[外電] 暗號門與太空人驚人的數據(上)
原文: Does electronic sign stealing work? The Astros' numbers are eye-popping
來源: The Athletics
Their strikeout rate plummeted--at a level unparalleled in the last 100 years.
Their strikeout rate at home took an even more dramatic plunge — and that,
too, was unlike anything we’ve seen in the last century.
They developed an uncanny ability to lay off breaking balls below the strike
zone — an ability they hadn’t displayed before, and didn’t display on the
road. But at the same time, they began crushing every kind of pitch inside the
zone — at a rate that didn’t bear much resemblance to the way they’d handled
those very same pitches in the past.
他們的被三振率以百年來前所未有的幅度大幅下降，尤以在主場下降的幅度更為驚人。他們
展現出放掉好球帶下緣壞球的能力，同時將好球帶內的球打得更加結實，但在此之前，他們
從未以此見長。
These were your 2017 Houston Astros. Remember back, oh, a few months ago, when
we just referred to them as the World Series champs? Those were the days. Now
we look at them and ask: Were they really that good? How much did they owe to
pilfering signs and thumping on trash cans?
So here at The Athletic, we've been digging through the data, looking for
answers to those questions. We can say, with total assurance, that what we
found was intriguing. It strongly suggests that what they were doing worked.
這就是2017年的太空人。幾個月前人們還以世界冠軍稱呼他們，現在大家只想問，他們真的
有那麼強嗎？這當中有多少比例是來自太鼓達人與偷暗號？
The Athletics分析了一些數據並嘗試回答這個問題，結果非常耐人尋味，並強烈支持他們
的確以此獲得了一些優勢。
But here is what we can't say with any assurance: We can't say exactly how
well it worked — because we don't live in a What If world. We can't say the
Astros couldn't have won the World Series without it, because we can't
possibly know that. We can't say they wouldn't have dramatically improved
their contact rate if they'd never banged on those trash can lids — because
they did, in fact, bring in more contact-oriented hitters and did, in fact,
change their hitting philosophy heading into that season.
So bear that in mind as you read this. We're about to present a series of
numbers that look anything but normal, especially when you frame them in the
context of everyone and everything around them — and when you view them
through the prism of history. But we’re hesitant to jump to any firm
conclusions. So read them. Think about them. Decide for yourselves.
但我們無法確定影響有多大，因為我們無法觀察到另一個他們沒有作弊的平行時空，所以我
們不能很有把握地說太空人如果沒作弊就拿不到世界冠軍，這是不可能證明的。我們也不能
確定上述的現象完全是因為作弊，因為2017年他們確實補強了一些contact hitter，攻擊策
略也有所改變。
所以，在閱讀以下這些看起來非常不自然的數據的時候，請記得我們無法推論出100%的結論
。事實與真相為何，每個人心中自有答案。
Their plummeting strikeout rate 大幅下滑的三振率
A tale of two very different seasons: In 2016, Astros hitters struck out 1,452
times. That was the second-most in the American League, the fourth-most in
baseball and the eighth-highest total in the history of baseball. In 2017, they
set out to change that.
They changed it, all right.
They didn’t just strike out less — or even a lot less. They struck out so
much less, it’s fair to use the word “historic” to describe it.
2016年，太空人打者們被三振了1452次，這數字高居美聯第二、MLB第四，而且是聯盟史上
第八高的紀錄。2017年，他們誓言改變。
而他們真的做到了。
他們的被三振不只是少了一些，甚至只說少了很多都還客氣了。準確一點說，他們被三振率
的進步只能以「史無前例」來形容。
They cut their strikeouts by 365 whiffs — in one year. They went from punching
out 1,452 times in 2016 to a mere 1,087 in 2017, which meant they transformed
themselves from a team that was striking out at one of the highest rates in
history to a team that struck out less than any team in baseball that season.
And they made that transformation in just one year — not to mention a year
when the other 29 teams combined to strike out nearly 1,500 more times than they
whiffed the year before.
So how unusual is it for any team to cut down on their strikeouts by that much?
We asked STATS Perform to take a look. How many other teams in the entire
live-ball era (1920 to present) have ever reduced their team strikeout total by
365 in one year in non-strike-shortened seasons?
That would be none. In fact, no other team in history has even come within 80
strikeouts of doing that.
