to write.
以下數據用的是Diff/1000。他的意思是說，每1000球中，被判的好球和理應是好球的數
字差異。這數據是根據FanGraphs上的選球數據來的，並不像PITCHf/x分析那麼嚴謹，但
這樣分析簡單多了，而且大致上是正確的。大於零表示被判的好球比預期的還多，負數就
表示被判的好球比較少，好球帶比較小。
In the table below, you’ll find data for all the teams. They’re ranked in
descending order of Diff/1000, but what you’ll also find is 2012 Diff/1000,
and the year-to-year difference between the two. Which teams are doing well?
Which teams are doing poorly? Which teams are doing differently from how they
did a season ago?
下表是各隊的數據，以今年至今Diff/1000由高到低來排序。另外附上2012年的數據，以
及兩個年度的差異。
http://snag.gy/rT0XG.jpg
If you’ve read about this research before, you won’t be surprised to see
the Brewers at the top of the list. Jonathan Lucroy appears to be one of the
game’s premier receivers, and when you combine that with his offense, he
seems like a borderline star player. Yadier Molina‘s good, Chris Stewart‘s
good, Russell Martin‘s good,Ryan Doumit‘s bad. This isn’t a strength of
the Marlins — few things are — and the Mets haven’t gotten a lot from John
Buck. Most teams are within 10 of league average.
It’s interesting when you look at the final column. The second-most-improved
team is the Tigers, at plus-eight. Then the Pirates are all the way up there
at plus-21, having signed Martin away from the Yankees in free agency even
though the Yankees didn’t have a sound backup plan. Martin has
singlehandedly turned this from a weakness into a strength. At the other end,
the Mets bring up the rear at minus-13. The Braves are at minus-12, a
consequence of missing Brian McCannand playing Gerald Laird. Atlanta also
lost David Ross. The Reds are at minus-11, and that does it for the
double-digits. You’re free to explore to your heart’s content.
如果你之前對這略有所知的話，釀酒人排第一應該不意外吧，Lucroy的接捕能力很好。
Yadier Molina不錯，Chris Stewart不錯，Russell Martin不錯，Ryan Doumit很遜。馬
林魚不怎樣，John Buck的大都會也不怎樣。看最後一欄，進步次多的是老虎，比去年多
了8球，進步最多的海盜比去年進步了21球，今年簽下的Russell Martin一手扭轉了海盜
的這數據。大都會比去年退步13球，勇士退步12球，因為Brian McCann缺席，Laird代打
一陣子，而且也少了David Ross。紅人退步11球。
Moving on, let’s look at some individual pitchers. So far, 136 pitchers have
thrown at least 1,000 pitches, and below, the top 10 and the bottom 10 in
Diff/1000.
Top Ten, Diff/1000
看一下投手吧。截至目前為止，有136個投手投超過1000球，以下是Diff/1000的前十名和
最後十名。
http://snag.gy/9UuHT.jpg
This should be a link to a spreadsheet for all 136. Unsurprisingly, a Brewer’
s in the lead; unsurprisingly, a Marlin is in last. If it makes you feel any
better, or worse, Koehler has a better ERA than does Estrada; the Marlins
have a better ERA than do the Brewers. Everything doesn’t come down to
pitch-receiving, but somestuff comes down to pitch-receiving. It’s possible
that without the catchers, the Marlins would look even better, and the
Brewers would look even worse. It’s all very complicated and we can’t speak
to a Lucroy-less Milwaukee pitching staff.
不意外，第一名是釀酒人投手，最後一名是馬林魚投手。不知道你會怎麼想，Koehler的
ERA比Estrada好，馬林魚團隊的ERA也比釀酒人好。接捕不是球賽最關鍵的元素，但或許
會改變某些情況。如果捕手不同，馬林魚可能會更好，釀酒人可能會更差。
主審對Estrada很大方
http://goo.gl/wvFmt
主審對Koehler很小氣
http://goo.gl/GFlKE
捕手的反應 XD
http://goo.gl/PaKWZ
That was in a 3-and-0 count, too, when umpires tend to be at their most
charitable. Koehler nailed his spot. It wasn’t a tricky pitch. The catcher
didn’t do anything particularly distracting. Ball four. All right. Now you
know this about Tom Koehler.
Above, you can see how Lucroy helped get the outside strike by making sure
his body momentum was leaning in toward the plate. Then you can see Rob
Brantlycatch a curveball all sloppy-like, driving the ball into the ground
even though it crossed the front plane in the strike zone. Granted,
curveballs aren’t easy to catch, but they should be easier than that.
What you can’t do with these numbers — which you can do with PITCHf/x — is
split them up by catcher. So simply based on this, we don’t know how much
credit should go to Lucroy or how much should go to Martin Maldonado. But you
can get good ideas, especially if you’re already familiar with
catcher-framing reputations that’ve developed in the past couple years. As
always, it’s hard to say how much all this matters, and it’s hard to
isolate the various contributing factors. But, still, it stands to reason you
’d rather have positive numbers than negative ones. For all the things the
Brewers have done wrong, they’ve done this right. And for all the things the
Marlins have done wrong, this is apparently among them.
和PITCHf/x相比，這一篇文章用的數據缺點就是沒辦法切出捕手個別的數據，所以釀酒人
的好表現，哪些是Lucroy的功勞哪些是Maldonado的功勞我們不知道。不過你可以得到一
個不錯的概念，尤其是如果你本來就對這幾年發展的接捕能力有點熟悉的話。真的很難說
我們討論的這些影響比賽多少，而且也很難把每一個變數都分開來探討，但是，這還是可
以推理出，你應該還是會比較想要有好的數據。釀酒人搞砸很多事情，但這一點他們做得
不錯。馬林魚搞砸很多事情，這是其中之一。