ESPN- The Great Analytics Rankings... The Twins are not so Great

[jokin]

Interesting and comprehensive study of all 122 teams in the four major team sports, ESPN  "unleashed our experts and an army of researchers to rate 122 teams on the strength of each franchise's analytics staff, its buy-in from execs and coaches, its investment in biometric data and how much its approach is predicated on analytics."

 

No surprise here.... The Twins didn't fare too well in the analysis-  labeled as "skeptics" (seems about right to me, the Twins were whacked on lack of Ks, but they really were just as vulnerable to an attack on the planned 2015 OF, but ESPN let them off easy in that case), but hey, at least they were not in the Bottom Ten.

 

Here's the overall baseball-only list:

 

ALL-IN
Boston Red Sox
Chicago Cubs
Cleveland Indians
Houston Astros
New York Yankees
Oakland A's
Pittsburgh Pirates
St. Louis Cardinals
Tampa Bay Rays
 

BELIEVERS
Baltimore Orioles
Kansas City Royals
Los Angeles Dodgers
New York Mets
San Diego Padres
Toronto Blue Jays
Washington Nationals
 

ONE FOOT IN
Chicago White Sox
Los Angeles Angels
Milwaukee Brewers
San Francisco Giants
Seattle Mariners
Texas Rangers
 

SKEPTICS
Arizona Diamondbacks
Atlanta Braves
Cincinnati Reds
Colorado Rockies
Detroit Tigers
Minnesota Twins
 

NONBELIEVERS
Miami Marlins
Philadelphia Phillies

 

 

Here are the baseball teams in the overall Top Ten ( some surprises):

 

# 2 Houston

# 4 Tampa Bay

# 5 Boston

# 6 New York Yankees

# 9 Oakland

 

Here are the baseball teams in the overall Bottom Ten:

 

# 115 Miami

# 122 Philadelphia

 

 

 

 

 

 

 

 

 

 

Ryan is conversant on sabermetric talking points like batting average on balls in play (BABIP), and now says he never makes a move without consulting with Jack Goin, his manager of baseball research who earned an MBA from the local University of St. Thomas. While the Twins are widely regarded as one of the least sabermetrically inclined teams, Goin says the team is more middle of the road.

 

Between the lines, the signals are mixed at best: The Twins employed defensive shifts more than the average team in 2014, increasing their usage fivefold. But their pitching staff has the lowest strikeout percentage over the past four seasons -- and the highest ERA (outside of Colorado) -- showing an overemphasis on "pitching to contact."                                                                                                                                                                                                                                                                            

New manager Paul Molitor will bring more openness to analytical discussions, but with Ryan keeping sabermetrics at arm's length, this crew has a long way to go to catch its division rivals in Cleveland and Kansas City.

[gunnarthor]

Here's the link: http://espn.go.com/espn/feature/story/_/id/12331388/the-great-analytics-rankings#!mlb

 

And the other article trying to determine if stats = wins http://espn.go.com/blog/sweetspot/post/_/id/55317/checking-out-the-mags-analytic-rankings

 

Interesting that the Royals, with Dayton Moore and Ned Yost narratives, are listed as believers.

[Parker Hageman]

On the surface, feels like a very non-scientific study. We will never truly know the depth of any team's use of analytics unless they are like the Astros that love to discuss their department/efforts. That being said, I'm not sure if the rankings are wrong, either. 

 

I've said this regularly but I do feel like there is some progress in this area locally with the Twins. Consider these three conversations:

 

Rob Antony, March 2010: http://www.startribune.com/sports/twins/blogs/88887222.html

 

Jack Goin, January 2013: http://twinsdaily.com/articles.html/_/minnesota-twins-news/minnesota-twins/where-are-the-twins-at-with-statistical-analysis-r1093

 

Terry Ryan and Jack Goin, October 2014: http://twinsdaily.com/_/minnesota-twins-news/twins-research-team-wont-tip-their-pitches-r3131

 

I don't think we'll ever see a situation where this organization turns into a industry leader but it feels like they are doing things to catch up. 