他們的被三振數在一年內整整減少了365次，從2016年的1452次降到了2017年的1087次。也
就是說他們從歷史上最容易被三振的球隊搖身一變成了全聯盟三振率最低的球隊。而其他29
隊的總三振數則比前一年多了1500次。
所以這有多不尋常呢？我們搜尋了從1920年至今的紀錄，想看看歷史上有多少球隊能在一年
內減少365次三振。
答案是……從來沒有。事實上，他們這個紀錄比第二名整整多了80次。
表一 https://imgur.com/y12LdbC
Of course, not all seasons in the live-ball era have been 162 games long. So
STATS also checked the biggest drops in strikeouts per game. The same five
teams led in that department, too. But STATS found that those 2016-17 Astros
were the only team in the last 100 seasons to cut their strikeout rate by more
than two strikeouts per game in one year. They went from 8.96 in 2016 to 6.71
in 2017 — a drop of 2.25 strikeouts per game.
當然，歷史上的球季並不一直都是162場，但以每場平均來看，太空人的紀錄依然相當驚人:
他們是百年來唯一一支在一年內每場平均三振數下降超過兩次的球隊(從8.96次下降到
6.71次)
But does that mean we can definitively say they made that dramatic a change
simply because they were stealing signs? We can’t. This was Alex Bregman's
first full season and Yuli Gurriel’s first full season. The Astros also
brought in Carlos Beltrán, Brian McCann, Josh Reddick and Nori Aoki that year.
All of those players had a track record of making more contact than the players
they replaced. Even preseason projections forecasted a significant decline in
their whiff rate.
但我們無法斷言這麼驚人的進步完全是因為偷暗號。這一年是Bregman和Gurriel的第一個完
整球季，打線中也新增了Beltran、McCann、Reddick，和青木宣青等人，這些人相較於他們
取代掉的球員都有更好的contact。季前預測其實就預期太空人的三振問題會有改善。
We also have no clear way to connect the dots between that dramatic increase in
contact rate and what sign-stealing techniques the Astros were using in any
specific game or games, how often they used those techniques, or how they
affected any specific player or players. Just this week, we did get a
game-by-game breakdown, from Astros fan Tony Adams, of how often they banged on
their trash-can lids in any given home game. And that at least gives us a
timeline that reveals how much that activity picked up between late May and
late September.
But remember, the report from the commissioner’s office this month says the
Astros also used other methods — including “clapping, whistling or yelling”
— to tip off hitters as to what pitch was coming. So it remains impossible to
measure how impactful every one of those techniques may have been. Others have
attempted to study whether it was possible to draw that connection, and they’
ve drawn conclusions that vary widely.
這一大段在講要準確估計偷暗號的影響有多大是不可能的…就不仔細翻了
We readily acknowledge that the most convenient explanation here isn’t the
only explanation. Nevertheless, when you view a historic change in teamwide
strikeout rates in the context of what the Astros have been accused of, and
punished for doing, it’s at least worth pointing out that their success in
cutting down on strikeouts was outside the realm of what’s considered “normal.
”
我們在此承認最直覺的解釋可能不是唯一的解釋。然而，將這個數字與暗號門事件放在一起
看，不得不說這麼大幅的三振下降率非常不自然。
Their eye-popping home/road splits 難以忽視的主客場差異
Let’s just accept the explanation that the Astros were able to cut their strike
outs because of an overhaul in their lineup and a more aggressive philosophy of
hitting. That's a perfectly logical reason for any team's strikeout decline.
But if that's all that was fueling this change, we have one big question:
Why did they strike out so much less at home than on the road?
讓我們姑且相信他們的三振率下降真的是因為打線重組和打擊策略改變吧，但如果是這樣的
話還有個大問題：
為甚麼他們在主場的三振比客場少這麼多？
表二 https://imgur.com/jarCoEO
In other words, they were able to cut down on their strikeouts both at home and
away from Houston. But if this was just about a team that decided it was time
to get more aggressive in swinging at strikes, why would they have been able to
reduce their strikeouts at home by 242 (nearly three fewer whiffs per game) but
only 123 on the road (or by about 1.5 per game)?
他們在主客場的被三振數都下降了，但如果這是打線和打擊策略改變的結果的話，為甚麼他
們在主場的被三振數下降了242次(將近每場下降3次)，但客場卻只下降了123次(大約每場下
降1.5次)?
