How Sam Mondry-Cohen Went From Intern to Nats Assistant GM

Sam Mondry-Cohen was between his junior and senior years at the University of Pennsylvania when he first began working with the Washington Nationals. He’s come a long way since then. An unpaid intern for six week in the summer of 2009, Mondry-Cohen now holds the title of Assistant General Manager, Baseball Research & Development.

His initial front office experience was the epitome of humble. The Nationals didn’t even have an actual internship program at the time. As Mondry-Cohen explained it, “They were basically there to babysit me. I don’t know that anyone was really looking for any work product.”

What they got was a second sabermetric voice at a time when analytics had yet to become mainstream. Mondry-Cohen may have been majoring in English at Penn — African-American literature was his main focus — but he was an avaricious reader of FanGraphs and Baseball Prospectus. He’d devoured The Book. In short, he was a nerd-in-training.

“I had the vocabulary, and a way of looking at the game, that wasn’t common back then,” recalled Mondry-Cohen. “The Nationals didn’t have an analytics department or an R&D department. They didn’t have any data analysts. Adam Cromie, who went on to become the assistant GM, was the Assistant Director of Baseball Operations. He was the one who appreciated my world view of baseball, and he did assign me a few projects.”

The Nationals told Mondry-Cohen that they wanted him back after he graduated, so he spent more time in Penn’s statistics department than he had previously. To a certain extent, he never left. He continued to work on projects for the Nationals during his senior year.

“They quadrupled my pay,” joked Mondry-Cohen. “I went from zero dollars to zero dollars.”

You Aren't a FanGraphs Member
It looks like you aren't yet a FanGraphs Member (or aren't logged in). We aren't mad, just disappointed.
We get it. You want to read this article. But before we let you get back to it, we'd like to point out a few of the good reasons why you should become a Member.
1. Ad Free viewing! We won't bug you with this ad, or any other.
2. Unlimited articles! Non-Members only get to read 10 free articles a month. Members never get cut off.
3. Dark mode and Classic mode!
4. Custom player page dashboards! Choose the player cards you want, in the order you want them.
5. One-click data exports! Export our projections and leaderboards for your personal projects.
6. Remove the photos on the home page! (Honestly, this doesn't sound so great to us, but some people wanted it, and we like to give our Members what they want.)
7. Even more Steamer projections! We have handedness, percentile, and context neutral projections available for Members only.
8. Get FanGraphs Walk-Off, a customized year end review! Find out exactly how you used FanGraphs this year, and how that compares to other Members. Don't be a victim of FOMO.
9. A weekly mailbag column, exclusively for Members.
10. Help support FanGraphs and our entire staff! Our Members provide us with critical resources to improve the site and deliver new features!
We hope you'll consider a Membership today, for yourself or as a gift! And we realize this has been an awfully long sales pitch, so we've also removed all the other ads in this article. We didn't want to overdo it.

After graduating, he returned to what was now a formal internship program — a paid internship program — and when the 2010 season ended he was hired as a full-time employee. The job title he was offered was Assistant in Baseball Operations. That didn’t fit his self-identity. Mondry-Cohen felt that the work he’d been doing, as well the work he wanted to do, was more akin to that of an analyst.

“There weren’t many people around baseball with that title,” said the now-32-year-old executive. “There were some — Farhan Zaidi having been the first — and I remember our assistant GM kind of pushing back and saying that it was an unorthodox, if not goofy, title. Was I sure I wanted it? I said that I did. He begrudgingly said, ‘Sure. We’ll still pay you $35,000 a year, but you can be an analyst,’ so my first full-time job in baseball was as an Analyst in Baseball Operations.”

He was among Mike Rizzo’s earliest wave of hires. The longtime GM had the interim tag removed from his own title during Mondry-Cohen’s internship, and he didn’t want to lose what he had. Rizzo cut his chops with scouting, but he was astute enough to recognize what the Penn product brought to the table.