[70charger]

Doesn't this seem like a trailing statistic at best? For example, the Twins' homegrown pitchers that are low-K guys (and I'm including the low-K guys of the last couple of years) were drafted under a different GM regime, and a different minor league coordinator, and probably with different scouts. There's also a big element of luck here too.

 

It seems hard to ding the current Twins on non-K pitchers when the free agents who rack up the major strikeout numbers aren't in a realistic price range (and would be a terrible investment according to advanced statisticians themselves!), and the recent draftees have been anything but high-contact soft tossers. They just haven't hit the majors (yet).

 

Baseball isn't basketball. You can't draft a LeBron and, voila, make the playoffs. I doubt the value of this study insofar as it regards baseball teams, and insofar as it measures current on-field personnel instead of overall strategy and/or entire-system makeup.

[jimmer]

I thought Arizona would be in the non-believers considering they were apparently the last to even have any kind of analytics person and considering Dave Stewart's comments on the subject when they were pursuing Shields.  I actually think that strategy backfired on him.

[nicksaviking]

Doesn't this seem like a trailing statistic at best? For example, the Twins' homegrown pitchers that are low-K guys (and I'm including the low-K guys of the last couple of years) were drafted under a different GM regime, and a different minor league coordinator, and probably with different scouts. There's also a big element of luck here too.

 

It seems hard to ding the current Twins on non-K pitchers when the free agents who rack up the major strikeout numbers aren't in a realistic price range (and would be a terrible investment according to advanced statisticians themselves!), and the recent draftees have been anything but high-contact soft tossers. They just haven't hit the majors (yet).

 

Baseball isn't basketball. You can't draft a LeBron and, voila, make the playoffs. I doubt the value of this study insofar as it regards baseball teams, and insofar as it measures current on-field personnel instead of overall strategy and/or entire-system makeup.

 

I don't think there's all that much luck to it, the Twins have been among the worst at K's for a long time, and it wasn't just under Smith.  Johan Santana was a godsend and an anomoly, but this is the same club that was trying to win championships with Radke, Milton, Mays, Redmond and Reed. 

 

The philosophy clearly has changed, but they did seem to be amongst the last teams to see the value in the strikeout, which is odd considering how much they seemed to dislike batters who wiffed too much. 

 

I will concede that there is the possiblity that the problem wasn't that they didn't look hard enough for strikeout pitchers, but instead just were not able to identify them in the draft or develop them to become effective MLBers.  However, if that was/is the issue, this club has even bigger problems and needs to bring in different help.

[nicksaviking]

I thought Arizona would be in the non-believers considering they were apparently the last to even have any kind of analytics person and considering Dave Stewart's comments on the subject when they were pursuing Shields.  I actually think that strategy backfired on him.

 

Yeah, they also got rid of young players with upside in exchange for Kirk Gibson's preferred "gritty" players.  I'm guessing ESPN is taking into consideration the fact that they got rid of those philosophies as a step in the right direciton.

[Parker Hageman]

The DBacks are certainly interesting when it comes to the use of analytics. Here is what Stewart and LaRussa said about them in November 2014:

 

 

The numbers will have their place. So, too, will La Russa and Stewart and Dave Duncan (who was out in front of some of this stuff a long time ago, but didn't give it a name) and the new manager, Chip Hale. While the trend is running toward increased statistical analysis, certainly in the general manager's chair, Stewart is asked about his familiarity with that end of things and he says, "I want to know about it and I'm going to educate myself in it."

That said, he added, "We're not going to be an organization that's going to [run on] 70 percent metrics. That's not going to happen."

Said La Russa: "We'll use it. It stops before the first pitch is thrown."

Beyond that, he said, "It's not possible. The game is too dynamic. Men versus men. It changes every day."

It makes sense. In-game decisions, he said, fall under the category of preparation, not analytics.