STATS examined this question, too, in various ways. First off, this wasn’t
just the biggest year-to-year drop in strikeouts that STATS found in full
(non-strike) seasons. It was the largest in any length season in the live-ball
era. But when STATS looked only at full seasons, it found the Astros reduced
their strikeouts at home by 83 more K’s than the next-closest team in the last
100 years:
即使把罷工球季都納入計算，這數字也是史上最高的。但如果不看罷工球季的話，太空人主
場三振的下降次數比第二名還多了83次:
表三 https://imgur.com/Uv5XdHB
Or let’s look at this another way — by the decline in strikeouts per game,
which allows us to compare seasons of any length. Once again, STATS reported
that no other team in the live-ball era was able to cut its strikeout rate at
home by even close to this degree.
即使以每場平均來看，太空人依然穩居龍頭
表四 https://imgur.com/bJWvj9P
But STATS did uncover one area in which the Astros didn’t rank No. 1. It found
the Astros were one of 14 teams in the live-ball era in which there was at
least a 119-strikeout differential between the fluctuation in their home-park
strikeouts and their road strikeouts. (For the record, this was a study of only
teams with a bigger strikeout drop at home.)
However, the Astros were the only team on that list that significantly reduced
their strikeouts both at home and on the road. Just one other team cut its
whiffs in both places. And that was the 1961-62 Angels — a franchise in just
its second year of existence, and one that reduced its strikeouts on the road
by only 11.
So essentially, STATS found essentially no parallel between the 2017 Astros and
any other team in the live-ball era, in the ability to appreciably cut down on
strikeouts everywhere — but by a much larger margin at home. We’ve seen no
evidence, from the commissioner’s report, that the Astros were bringing their
trash cans with them on the road. Yet a different form of illegal
sign-stealing, using video feeds in the replay room, was available to the team
on the road as well as at home. Once again, however, the exact impact of any of
those methods is virtually impossible to measure.
總之，太空人的三振率在主客場都大幅下降，尤其在主場下降的幅度更是讓人驚異。由於在
客場也可能有種種的作弊方式，所以作弊帶來的確切影響仍然是難以準確估計的。
Their ability to stop chasing 他們不再追打壞球了
It sounds like such a simple goal: Let’s swing aggressively at the strikes and
stop chasing the pitches that aren’t strikes. It’s what every team aspires to
— but only the best teams do consistently.
It just happened to become one of the things the 2017 Astros did best.
We sifted through the Statcast data, and here is some of what we found:
「積極攻擊好球帶內的球，並不要追打壞球」。聽起來簡單的目標，但只有最好的球隊才能
穩定地做到這點。而這剛好是2017年的太空人做得最好的一件事。
我們從資料中發現以下事實:
‧In 2016, the Astros swung at 36 percent of all the breaking balls they saw
at home that were below the bottom of the strike zone. In 2017, they
dramatically improved that rate — by chasing only 27 percent of those same
pitches.
2016年，太空人在主場追打了36%好球帶下緣的壞球，而2017年，他們戲劇性地進步了: 只
追打了27%好球帶下緣的壞球。
‧On the road, they were actually better, though they didn’t improve as much
compared with the year before. In 2016, they chased 27 percent of those
breaking balls below the zone away from home. In 2017, they pared that rate
down to 25 percent.
在客場，他們其實表現得更好，但進步幅度不如主場來的大。2016年，他們在客場追打了
27%好球帶下緣的壞球，而2017年則進步到25%
‧ Four of their returning hitters made remarkable improvements in their
ability to lay off those pitches. Marwin González went, incredibly, from a
40-percent swing rate to just 10 percent … Carlos Correa improved from 27
percent to 15 percent … Jake Marisnick dropped from 30 percent to 13 percent …
George Springer went from 18 percent to 12 percent. We’ll mention here that
according to Tony Adams’ work at signstealingscandal.com, those four players
were among the Astros who were at bat for the highest percentage of trash-can
bangs that season. And there were more of those bangs while González was
hitting (147) than any other hitter. But this is just one more occasion where we
’ll refrain from drawing any firm conclusions from those seemingly related
facts.
