Ben Lindbergh's Effectively Wild podcast on Fangraphs is out with an episode today that features a lengthy interview with Daniel Adler, the Twins' Director of Baseball Operations.

 

Check it out here.

Lots of good stuff, interesting guy and liked the football vs. baseball comparisom.

 

From a specific Twins perspective, what struck me the most was his descruption of the previous front office as "lean and strong" which seems pretty accurate. I also was intrigued by his description of the Twins analytics department before Falvey took over. It was a little light on personnel but also was very advanced in some areas, partially because of advantages gained by integrating technology somewhat late.

I saw this... I'd never heard of this guy, and his role makes me wonder even more about what St. Peter does.

 

I saw this... I'd never heard of this guy, and his role makes me wonder even more about what St. Peter does.

 

St. Peter does the business side of the organization (marketing, tickets sales, vendors, Target Field management). His and Adler's jobs aren't remotely close.

 

St. Peter does the business side of the organization (marketing, tickets sales, vendors, Target Field management). His and Adler's jobs aren't remotely close.

 

The Director of Baseball Operations has nothing to do with the person who was the long-time President of Baseball Operations and is now the President and CEO? Interesting organization structure.

 

The Director of Baseball Operations has nothing to do with the person who was the long-time President of Baseball Operations and is now the President and CEO? Interesting organization structure.

 

Titles vary, but every organization splits business and baseball operations.

Edited January 9, 20188 yr by drjim

This is a good interview, even though it wasn't as Twins-centric as I had hoped.

 

One thing that stuck in my mind...
Adler mentioned that the Twins actually had solid technical infrastructure created prior to Falvey's arrival, with a system that was running in the cloud and provided a systematic method for storing and accessing player data, models, etc. This comment corroborates other info I've heard from people within the front office. While Jack Goin's tenure with the Twins left some things to be desired on the research and analytic side, I think it is pretty clear that he took the job of creating/designing the technical infrastructure to support the research and analytics seriously, and everything I've heard/read indicates that he did a good job in that area.

 

 

This is a good interview, even though it wasn't as Twins-centric as I had hoped.

 

One thing that stuck in my mind...
Adler mentioned that the Twins actually had solid technical infrastructure created prior to Falvey's arrival, with a system that was running in the cloud and provided a systematic method for storing and accessing player data, models, etc. This comment corroborates other info I've heard from people within the front office. While Jack Goin's tenure with the Twins left some things to be desired on the research and analytic side, I think it is pretty clear that he took the job of creating/designing the technical infrastructure to support the research and analytics seriously, and everything I've heard/read indicates that he did a good job in that area.

 

I wouldn't want to run analytics in a cloud. You'd be stuck with the views that the web programmers are able to give you. If you want to see something in a new way, you're stuck waiting for someone to program it, and they might have higher priority stuff to do or some other yob might deny the request because he or she doesn't understand your new theory. Most of us would rather have the raw data and the tools to put it together ourselves, or at least have our assistant do it, in a tool that is more nimble.

A player database in the cloud, sure. I'm sure everybody has that. But download the raw data so you can test premises quickly.

Edited January 9, 20188 yr by Doomtints

 

I wouldn't want to run analytics in a cloud. You'd be stuck with the views that the web programmers are able to give you. If you want to see something in a new way, you're stuck waiting for someone to program it, and they might have higher priority stuff to do or some other yob might deny the request because he or she doesn't understand your new theory. Most of us would rather have the raw data and the tools to put it together ourselves, or at least have our assistant do it, in a tool that is more nimble.

A player database in the cloud, sure. I'm sure everybody has that. But download the raw data so you can test premises quickly.

I think you're misunderstanding what cloud computing is.

It's simply a remote server (or more accurately, a series of remote servers). A user can have as much or as little access as an administrator grants them.

 

A properly set up cloud computing setup used for data crunching is negligibly different from a local network setup, the only difference is where the server is hosted.

 

But in a cloud setup, it's a lot easier to set it up to work anywhere, not only a local office. They're made to be accessed from anywhere and there's less hassle with opening ports, securing the system, and granting outside user access.

 

I think you're misunderstanding what cloud computing is.

It's simply a remote server (or more accurately, a series of remote servers). A user can have as much or as little access as an administrator grants them.

 

A properly set up cloud computing setup used for data crunching is negligibly different from a local network setup, the only difference is where the server is hosted.

 

But in a cloud setup, it's a lot easier to set it up to work anywhere, not only a local office. They're made to be accessed from anywhere and there's less hassle with opening ports, securing the system, and granting outside user access.