 

"It's not that we devalue it," La Russa said. "We value it when it's used appropriately. We do not value its intrusion into the game. You've got to allow your uniformed people to observe the dynamics of the competition."

 

Then, of course, Stewart said this in January 2015:

 

 

“I think James is a throwback guy by the way he goes about his business and the innings he pitches,” Stewart said. “I think the fact that Tony (La Russa) is here and that we have more baseball people – he probably sees us as a true baseball team vs. some of the other teams out here that are geared more toward analytics and those type of things.

 

“Sometimes, there are concessions the player will make to be here. It’s the case that he likes what we’re doing with our organization from our end, all we can hope is that there will be concessions enough that he can be here.”

 

LaRussa is undoubtedly one of the smartest guys in the game and he did rely on metrics and analysis that he and Oakland organization flourished with in the late 1980s/early 1990s but at the end of the day, he's right -- for the most part metrics flies out the window once the game starts. 

[jimmer]

'LaRussa is undoubtedly one of the smartest guys in the game and he did rely on metrics and analysis that he and Oakland organization flourished with in the late 1980s/early 1990s'.

 

Were the As even using metrics that much in the 90s?  Didn't they really just grab on in the late 90s/early 00s?

 

And I'm not sure they should fly out the window once the game starts unless we don't want to do defensive shifting or any metrics when determining offensive in-game strategy?  Even if you use it to prepare (pre-game), so the work is done, it's still being applied in the game, one hopes.  If it isn't, what's the point?

[Otto von Ballpark]

Doesn't this seem like a trailing statistic at best? For example, the Twins' homegrown pitchers that are low-K guys (and I'm including the low-K guys of the last couple of years) were drafted under a different GM regime, and a different minor league coordinator, and probably with different scouts. There's also a big element of luck here too.

 

It seems hard to ding the current Twins on non-K pitchers when the free agents who rack up the major strikeout numbers aren't in a realistic price range (and would be a terrible investment according to advanced statisticians themselves!), and the recent draftees have been anything but high-contact soft tossers. They just haven't hit the majors (yet).

 

Baseball isn't basketball. You can't draft a LeBron and, voila, make the playoffs. I doubt the value of this study insofar as it regards baseball teams, and insofar as it measures current on-field personnel instead of overall strategy and/or entire-system makeup.

I hear you -- but isn't trailing going to be a permanent characteristic of the skeptic team? Right now it is K rate, in a few years it could be something else that we are playing catch-up in.

[Parker Hageman]

Were the As even using metrics that much in the 90s?  Didn't they really just grab on in the late 90s/early 00s?

 

 

The A's were frontrunners when it came to technology. Former manager Steve Boros would consult his Apple II for information. 

 

http://upload.wikimedia.org/wikipedia/en/0/0c/Steve_Boros_and_His_Apple_II.png

 

From Wikipedia:

 

During the 1983 season, Boros gained note for his pioneering use of "sabermetrics" and a computer to help guide his managerial decisions. In his book on baseball innovations, historian Peter Morris wrote that Boros was "the first manager to make extensive use of computers in his decision-making process."After every game, Boros had pitch-by-pitch data fed into a mainframe computer at Digital Equipment Corporation in Philadelphia. He would retrieve the data before the next game, analyzing how each pitcher and hitter matched up.

 

Jay Alves, a "sabermetrician" hired by Boros to run the computer system,[56] later recalled: "With Steve, we tracked (on paper) virtually every pitch and where it was hit. Then I'd type it into the computer after the game, but that would take a couple of hours. That's how long it takes me to do it with pencil and paper."

 

 

 

And Sandy Alderson in that era (from a Mike Berardino Baseball America article from 2003):

 

 

 

What went into the Total Offensive Production (TOP) figure Beane inherited from A’s predecessor Sandy Alderson?

 

"That wasn’t even ours," Beane says. "That came from someone we used to employ a number of years ago."