2016就在太空人的球員中，其中4人的進步格外顯著:
Marwin González : 從40%的追打率降到10%
Carlos Correa: 從27%降到15%
Jake Marisnick: 從30%降到13%
George Springer: 從18%降到12%
同時，在之前美國鄉民整理的資料中，這4人也是打擊前Bang聲最多的其中幾人
‧ We also analyzed the data from signstealingscandal.com to see how the Astros
’ “Reach Rate” (i.e, balls swung at that were definitively outside the
strike zone) changed at home after the trash-can banging began in late May. As
a team, the Astros reduced their rate of pitch-chasing — from more than 25
percent before the banging to under 20 percent afterward. Obviously, not all
players used the can-banging assistance the same way, and
signstealingscandal.com‘s research indicated some players barely used it at
all. And that usage also varied from game to game, as the chart below —
breaking down that Reach Rate by game — would indicate.
我們又比較了在Bang聲在五月底出現之前跟出現之後的資料。我們發現在Bang聲出現前，追
打率超過25%，而出現後，追打率下降到低於20%
Overall, the Astros improved their knack for laying off non-strikes during the
period in which the banging scheme was in effect. The numbers show that. Let’s
say this again, though: Whether that’s a clear cause and effect is impossible
to determine with any certainty.
總之，在Bang聲期間太空人的追打壞球率顯然較低，但我們仍然不能肯定地說Bang聲一定就
是造成差異的原因。
Their ability to crush the strikes 他們把好球打得更遠
Of course, the best hitters in the game do more than merely lay off the pitches
that aren’t strikes. What really matters is how much damage they inflict on
the pitches that are strikes. And Statcast data tells us the 2017 Astros
improved their ability to unleash that damage more than any other offense in
the game.
當然了，只是能放掉壞球還稱不上是好的打者，重點還是在攻擊好球帶內的球並對對手造成
傷害。而數據顯示2017年的太空人在此能力上的進步再次高居榜首。
‧ In 2016, the Astros slugged only .438 on breaking balls inside the zone at
home. In 2017, they slugged .508 at home on those same pitches.
2016年，太空人在主場隊好球帶內的變化球長打率是 .438。2017年則進步到了 .508
‧ But the Astros made just as big an improvement on fastballs. In 2016, they
slugged .503 on fastballs inside the zone at home. The next year, they spiked
that slugging percentage, on the same pitches, by 74 points — to .577.
他們對好球帶內速球的攻擊進步幅度也一樣大: 長打率從2016的 .503進步到2017的 .577
---- 進步了7.4個百分點
‧ So how many other teams slugged at least 70 points better at home that year
against both fastballs and breaking balls, compared with the year before? Yes,
that would be none.
所以2017年還有多少其他隊伍對好球帶內的速球與變化球的長打率都比前一年提高7個百分
點以上的? 沒錯，完全沒有。
‧ We should note that the Astros also made big jumps in games away from
Houston. Against breaking balls in the strike zone on the road, they went from
a .461 slugging percentage to .574. On fastballs, they went from .540 to .602.
他們在客場對好球帶內的球的攻擊能力進步幅度同樣驚人。變化球的長打率從 .461提升到
.574 ; 速球則從 .540進步到 .602
‧ Overall, their team slugging percentage against breaking balls in the strike
zone — home and away — jumped by nearly 100 points, from .450 to .544.
Curious how many other teams improved their slugging against breaking balls
that much in 2017? Once again, it was zero. And as this chart of the four
seasons from 2015-18 shows, the Astros’ slugging against those same pitches
didn’t reach that level in any other season.
加總來看，太空人在2017年對好球帶內變化球的長打率從 .450提高到了 .544 ---- 將近提
升了10個百分點。好奇還有其他隊進步這麼多的嗎? 答案依然是沒有。而在2015 – 2018年
間，太空人從來沒有一年像2017一樣這麼有效地攻擊這類球路。
Again, don’t lose sight of the fact that this was a team that upgraded its
lineup significantly that season — and specifically emphasized this exact
approach, to be more aggressive and do more damage on pitches in the zone. But
there is also no disputing that no team improved its ability to execute that
approach that season. How much did it help that they often knew what pitch was
coming? That’s something Statcast can’t tell us.
