How does an MLB R&D department differ from what's publicly available on FanGraphs or Baseball Reference?

Public analytics sites provide league-wide metrics with a data lag and no access to proprietary data streams. An MLB R&D department has real-time access to full Statcast/Hawk-Eye data (including camera-based ball tracking beyond what Baseball Savant publishes), proprietary Trackman data from affiliated minor-league parks, and internal player development metrics not available publicly. The R&D department also builds custom models calibrated to the club's specific roster construction philosophy, not generic league-wide frameworks.

What is the relationship between R&D and the coaching staff?

The most effective R&D directors treat the coaching staff as clients, not recipients of unsolicited analysis. Coaches identify the questions they need answered — 'Should this pitcher throw his curveball more in two-strike counts against left-handed hitters?' — and the R&D team provides data-backed answers in a format coaches can use in pre-game preparation and in-game decision-making. The relationship requires trust built over time, which means the R&D director must speak baseball language, not data science jargon, and deliver analysis at the pace the game demands.

How does Statcast data feed into R&D models?

MLB's Statcast system, powered by Hawk-Eye cameras and Doppler radar, generates position coordinates, velocity vectors, and spin data on every ball in play and every pitch. The R&D department accesses this through the Baseball Savant API and the Commissioner's Office's full-data feeds. Raw Statcast data is transformed through the club's data pipeline into the inputs for projection models: expected batting average (xBA), expected slugging (xSLG), sprint speed, catch probability, route efficiency, and dozens of other derived metrics. The R&D director designs the model architecture on top of these inputs.

How has the pitch-clock changed R&D priorities?

The 2023 pitch clock (15 seconds with bases empty, 18 seconds with runners) has generated new research questions that R&D departments are actively working on: Does the pitch clock favor power pitchers (who rely less on hold time)? Are there count-specific timing patterns that catchers and pitchers can exploit legally? How does the pitch clock interact with pitcher fatigue models across a 162-game season? PitchCom — the electronic pitch-communication system — has also generated data on sequence selection that was previously unobservable.

How is AI changing the Director of R&D role?

Large language models are beginning to appear in baseball operations contexts — AI tools that enable natural language queries against the club's Statcast database, or that summarize multiple scouting reports on a prospect into a synthesized evaluation. Machine learning has been central to baseball R&D for years (neural network pitch-classification models, gradient boosting for player projection), but the LLM era is accelerating the speed at which research outputs can be communicated to non-technical users. Directors increasingly manage AI model governance — ensuring that models are tested, validated, and not blindly trusted — alongside the traditional research function.

Sports

MLB Director of Research and Development

Last updated May 17, 2026

At a glance

Salary (USD)$320K

$200K low$500K high

Read time: 9 min
Last updated: May 17, 2026

Salary methodology

Our proprietary model combines official data from sources such as the U.S. Bureau of Labor Statistics and industry compensation reports, along with publicly available job postings, posting details, and other market signals, to identify what we believe is a representative range for this role.

These figures are directional and provided for informational and educational purposes only. Actual compensation varies by employer, location, experience, certifications, and negotiation, and should not be relied upon for hiring, salary-negotiation, or financial- planning decisions.

Role-specific factorsCompensation reflects both baseball market rates and competition from the broader data science and technology industry for quantitative talent. Large-market clubs compete for top researchers who also have Silicon Valley or quant finance opportunities. Directors managing teams of 5-15 analysts and data scientists earn at the higher end. Some clubs provide performance bonuses tied to research output adoption — specifically, models that are put into production use by the GM or coaching staff.

The MLB Director of Research and Development leads the analytical engine of a club's baseball operations — building predictive models for player evaluation, pitch design, roster optimization, and opponent game planning using Statcast, Hawk-Eye, Trackman, and proprietary data. The role sits at the intersection of statistical research, data science, and baseball operations, requiring both technical depth and the ability to translate complex models into practical decisions that the GM, coaching staff, and player development system can use.

Role at a glance

Typical education: Master's or PhD in statistics, econometrics, machine learning, or related quantitative field; top programs include MIT, Carnegie Mellon, UC Berkeley, University of Chicago
Typical experience: 5-10 years in quantitative analysis, with 3-6 years specifically in baseball or sports analytics
Key certifications: No formal certifications required; PhD or Master's in quantitative field standard at top clubs; public baseball analytics portfolio (SABR conference presentations, open-source projects) valued
Top employer types: All 30 MLB clubs; MLB Advanced Media; sports data vendors (Sportradar, TrackMan, Baseball Cloud); sports betting analytical operations
Growth outlook: Growing; MLB R&D teams have expanded steadily for 15 years; Hawk-Eye expansion to affiliates and AI tool adoption are driving further scope increases
AI impact (through 2030): Transformative — LLMs enabling natural-language querying of baseball databases; ML has been central to baseball R&D for a decade; AI model governance and LLM integration are new director responsibilities.