 

OK, so what went into TOP, a statistical tool that Alderson began using in the mid-1980s? This tool once led Alderson to dump Dave Kingman and his .255 OBP for an aging Reggie Jackson and snare the likes of Doug Jennings (Rule 5 draft) and Geronimo Berroa (six-year free agency) when others wrote them off.

 

 

Also from that same article, a great quote from a former DBacks GM on statistics:

 

 

"Maybe I’m hopelessly old school in this regard, but to me statistics that you can derive from sort of the basic building blocks I think have real value," Garagiola says. "I look at ‘Baseball Prospectus’ from time to time, and some of those stats are so arcane, dense, impenetrable–whatever the word you want to use is.

 

"I guess this is meaningful to somebody, but not me. It drills down so deeply, it’s like, ‘OK, when you hit the bottom, there are three people in the world that this matters to.’ "

[jimmer]

GREAT INFO, THANKS!

[nicksaviking]

 

"I guess this is meaningful to somebody, but not me. It drills down so deeply, it’s like, ‘OK, when you hit the bottom, there are three people in the world that this matters to.’ "

 

All of whom are Twins Daily readers I guess.

[markos]

On the surface, feels like a very non-scientific study. We will never truly know the depth of any team's use of analytics unless they are like the Astros that love to discuss their department/efforts. That being said, I'm not sure if the rankings are wrong, either. 

 

I agree with this. I think the overall trend-line is correct, though enough teams are secret about their work that it wouldn't surprise me if there are teams that are severely misplaced. It (necessarily) skews heavily on visible features: quantity of job-titles, in-game decision making (bunting, defensive shifts, etc), and perceived relationship between front-office and managers. Those are important factors, but I think the real value lies in calculating player value and making future projections. And most teams are pretty quiet about their proprietary projection models. It wouldn't surprise me if in 5-10 years we find out that one of the successful teams near the bottom of this list (maybe the Giants?) had been quietly utilizing really good models for player projections for years. 

[jokin]

The A's were frontrunners when it came to technology. Former manager Steve Boros would consult his Apple II for information. 

 

http://upload.wikimedia.org/wikipedia/en/0/0c/Steve_Boros_and_His_Apple_II.png

 

From Wikipedia:

 

 

And Sandy Alderson in that era (from a Mike Berardino Baseball America article from 2003):

 

 

 

What went into the Total Offensive Production (TOP) figure Beane inherited from A’s predecessor Sandy Alderson?

 

"That wasn’t even ours," Beane says. "That came from someone we used to employ a number of years ago."

 

OK, so what went into TOP, a statistical tool that Alderson began using in the mid-1980s? This tool once led Alderson to dump Dave Kingman and his .255 OBP for an aging Reggie Jackson and snare the likes of Doug Jennings (Rule 5 draft) and Geronimo Berroa (six-year free agency) when others wrote them off.

 

 

Also from that same article, a great quote from a former DBacks GM on statistics:

 

 

"Maybe I’m hopelessly old school in this regard, but to me statistics that you can derive from sort of the basic building blocks I think have real value," Garagiola says. "I look at ‘Baseball Prospectus’ from time to time, and some of those stats are so arcane, dense, impenetrable–whatever the word you want to use is.

 

"I guess this is meaningful to somebody, but not me. It drills down so deeply, it’s like, ‘OK, when you hit the bottom, there are three people in the world that this matters to.’ "

 

Thanks for doing all the work, so I didn't have to...

[tobi0040]

I think the ranking probably has the Twins rated right in my opinion.  To my knowledge, our analytics department is one person, who appears relatively young and has an MBA versus an advanced statistics degree and/or outside application.  I believe he was with the Twins in ticket sales prior to going back to school.  I don't mean any disrespect.  But I am 31 and have an economics degree and an MBA and I would feel vastly out gunned relative to other teams.

 

Other teams have Phd's, quants from NASA, statisticians, much larger departments, guys plucked from other organizations that have good reputations, etc.