 

We are responding to second- or even third-hand information so we don't need to debate it, but I think the only reason someone would bring up a "cloud" in this instance is if they are referring to an interface built to spit out the information they want to see, a-la Salesforce. Because, as you said, if it's just a storage device it doesn't matter.

Edited January 10, 20188 yr by Doomtints

 

We are responding to second- or even third-hand information so we don't need to debate it, but I think the only reason someone would bring up a "cloud" in this instance is if they are referring to an interface built to spit out the information they want to see, a-la Salesforce. Because, as you said, if it's just a storage device it doesn't matter.

I don't think that's the case at all. To the layman, the "cloud" is a myriad of things and often used colloquially as "data stored outside our internal network" because they heard a tech guy use the word once and it stuck.

 

There are loads of reasons to use cloud servers in a large organization. It's easily accessible, it's fast *everywhere* (no need for a single gateway or bottleneck, you pull from the closest server), a lot of the security is managed by an outside vendor, and you can get massive numbers-crunching horsepower without the overhead of server hardware, connection speed, and maintenance.

 

A cloud is simply a network of servers that mirror data. You can set up those servers to have as much brute force power or to do anything a normal server would do in your local office. To the end user, a properly set up cloud server(s) will look and feel identical to a local network server, except it works better outside the network.

 

I think you're misunderstanding what cloud computing is.

It's simply a remote server (or more accurately, a series of remote servers). A user can have as much or as little access as an administrator grants them.

 

A properly set up cloud computing setup used for data crunching is negligibly different from a local network setup, the only difference is where the server is hosted.

 

But in a cloud setup, it's a lot easier to set it up to work anywhere, not only a local office. They're made to be accessed from anywhere and there's less hassle with opening ports, securing the system, and granting outside user access.

 

That and massive boost in computing power.

 

We are responding to second- or even third-hand information so we don't need to debate it, but I think the only reason someone would bring up a "cloud" in this instance is if they are referring to an interface built to spit out the information they want to see, a-la Salesforce. Because, as you said, if it's just a storage device it doesn't matter.

 

I think you're perspective on the cloud is a bit off. Brock hit the nail on the head.

 

Not to mention that the Twins are likely a small organization in terms of IT footprint. While they can certainly afford the outlay to have all their devices in house, I suspect the ROI for cloud usage is pretty simple as it tends to ROI very well for a small organizations.

anyone want to see my win.ini?

I think you're perspective on the cloud is a bit off. Brock hit the nail on the head.

 

Not to mention that the Twins are likely a small organization in terms of IT footprint. While they can certainly afford the outlay to have all their devices in house, I suspect the ROI for cloud usage is pretty simple as it tends to ROI very well for a small organizations.

Yeah, you can buy a lot of server if you don’t have to pay a sysadmin $75k a year.

anyone want to see my win.ini?

There goes your political career.

 

anyone want to see my win.ini?

 

win.ini win.ini chick.in. din.ini

 

I think you're perspective on the cloud is a bit off. Brock hit the nail on the head.

 

Not to mention that the Twins are likely a small organization in terms of IT footprint. While they can certainly afford the outlay to have all their devices in house, I suspect the ROI for cloud usage is pretty simple as it tends to ROI very well for a small organizations.

 

The Twins have 623 employees which does not include players. They have an adequate IT staff.

 

I think both of you are still missing what I'm saying. It's not an accomplishment in analytics to be using the cloud as a data storage device, which is what you are defending. Whether the data is on your hard drive or on the cloud, it's no difference. It's what you do with the data. The fact that the cloud was brought up as a great accomplishment in the world of analytics means to me that what the poster was stating is they have a tool built to spit out what they want to see. This is limiting because they're not seeing anything else without talking to a programmer.

 

The Twins have 623 employees which does not include players. They have an adequate IT staff.

 

I think both of you are still missing what I'm saying. It's not an accomplishment in analytics to be using the cloud as a data storage device, which is what you are defending. Whether the data is on your hard drive or on the cloud, it's no difference. It's what you do with the data. The fact that the cloud was brought up as a great accomplishment in the world of analytics means to me that what the poster was stating is they have a tool built to spit out what they want to see. This is limiting because they're not seeing anything else without talking to a programmer.

 

I think you are much too hung up on this. My takeaway from the interview is that since the Twins were a little late to developing analytical systems, it allowed them to build state of the art systems, which presumedly would include more agility and computing power, and not be bogged down with outdated systems that would either need to be rebuilt and/or transferred. Depending on how deep into analytics early adopters were (and I assume they were quite deep), this could be somewhat problematic. Much better to jump in with clean infrastructure, no?