Duties and responsibilities

Lead an R&D team of analysts and data scientists building predictive models for player evaluation, roster construction, and in-game strategy
Develop and maintain Statcast-based player projection systems covering offensive, pitching, and defensive performance for both MLB and minor-league players
Build pitch-design models that identify optimal pitch shapes, spin-axis adjustments, and repertoire additions for individual pitchers based on their release characteristics
Produce opponent game-planning analytical packages for the coaching staff, integrating Hawk-Eye ball-tracking data with historical tendency analysis
Collaborate with the Director of Baseball Systems to productionize research models into tools the GM, coaches, and scouts can use operationally
Evaluate proposed trades, free-agent signings, and draft picks using projection models, presenting risk-adjusted value assessments to the GM
Design and execute research projects addressing specific baseball operations questions: shift restriction impact, pitch-clock effects on pitcher workloads, defensive positioning optimization
Maintain the club's player WAR modeling system, calibrating it to Statcast data and validating against public baselines (FanGraphs, Baseball Reference)
Present research findings at internal front-office meetings and represent the club at external analytics conferences (MIT Sloan, SABR Analytics)
Recruit, hire, and develop analysts and data scientists with quantitative backgrounds and genuine baseball domain interest

Overview

The R&D director is baseball's equivalent of a quantitative research chief — leading the team that builds the analytical infrastructure behind every data-driven decision in the front office. When the GM is evaluating whether to offer a $90M free-agent contract, when the hitting coach wants to know if a prospect's Trackman profile projects well at the major-league level, or when the pitching coach wants a second pitch-design recommendation for a starter with a stagnating slider — the R&D department is the source of the analytical answer.

The role has evolved dramatically since the Moneyball era. Early baseball analytics teams were primarily running regression analyses on publicly available statistics. Modern R&D departments work with full Statcast data (ball position, velocity, spin, and camera-derived biomechanical measurements at 300 frames per second), Trackman pitch data from all 120 affiliated minor-league parks, proprietary video analytics, and increasingly wearable biometric data from training programs. The scale and complexity of the data environment require professional data engineers and scientists, not just statistically sophisticated front-office generalists.

Pitch design has become a primary R&D function at most clubs. Using Rapsodo and Trackman spin-axis data, R&D teams build models that identify how a pitcher's physical characteristics (arm slot, spin axis, release point) interact with different grip configurations to produce specific ball-flight shapes. The goal is identifying which pitch shapes will generate the most swing-and-miss or weak contact at the major-league level given the pitcher's stuff profile. The Houston Astros' pitch-design program — credited with transforming pitchers like Justin Verlander post-trade and developing Lance McCullers Jr.'s curveball into an elite weapon — established the proof of concept; most clubs have since built comparable programs.

Player projection is the other pillar. The R&D department maintains a system that projects the future performance of both current roster players and external acquisition targets based on their Statcast profiles and aging curves. These projections inform every significant roster decision: the trade package assessment, the free-agent contract length negotiation, the prospect graduation decision. When the projections are calibrated well, they give the GM a quantitative anchor in what would otherwise be purely subjective negotiations.

The translation function — taking rigorous statistical work and making it accessible to coaches, scouts, and the GM without dumbing it down — is perhaps the hardest skill in the role. A pitch-design recommendation that arrives as a regression output is useless to a pitching coach. The same recommendation delivered as 'Your slider's spin axis at release is generating horizontal break of 9 inches; shifting your grip 3mm toward the inside seam projects to increase that to 12 inches and add 200 RPM of active spin' is actionable.

Qualifications

Directors of Research and Development in baseball hold some of the strongest quantitative credentials in professional sports. The role has attracted candidates from PhD programs in statistics, economics, operations research, and computer science — and from quant finance, tech, and government analytics careers.

Educational background:

Master's or PhD in statistics, econometrics, applied mathematics, machine learning, or a related quantitative field (most common among directors at top-market clubs)
Bachelor's in computer science or data science with exceptional analytical project experience is possible for the most talented candidates
Some directors hold degrees in economics or physics with a self-taught statistics background supplemented by baseball analytics community work

Technical skills required:

Statistical modeling: experience with regression, survival analysis, mixed-effects models, and Bayesian methods in sports or comparable applied contexts
Machine learning: proficiency with scikit-learn, XGBoost, or PyTorch/TensorFlow for classification, regression, and sequence modeling tasks
Python and SQL as primary languages; R experience valued for legacy codebases
Baseball data fluency: deep familiarity with Statcast data structure, Baseball Savant API, and the derived metrics (sprint speed, OAA, run value) built on Hawk-Eye tracking

Baseball domain knowledge:

Understanding of how pitching mechanics interact with ball-flight physics (seam-shifted wake, Magnus effect, spin efficiency)
Familiarity with the full player development pathway from amateur draft through major leagues
Ability to evaluate research projects for both analytical rigor and practical baseball relevance

Leadership:

Building and managing teams of 5–15 analysts and data scientists with diverse quantitative backgrounds
Prioritizing research projects against the front office's decision calendar
Communicating complex analytical results to non-technical audiences (coaches, scouts, the GM)

Career outlook

MLB R&D departments have expanded from zero-to-two person operations at most clubs in 2010 to teams of 5–20 analysts and data scientists at well-resourced organizations in 2026. The Director of Research and Development has become a standard front-office position, though the title varies — some clubs use Director of Analytics, VP of Baseball Research, or Director of Baseball Analytics.