我們還是得再次提醒讀者，別忘了2017年太空人的打線相對2016有顯著的補強，而他們也特
別強調要改變打擊策略。但無可爭論的是，在2017年沒有其他任何一支球隊在這項能力上進
步這麼多。偷暗號對他們的幫助究竟有多少? 這是數據沒辦法告訴我們的。
--
※ 發信站: 批踢踢實業坊(ptt.cc), 來自: 24.99.0.224 (美國)
※ 文章網址: https://www.ptt.cc/bbs/MLB/M.1581563349.A.6D5.html
推
02/13 11:11,
5年前
, 1F
02/13 11:11, 1F
文章後面還整理了一些個別球員的數據, 包括Springer, Correa, Gattis, Marisnick,
和Altuve。有空的話再補完
※ 編輯: ianshadow (24.99.0.224 美國), 02/13/2020 11:13:41
推
02/13 11:18,
5年前
, 2F
02/13 11:18, 2F
推
02/13 11:18,
5年前
, 3F
02/13 11:18, 3F
推
02/13 11:29,
5年前
, 4F
02/13 11:29, 4F
推
02/13 11:38,
5年前
, 5F
02/13 11:38, 5F
→
02/13 11:39,
5年前
, 6F
02/13 11:39, 6F
推
02/13 11:48,
5年前
, 7F
02/13 11:48, 7F
推
02/13 11:49,
5年前
, 8F
02/13 11:49, 8F
※ 編輯: ianshadow (24.99.0.224 美國), 02/13/2020 11:52:11
推
02/13 11:50,
5年前
, 9F
02/13 11:50, 9F
推
02/13 11:50,
5年前
, 10F
02/13 11:50, 10F
→
02/13 11:51,
5年前
, 11F
02/13 11:51, 11F
推
02/13 11:52,
5年前
, 12F
02/13 11:52, 12F
推
02/13 11:57,
5年前
, 13F
02/13 11:57, 13F
→
02/13 11:58,
5年前
, 14F
02/13 11:58, 14F
推
02/13 11:58,
5年前
, 15F
02/13 11:58, 15F
**應madboy版友要求修掉推文**
推
02/13 12:06,
5年前
, 16F
02/13 12:06, 16F
推
02/13 12:16,
5年前
, 17F
02/13 12:16, 17F
→
02/13 12:16,
5年前
, 18F
02/13 12:16, 18F
推
02/13 12:29,
5年前
, 19F
02/13 12:29, 19F
推
02/13 12:40,
5年前
, 20F
02/13 12:40, 20F
推
02/13 12:45,
5年前
, 21F
02/13 12:45, 21F
→
02/13 12:48,
5年前
, 22F
02/13 12:48, 22F
→
02/13 12:48,
5年前
, 23F
02/13 12:48, 23F
→
02/13 12:49,
5年前
, 24F
02/13 12:49, 24F
推
02/13 12:51,
5年前
, 25F
02/13 12:51, 25F
其實各種數據來看，是不支持阿土伯有參與作弊的喔
文章之後列出的個人數據裡面有包括阿土伯，是為了說明從數字來看阿土伯不像是有參與
推
02/13 12:51,
5年前
, 26F
02/13 12:51, 26F
→
02/13 12:52,
5年前
, 27F
02/13 12:52, 27F
推
02/13 12:55,
5年前
, 28F
02/13 12:55, 28F
→
02/13 12:57,
5年前
, 29F
02/13 12:57, 29F
→
02/13 12:58,
5年前
, 30F
02/13 12:58, 30F
※ 編輯: ianshadow (24.99.0.224 美國), 02/13/2020 13:02:56
推
02/13 13:01,
5年前
, 31F
02/13 13:01, 31F
推
02/13 13:03,
5年前
, 32F
02/13 13:03, 32F
推
02/13 13:05,
5年前
, 33F
02/13 13:05, 33F
→
02/13 13:06,
5年前
, 34F
02/13 13:06, 34F
還有 81 則推文
還有 5 段內文
→
02/14 12:36,
5年前
, 116F
02/14 12:36, 116F
→
02/14 12:37,
5年前
, 117F
02/14 12:37, 117F
→
02/14 12:37,
5年前
, 118F
02/14 12:37, 118F
→
02/14 12:39,
5年前
, 119F
02/14 12:39, 119F
→
02/14 12:39,
5年前
, 120F
02/14 12:39, 120F
推
02/14 12:52,
5年前
, 121F
02/14 12:52, 121F
→
02/14 12:56,
5年前
, 122F
02/14 12:56, 122F
→
02/14 12:57,
5年前
, 123F
02/14 12:57, 123F
→
02/14 12:57,
5年前
, 124F
02/14 12:57, 124F
→
02/14 12:57,
5年前
, 125F
02/14 12:57, 125F
推
02/14 14:33,
5年前
, 126F
02/14 14:33, 126F
→
02/14 14:33,
5年前
, 127F
02/14 14:33, 127F
→
02/14 14:33,
5年前
, 128F
02/14 14:33, 128F
→
02/14 14:33,
5年前
, 129F
02/14 14:33, 129F
罰款罰選秀非常合理，因為數字上看起來太空人確實有作弊
但Fiers也說了不是全隊都有參與
而支持阿土伯有參與的證據一個都沒有
所有目前出來的證據都是比較支持他沒有參與的
選擇不聽不看，反正說他有作弊就能大嘴特嘴說洋基本來根本不會被打那支再見HR
一口咬定他「一定」有作弊，這確實就是隨意誹謗
然後一直說只要有一次bang聲就是作弊的
是不是忘了bang聲的統計是球迷看影片一球一球聽聲音整理出來的?