[bird]

I think there was something written today about how Royals fans have had this misperception that KC is "out of date" regarding advanced metrics. I'd guess this type of misperception would be found in almost every market to some degree. The opinions are likely to be expressed by fans in every market on sites like this, where a good share of commenters are going to be huge believers in the benefits of using the statistics they so fervently enjoy.

 

Just to be clear, like most of you, I believe Ryan and the Twins were not only late to the party, but continue to be pretty tepid about the process of incorporating systems to take advantage of the new information made available to them. That said, I'd probably place a lot less relative importance and value on it than most, perhaps due to my own ignorance.

[Mike Sixel]

How do you show someone how to win? Isn't talent the key? Mostly?

 

We clearly have no idea what team is using what most......but given the knowledge we have, it seems reasonable to say this is probably correct. 

 

Though, I agree, the Diamondbacks need to drop.....

[JB_Iowa]

I think the ranking probably has the Twins rated right in my opinion.  To my knowledge, our analytics department is one person, who appears relatively young and has an MBA versus an advanced statistics degree and/or outside application.  

It is defintely more than one person because Jack Goin, the Manager, Major League Administration and Baseball Research, was hiring for at least one  position last summer.  The ad is on a thread somewhere.

 

How many besides Jack, the new hire, and an intern is anybody's guess.

 

I don't dispute the ranking, though.

 

Edit:

 

A little insight from Parker:

 

http://twinsdaily.com/topic/15593-article-twins-research-team-wont-tip-their-pitches/?hl=%2Bjack+%2Bgoin

 

http://twinsdaily.com/topic/13852-article-where-are-the-twins-at-with-statistical-analysis/?hl=%2Bjack+%2Bgoin

 

And here's the thread on the job opening:

 http://twinsdaily.com/topic/11577-jack-goin-discusses-twins-job-opening-etc-was-all-right-all-you-geniuses/?hl=%2Bjack+%2Bgoin

 

and the fangraphs ad:

http://www.fangraphs.com/blogs/job-posting-twins-developer/

[nicksaviking]

How do you show someone how to win?..

I'd agree you probably can't teach guys how to win, I'd guess the idea is to teach guys the work habits and mind sets of winners.

 

Even if it ultimately proved futile, I'm all for the effort, especially for the young guys with talent.

[jimmer]

How do you show someone how to win?

 

 

Just watch some sports movies, you'll find out :-)

[jokin]

It is defintely more than one person because Jack Goin, the Manager, Major League Administration and Baseball Research, was hiring for at least one  position last summer.  The ad is on a thread somewhere.

 

How many besides Jack, the new hire, and an intern is anybody's guess.

 

I don't dispute the ranking, though.

 

Edit:

 

A little insight from Parker:

 

http://twinsdaily.com/topic/15593-article-twins-research-team-wont-tip-their-pitches/?hl=%2Bjack+%2Bgoin

 

http://twinsdaily.com/topic/13852-article-where-are-the-twins-at-with-statistical-analysis/?hl=%2Bjack+%2Bgoin

 

And here's the thread on the job opening:

 http://twinsdaily.com/topic/11577-jack-goin-discusses-twins-job-opening-etc-was-all-right-all-you-geniuses/?hl=%2Bjack+%2Bgoin

 

and the fangraphs ad:

http://www.fangraphs.com/blogs/job-posting-twins-developer/

 

But how do we know if the position was in fact, filled? Not like there's a ton of transparency or "pitch-tipping" concerning developments of this Twins strategic division of their operation.

[JB_Iowa]

But how do we know if the position was in fact, filled? Not like there's a ton of transparency or "pitch-tipping" concerning developments of this Twins strategic division of their operation.

 

 Next time Jack is around, ask him.

[70charger]

I hear you -- but isn't trailing going to be a permanent characteristic of the skeptic team? Right now it is K rate, in a few years it could be something else that we are playing catch-up in.