“Mike wanted to have some kind of sabermetric voice, as he could kind of see the way the game was going,” Mondry-Cohen explained. “He wanted that voice to be represented some way in the front office, and it’s what Adam Cromie and I were able to provide for him. To this day, he wants a data-driven opinion.”

The Nationals established an actual analytics department following the 2013 season. Mondry-Cohen was promoted and given the title Manager of Baseball Analytics… which isn’t exactly what he wanted. As much as he’d preferred “analyst” when he was first hired, he believed that the department he’d now be leading should be called Research and Development. A year later, he got his wish.

Washington’s lone year with an “Analytics Department,” per se, saw a pair of heady hirings. One came indirectly from Tufts University and Columbia Business School. The other came indirectly from the website you’re currently reading.

“We brought on Mike DeBartolo as an analyst,” explained Mondry-Cohen. “Josh Weinstock, who had been working for me as an intern, came on as an analyst as well. Josh had written publicly at FanGraphs, which is where I’d seen his work.”

In part because he considers his own skill set to be “much more on the baseball side than on the machine-learning, pure-quant side,” Mondry-Cohen typically looks to hire people highly proficient in computer programming.

“For some people that’s maybe not the most fun, because it’s a little disconnected from baseball,” observed Mondry-Cohen. “Presumably they want to work in baseball because they love baseball, but a lot of what they need to be able to do doesn’t necessarily have anything to do with baseball. The size and scope of data sets we see in today’s game require you to have real computer programming and data modeling skills. That’s something that’s really changed since I started out 10 years ago. It’s a lot more than linear regressions and multi-linear regressions now.”

The one-time intern has another valuable piece of advice for those looking to join a baseball R&D team: Don’t wait to get hired.

“If you want to be a surgeon, you can’t just go into an operating room and start cutting people up,” Mondry-Cohen told me. “If you want to be a baseball analyst, you can start analyzing baseball data tonight. There’s a lot of public baseball data, and the sorts of modeling we do… a lot of it is open source, and free. You can use these programs like R from home. And there can be an audience for it. You don’t have to do this work and just put it in a closet. You can post it on your own blog, you can post it at GitHub, you can maybe get it published on the FanGraphs Community Blog. So if you want to do baseball analysis, my advice is to start doing it now. If you don’t have the technical skills to do things like scrape publicly available Statcast data, start there. Get those technical skills, then jump in. You don’t have to wait for a job.”

Mondry-Cohen’s R&D department has grown organically and currently comprises 13 people. In all likelihood, it will continue to grow organically. His career certainly has.





David Laurila grew up in Michigan's Upper Peninsula and now writes about baseball from his home in Cambridge, Mass. He authored the Prospectus Q&A series at Baseball Prospectus from December 2006-May 2011 before being claimed off waivers by FanGraphs. He can be followed on Twitter @DavidLaurilaQA.

23 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Smiling PolitelyMember since 2018
6 years ago

Hooray for English majors!

Bradley WoodrumMember since 2020
6 years ago

There are dozens of us! Dozens!!!

gafford2475
6 years ago

Thank you so much for writing this article! I’m a current senior and have been looking to find out more information about how certain people got started in the baseball ops business and this really helped me a lot!

NATS FanMember since 2018
6 years ago

Well I would not call the Nationals Analytics dept the best and I am a huge NATS fan! but its cool we have 13 people now.

ryanredsox
6 years ago

Would absolutely love to here from more front office people on how they got into baseball. Great article!

Mac
6 years ago

“You don’t have to wait for a job” also means – companies are wanting you to take on unpaid training yourself. You take on the risk of learning these skills with no guarantee of employment down the road. Which, hey, it’s a very competitive world out there. Kind of is what it is but also gets very frustrating at times.

docgooden85Member since 2018
6 years ago
Reply to  Mac

It must be nice to be able to afford to work for free. Only the already-wealthy need apply.

JLanDC
6 years ago

Meh.

dukewinslowMember since 2020
6 years ago

This is a good interview, and good for this guy BUT how is “Ivy league guy works for free for a long ass time because he can” man bites dog news in sports? This is the same story as every other front office person. “Well I come from material circumstances that mean I don’t have to worry about money and lo and behold going to an Ivy league is a better credential than an actual degree with skills that apply here hey here’s some stuff I learned in ggplot2 what’s a table” is what…. 90% of analytically minded FO’s?