這麼多打數，偶爾有幾球前有敲擊聲，你確定那一定是敲垃圾桶的聲音?
有些打者出現敲擊的次數太頻繁了，球場噪音不可能那麼巧每次擊球前都出現
這時候我們才認為這個打者很可疑
就像你要抓學生是否有作弊偷看旁邊的答案(假設兩人考的是不同考卷)
50題選擇題裡面有1題答案一樣，你會說這就是他有作弊的證據嗎?
不可能，一定是太多題答案都一樣，沒作弊剛好發生這種情形的可能性太低
才構成支持他有作弊的證據
這道理是一樣的
那個統計不能提供阿土伯有沒有用其他方法作弊的資訊
但是絕對能證明他沒有用垃圾桶敲擊作弊
再說一次
bang聲統計、聯盟調查結果、與2016成績比較，都是比較支持他沒作弊
支持他有作弊的客觀證據 一個都沒有
但如果不聽不看，覺得我說他有就是有，那當然也是個人自由
推
02/14 14:55,
5年前
, 130F
02/14 14:55, 130F
→
02/14 14:55,
5年前
, 131F
02/14 14:55, 131F
→
02/14 14:55,
5年前
, 132F
02/14 14:55, 132F
→
02/14 14:55,
5年前
, 133F
02/14 14:55, 133F
推
02/14 15:02,
5年前
, 134F
02/14 15:02, 134F
※ 編輯: ianshadow (24.99.0.224 美國), 02/14/2020 18:54:16
※ 編輯: ianshadow (24.99.0.224 美國), 02/14/2020 19:02:45
→
02/14 20:18,
5年前
, 135F
02/14 20:18, 135F
→
02/15 04:55,
5年前
, 136F
02/15 04:55, 136F
→
02/15 04:55,
5年前
, 137F
02/15 04:55, 137F
→
02/15 04:55,
5年前
, 138F
02/15 04:55, 138F
→
02/15 04:55,
5年前
, 139F
02/15 04:55, 139F
→
02/15 04:55,
5年前
, 140F
02/15 04:55, 140F
→
02/15 04:55,
5年前
, 141F
02/15 04:55, 141F
→
02/15 04:55,
5年前
, 142F
02/15 04:55, 142F
→
02/15 04:55,
5年前
, 143F
02/15 04:55, 143F
→
02/15 04:55,
5年前
, 144F
02/15 04:55, 144F
推
02/15 08:51,
5年前
, 145F
02/15 08:51, 145F
→
02/15 08:52,
5年前
, 146F
02/15 08:52, 146F
→
02/15 08:53,
5年前
, 147F
02/15 08:53, 147F
→
02/15 08:54,
5年前
, 148F
02/15 08:54, 148F
→
02/15 08:55,
5年前
, 149F
02/15 08:55, 149F
→
02/15 08:55,
5年前
, 150F
02/15 08:55, 150F
→
02/15 09:26,
5年前
, 151F
02/15 09:26, 151F
推
02/15 19:51,
5年前
, 152F
02/15 19:51, 152F
→
02/15 19:52,
5年前
, 153F
02/15 19:52, 153F