 

Don't take this as an attack, but this kind of argument presumes the sufficiency of the statement that the Twins are skeptics. I.e., as long as a team is skeptical, there will be something that they're catching up at.

 

But what if the perception of skepticism is incorrect? What if the Twins have changed over the last several years? My point is that it will take much, much longer for that to show up on the field than it would in pretty much any other sport. Which is to say, looking at the Twins' roster right now doesn't tell you as much as you'd think about the Twins right now. It tells you about the Twins of five-ish years ago.

 

That statement does, by the way, leave open the possibility that the Twins haven't changed. But we can't properly argue from the conclusion.

[jokin]

Don't take this as an attack, but this kind of argument presumes the sufficiency of the statement that the Twins are skeptics. I.e., as long as a team is skeptical, there will be something that they're catching up at.

 

But what if the perception of skepticism is incorrect? What if the Twins have changed over the last several years? My point is that it will take much, much longer for that to show up on the field than it would in pretty much any other sport. Which is to say, looking at the Twins' roster right now doesn't tell you as much as you'd think about the Twins right now. It tells you about the Twins of five-ish years ago.

 

That statement does, by the way, leave open the possibility that the Twins haven't changed. But we can't properly argue from the conclusion.

 

If the musical chairs in CF in 2014, ending up with a SS playing the position isn't enough evidence, the Terry Ryan press conference in welcoming back Torii Hunter and explaining how he analyzes defense should tell you all you need to know about the Twins and their "skeptical" view of the likely prospect of the Twins having the MLB's worst OF defense in 2015.

[bird]

I'm sorry, I don't want to start a squabble here, but can't we move on from taking Ryan's comments at Torii's presser out of context and using them as some sort of definitive statement about his ignorance or disbelief in our beloved statistics? This doesn't help us get to the truth about what's going on with the team's use of advanced metics. It only serves to (I think) unfairly castigate Ryan, especially when we refuse to also discuss his very informed and valid observations on the subject.

[70charger]

Now you're changing the conversation to centerfield defense. Fine. Even as far as that's concerned, I think their plan A, Aaron Hicks, is pretty good as far as defense goes. When he didn't hit, they got caught short a fielder, but I just don't see how that has to do with belief in math.

 

Again, I'm fully willing to leave open the possibility that the Twins lag the rest of the world in stats. What I'm attempting to do is ask whether the measurements taken by this particular study are useful in proving that point.

[tobi0040]

I'm sorry, I don't want to start a squabble here, but can't we move on from taking Ryan's comments at Torii's presser out of context and using them as some sort of definitive statement about his ignorance or disbelief in our beloved statistics? This doesn't help us get to the truth about what's going on with the team's use of advanced metics. It only serves to (I think) unfairly castigate Ryan, especially when we refuse to also discuss his very informed and valid observations on the subject.

 

Here are his actual comments, then the UZR data.  I see a clear disconnect between Terry's comments and Torii's UZR ratings.   "More than adequete" does not equal dead last in the league to me.  And taking the worst RF and putting him in center and saying "You would not miss much" strikes me as a stretch.  Imagine Arcia in center.   Then he goes out of his way to say that he "shakes his head" most of the time regarding defensive metrics.

 

I think his decision and comments suggest he places way more emphasis on watching people than metrics.  I think this is a fair statement.

 

http://blogs.twincities.com/twins/2014/12/04/twinsights-terry-ryan-defensive-metrics-twins-ignore/

[Shane Wahl]

The Twins will need to have a different GM in order to ever get serious about intelligent statistical analysis in baseball. That is simply a fact. The Twins are lucky that the other aspect of assessment--the eye test--has been done by someone very good at the top in Terry Ryan. That said, the overall antipathy towards statistical analysis by this organization is embarrassing. Ryan seems like a more eloquent and careful-with-words version of Charles Barkley. 

[jimmer]

Not only did Terry say Torii defense was fine and he could play CF for a few days and not miss much, he also said he didn't think our OF defense was that bad last year.