Not to be too horrible here, but every junior employee being “Ivy league or Ivy adjacent dude decides to work for free for years because material circumstances mean he DOESN’T have to take a job at Mckinsey, Bain, Morgan Stanley, or whatever” is maybe a bad thing .

As long as people take peanuts to work in sports, sports will pay well below market rate for talent. That will work out for teams in some cases, but in all honesty they’ll never pull in the guys for whom pecuniary incentives matter- particularly folks who had to pay themselves for the kind of training clubs need. This likely limits the tool sets in front offices- you’ll probably never see an I/O structural model, or if you do, it will be 10 years out of date. This is a skills based argument, but the same mechanisms of accepting essentially free labor above all else will keep front offices privileged and homogenous, in addition to likely underqualified (however you choose to define it).

HappyFunBallMember since 2019
6 years ago
Reply to  dukewinslow

Alternatively, as long as WAY more people want to work in sports than there are jobs in sports, clubs will continue to hire the best and the brightest for peanuts.

And while some certainly do lag behind the times, plenty of them do have an abundance of very well qualified people working there. The argument that folks who paid their own way through school are more virtuous and hence more skilled than those that came from money is idealistic and fallacious. Lazy trust fund kids who have connections and happen to like baseball might get their foot in the door, but they won’t last very long.

dukewinslowMember since 2020
6 years ago
Reply to  HappyFunBall

The argument isn’t about virtue, the argument is that someone who gets a PhD (usually funded) or a terminal masters (debt) in a quantitative discipline is not going to be interested in maintaining the oath of poverty they had to maintain in graduate school.
Let’s not pretend that there are many undergrads out there with the skills it takes to advance the field.

If I were to propose an entirely neoclassical argument (I’m not), the unwillingness to offer appropriate incentives guarantees that baseball is oversampling from the underqualified, and is doomed to fall into a “local minima” in terms of performance- sure you look fine, but you’re leaving performance on the table. But I’m willing to concede that there are people who don’t need the money. There are just huge downsides to only sampling from people who don’t need the money, both in terms of the diversity of skills you can acquire, and the diversity in the office full stop. Maybe you get Michael Lewis (the Emory professor, not the moneyball guy) or Etan Green to parachute in occasionally, but you will never be hiring those guys out of school with bleeding edge skillsets without better incentives.

There are exceptions to this, but I don’t know anyone with an advanced degree in a front office who was not independently wealthy before besides Michael Edwards (and he was certainly comfortable enough). 86k in a front office vs. 250k at a software company (w/ options) is the choice (or a hell of a lot more than that in finance. Shit, ACADEMIA pays better than most teams speaking from personal experience). And that’s for someone with hard skills.

HappyFunBallMember since 2019
6 years ago
Reply to  dukewinslow

And my argument is that you are misunderstanding the supply/demand imbalance that exists for the number of jobs available in front offices. Clubs that lag behind the skills curve tend to do so due to a lack of high-level institutional belief in the process, not due to a lack of willing and inexpensive analysts.

dukewinslowMember since 2020
6 years ago
Reply to  HappyFunBall

assuming an even distribution of talent between the people who care about money versus don’t (which is ludicrous, but ok), the population is smaller in those who don’t care about money (if you’re going to argue that those populations are even and incentives are irrelevant… I think you’re in the wrong place).

For n jobs, if baseball is oversampling from the “don’t care about money” group, a given team has a much higher probability of acquiring “low quality talent” for n jobs than they would if they were sampling from the whole population. It’s just simple math. That’s if the talent distribution is EVEN. This is laughable, if we’re talking undergrad trusties and people who graduated from school and made their money over a decade or so versus folks coming straight out of grad school.