 

The analogy that sprung to my mind was telephone infrastructure in the US vs most of the rest of the world.

 

Whether the data is on your hard drive or on the cloud, it's no difference. It's what you do with the data. The fact that the cloud was brought up as a great accomplishment in the world of analytics means to me that what the poster was stating is they have a tool built to spit out what they want to see. This is limiting because they're not seeing anything else without talking to a programmer.

Bolded: well, sure, of course. What you do with information is always more important than how you store said information.

 

The rest: Is it a "great accomplishment" or "a modern system of data storage and distribution"? You're not only reading between the lines, you're adding a hell of a lot of speculation to what was a pretty simple statement. There's absolutely nothing to indicate that "cloud storage" equals "a limited web interface that requires programmer input to alter".

 

In fact, as an actual web developer, I know there are ways to create a web interface that basically changes itself as more data is entered into the system.

 

If there's a web interface at all, which I doubt there is as web interfaces don't make much sense for this kind of work (at least the way I imagine the system working based on previous experience). There are loads of tools that can be installed locally to pull data directly from a remote source.

 

If there's a web interface at all, which I doubt there is as web interfaces don't make much sense for this kind of work (at least the way I imagine the system working based on previous experience). There are loads of tools that can be installed locally to pull data directly from a remote source.

 

The much heralded Astros system is a web interface which was hacked, if you recall. Seeing as media and baseball were incredibly impressed with what the Astros built, would it *truly* surprise you if other teams mimicked it as being forward-thinking?

 

You're putting too much faith into the metrics abilities for a team which has never proven competence in the area.

 

The much heralded Astros system is a web interface which was hacked, if you recall. Seeing as media and baseball were incredibly impressed with what the Astros built, would it *truly* surprise you if other teams mimicked it as being forward-thinking?

 

You're putting too much faith into the metrics abilities for a team which has never proven competence in the area.

1. Do you have a link stating the Cardinals gained access to a web interface? I remember reading they accessed a database (databases?) and email but never recall seeing how that information was accessed, whether via web or a direct connection to the database.

 

2. The Twins have overhauled their analytics department, even going so far as to let go of the guy previously in charge. Both Levine and Falvey come from more technically-adept organizations, as do most of their recent hires. What the Twins did or did not do in the past is becoming less relevant by the day.

 

What the Twins did or did not do in the past is becoming less relevant by the day.

 

How so? They still haven't done anything, 12+ months after joining the team.

How so? They still haven't done anything, 12+ months after joining the team.

So you start the conversation by saying the Twins have always been bad at analytics and follow it up by saying they haven’t “done anything”, which extends far beyond analytics.

 

We know they’ve done a lot in analytics based only on their hires and shuffling of the organizational hierarchy. Naturally, we don’t know the details of what they’ve done or what they’re doing because teams don’t reveal much about the inner workings of their analytics departments.

 

Also, why’d you ignore my request for a link about the Astros’ web interface? I’ve scanned a dozen articles now and not a single mention of how the data was accessed with specifics about the interface.

 

How so? They still haven't done anything, 12+ months after joining the team.

 

 

I would respectfully submit that you have negligible knowledge about what's been done behind the scenes. 

 

I would respectfully submit that you have negligible knowledge about what's been done behind the scenes. 

 

Nor do I care about what's been done behind the scenes. I'm talking about key signings since they arrived. 

 

So you start the conversation by saying the Twins have always been bad at analytics and follow it up by saying they haven’t “done anything”, which extends far beyond analytics.

 

Analytics are pointless if they are not using them, right? Why would we buy into the Twins becoming metrics geniuses when there are no results from whatever-it-is-they-are-doing?

Analytics are pointless if they are not using them, right? Why would we buy into the Twins becoming metrics geniuses when there are no results from whatever-it-is-they-are-doing?

So you’re judging a 5+ year plan on two months of little action in an offseason where there has been shockingly little action from MLB at large.

 

And to do so, you need to ignore the Twins’ 2017 draft, which employed a vastly different strategy than any draft under Ryan or Smith.

 

Hell, earlier in this very thread you said you’d never heard of Adler until now. Maybe it’s possible you’re not exactly an authority on the Twins’ current use of analytics.

 

So you’re judging a 5+ year plan on two months of little action in an offseason where there has been shockingly little action from MLB at large.

 

Untrue, I'm also basing it on last offseason.

 

For whatever reason people are swarming at the moment and taking narrow/bizarre interpretations of my statements and attacking them, so I will bow out of this conversation. I'll be back when people are ready to converse instead of whatever-this-is.