Salary range: $200K–$280K at small-market clubs with lean R&D teams; $280K–$380K at mid-market organizations; $380K–$500K at large-market clubs competing with tech industry for senior analytical talent. Some large-market clubs have created C-suite level positions (Chief Analytics Officer, Chief Data Officer) above the Director of R&D level, creating internal promotion paths.

The role's influence has grown significantly since the 2020s. The Astros' success demonstrated that R&D-driven roster construction and player development could produce sustained competitive windows. The Dodgers' investment in analytical infrastructure — reported to include one of the largest R&D teams in baseball — has become a model other clubs aspire to. Small-market clubs including the Rays and A's have used R&D advantages to compete above their payroll level, demonstrating the ROI of investment in this department.

The career ceiling beyond the Director level includes VP of Baseball Research, Chief Analytics Officer, and increasingly, General Manager — several GMs who rose through analytical careers now lead major-market clubs. Andrew Friedman (Dodgers) and Chaim Bloom (former Red Sox) represent the analytical-background-to-President-of-Baseball-Operations trajectory at scale.

Lateral transitions include tech company sports analytics roles, sports betting analytical leadership, and academic research positions. The quantitative skills developed in baseball R&D translate across industries, making former baseball directors competitive in data science leadership markets broadly.

Sample cover letter

Dear [General Manager / VP of Baseball Operations],

I am applying for the Director of Research and Development position with [Club]. As a Senior Research Analyst at [Organization] for the past five years, I have built projection models, pitch-design tools, and game-planning analytical packages that have been adopted into the front office's daily decision-making workflow.

My most impactful recent work was a pitch-design recommendation system built on Trackman spin-axis data from our affiliates that identified 11 pitchers whose curveball spin profiles projected to benefit from grip modifications. Coaches worked with three of those pitchers during spring training; all three posted career-best curveball whiff rates in the subsequent season. I also maintain our internal player projection model — calibrated against full Statcast data — that the GM references in trade evaluation meetings.

I write primarily in Python, work with our baseball systems team on productionizing models into coach-accessible applications, and present research findings in front of the full coaching staff three times per season. I believe the research director must earn the coaching staff's trust by delivering answers they can actually use, not just technically correct outputs.

I would welcome the opportunity to present a recent research project — specifically our shift-restriction impact analysis from 2023 and its ongoing roster construction implications — as a demonstration of how I approach baseball operations questions.

Thank you for your time.

Sincerely, [Your Name]

Frequently asked questions

How does an MLB R&D department differ from what's publicly available on FanGraphs or Baseball Reference?: Public analytics sites provide league-wide metrics with a data lag and no access to proprietary data streams. An MLB R&D department has real-time access to full Statcast/Hawk-Eye data (including camera-based ball tracking beyond what Baseball Savant publishes), proprietary Trackman data from affiliated minor-league parks, and internal player development metrics not available publicly. The R&D department also builds custom models calibrated to the club's specific roster construction philosophy, not generic league-wide frameworks.
What is the relationship between R&D and the coaching staff?: The most effective R&D directors treat the coaching staff as clients, not recipients of unsolicited analysis. Coaches identify the questions they need answered — 'Should this pitcher throw his curveball more in two-strike counts against left-handed hitters?' — and the R&D team provides data-backed answers in a format coaches can use in pre-game preparation and in-game decision-making. The relationship requires trust built over time, which means the R&D director must speak baseball language, not data science jargon, and deliver analysis at the pace the game demands.
How does Statcast data feed into R&D models?: MLB's Statcast system, powered by Hawk-Eye cameras and Doppler radar, generates position coordinates, velocity vectors, and spin data on every ball in play and every pitch. The R&D department accesses this through the Baseball Savant API and the Commissioner's Office's full-data feeds. Raw Statcast data is transformed through the club's data pipeline into the inputs for projection models: expected batting average (xBA), expected slugging (xSLG), sprint speed, catch probability, route efficiency, and dozens of other derived metrics. The R&D director designs the model architecture on top of these inputs.
How has the pitch-clock changed R&D priorities?: The 2023 pitch clock (15 seconds with bases empty, 18 seconds with runners) has generated new research questions that R&D departments are actively working on: Does the pitch clock favor power pitchers (who rely less on hold time)? Are there count-specific timing patterns that catchers and pitchers can exploit legally? How does the pitch clock interact with pitcher fatigue models across a 162-game season? PitchCom — the electronic pitch-communication system — has also generated data on sequence selection that was previously unobservable.
How is AI changing the Director of R&D role?: Large language models are beginning to appear in baseball operations contexts — AI tools that enable natural language queries against the club's Statcast database, or that summarize multiple scouting reports on a prospect into a synthesized evaluation. Machine learning has been central to baseball R&D for years (neural network pitch-classification models, gradient boosting for player projection), but the LLM era is accelerating the speed at which research outputs can be communicated to non-technical users. Directors increasingly manage AI model governance — ensuring that models are tested, validated, and not blindly trusted — alongside the traditional research function.

See all Sports jobs →