Now, maybe that less talented/less skilled group is “good enough,” (the local maxima has equivalent welfare to the global maxima) and the opportunity cost is miniscule. It seems, however like you could extract excess returns from, well, sampling from a more talented more populous distribution and better matching needs to skills. In leagues where performance is truly matched to some major disincentives (i.e. relegation) salaries are rapidly approaching market levels, suggesting that teams in a more competitive market believe that paying more gets you, in essence, better AAV.

Sure, you can outperform other baseball teams by maintaining things the way they are (the local minima), but that only works as long as teams “collude” to stay in the space. The second someone breaks the bank, they could build up a huge, potentially difficult to surmount, advantage (we’re seeing this for a lot of reasons in some other sports- proprietary knowledge that’s building a rather sizable moat using what appears to be quantum computing, judging by the hires)

HappyFunBallMember since 2019
6 years ago
Reply to  dukewinslow

You’re right. MLB clubs are primarily run by fools who have no idea what they’re doing. What the heck are they thinking?

Shalesh
6 years ago
Reply to  dukewinslow

Duke, please provide some examples of what MLB is missing by not paying $250K to candidates who would otherwise be doing I/O Structural Models or quantum computing at Google rather than just paying $86k to someone who finds insights by applying machine learning algorithms to StatCast data in his/her spare time? It seems that the data sets size and complexity in baseball are orders of magnitude smaller than what Google & Facebook need to do and the passion of candidates who avidly follow baseball is important, but maybe you have some ideas.

dukewinslowMember since 2020
6 years ago
Reply to  Shalesh

That’s a really good question. From my formal work in soccer where I have the data, which is insanely dynamic, there’s a lot of work to do in recruitment, especially at finding undervalued good players on bad teams (structural models allow you to simulate the counterfactual of “what if this player is on my team”). Even simple (but tricky at the margins) instrumentation in game state can get you a long way if you simplify things enough. However, i think structural IO models have a lot of promise in producing much much better models of defensive contribution with stat cast data.

The machine learning question is a big one actually- I just spent the last couple of weeks working with a bunch of “off the shelf” classification algorithms to see what the state of the field is for my analytics students. The key there is that off the shelf solutions have, to me, unacceptable error rates in high value decisions. Writing your own classification or prediction algos has value in that space. I wished we got more technical detail on the Astros cheating out of curiosity’s sake, because pitch identification is a place where you’d think a simple model would work, but there are a lot of lower level nodes (I.e. stuff going on) that would perhaps require multilevel network models.

I can’t speak to what quantum and applied physics guys think they can add, but I wouldn’t be surprised to see Edwards and company end up at the Red Sox at some point.

dukewinslowMember since 2020
6 years ago
Reply to  dukewinslow

I forgot about what I think is the most important thing a team could build, and maybe they already have it- and EVSI tool. Expected value of sample information basically tells you whether the “juice” (value added) that additional information provides in a decision framework is worth the “squeeze” of looking for it. EVSI has kind of fallen out of favor in MBA education, for some reason, but with analytics and everything I find that students really love learning how to use the tools. Building a tool for baseball would, I think, be a really big deal.

insidb
6 years ago
Reply to  dukewinslow

In what world is U Penn “Ivy League?”

I’d be hard-pressed to accept that it’s even “Ivy League Adjacent.”

dukewinslowMember since 2020
6 years ago
Reply to  insidb

I mean being part of the penn state system HAS to count for something (old office mate played ball there, I know all the trolls)

Shalesh
6 years ago
Reply to  insidb

Is this a serious comment? School with Wharton is not an Ivy???

Ivy League: Brown, Columbia, Cornell, Dartmouth, Harvard, Princeton, University of Pennsylvania, Yale University.

dukewinslowMember since 2020
6 years ago
Reply to  Shalesh

No it’s not serious. Penn grads get really uptight about being called a state school, which happens a lot ( Cornell actually IS a state school in some ways)

19piercec
6 years ago
Reply to  insidb

Sports conference

insidb
6 years ago

This is a wonderful story of persistence and progress!

Speaking of climbing the MLB ladder, I hear that Houston, Boston, and New York (NL) just executed fast-track advancement plans for Manager, GM, AGM…