We build the decision tools your front office wishes it had
Fat Cat Data is a quantitative AI consultancy for college and professional sports organizations. You describe the tool you need. We scope it, price it as a single project, and deliver working software — a proof of concept first, an MVP when you're ready, and code you own at the end.
High-fidelity flat vector product mockup in enterprise SaaS marketing style, wide hero composition: three overlapping web-application windows floating with subtle drop shadows on a soft pale gray (#f3f6fa) field — centered and largest a draft-board window with a top navigation bar, ranked player rows as white cards with avatar circles and stat bars, and a right slider panel; behind-left a chat briefing window with message bubbles and small citation chips; behind-right a trip-planner window with venue dots joined by one royal blue (#004cce) route line. White (#ffffff) panels with thin (#e3e9f1) hairline borders; charcoal (#16212e) primary marks; slate (#5d6b7b) secondary marks; blue used only for one active tab, one button, one highlighted row, and the route. All text as greeked placeholder bars — no legible words, letters, or numbers. Crisp straight-on geometry, generous spacing, modern and corporate, no people, no photographs Applied science, delivered as software
We work at the intersection of technology, analytics, and artificial intelligence, with a specialty in Quantitative AI: heavy statistical modeling, mathematics, physics, and applied science brought to bear on sports decision-making — across the full player lifecycle, from draft analysis through player development.
Quantitative modeling
Deterministic, auditable models built from statistics, mathematics, and physics — ranking engines, valuation curves, optimization routines. In our tools the model, not the AI, produces every number.
Fat Cat Agentic AI
AI assistants that read your evidence, explain the model's output, draft the brief, and propose the next move — grounded in your data, with every claim traceable to its source. The AI narrates; it never invents the numbers.
PoC and MVP builds
Login-gated web tools your staff actually use: interactive boards, chat workspaces, automated nightly pipelines, exports, and usage dashboards — deployed privately for your organization.
From a conversation to a tool your staff uses
Engagements are scoped and priced as a single project. You always know what you're paying and what you're getting before we write a line of code.
High-fidelity flat vector product mockup in enterprise SaaS marketing style: a two-stage composition on a soft pale gray (#f3f6fa) field — lower-left a simple chat-bubble card containing three greeked request lines, a smooth thin arrow arcing up-right into a complete application window with a ranked board, a small line chart, and a royal blue (#004cce) primary button. White (#ffffff) panels with thin (#e3e9f1) hairline borders and very subtle drop shadows; charcoal (#16212e) primary marks; slate (#5d6b7b) secondary marks; blue used only for the button and one highlighted row. All text as greeked placeholder bars — no legible words, letters, or numbers. Crisp straight-on geometry, generous spacing, modern and corporate, no people, no photographs Describe
You tell us what you want to build — in plain language, from your side of the operation.
Scope
We turn it into a defined project: deliverables, timeline, data requirements, and one price.
Proof of concept
A working system on real data — enough to judge the idea honestly, quickly.
MVP
The concept hardened into a deployed, login-gated tool your staff can rely on.
Handoff
Documentation, training, and transfer. You own what we built.
And after handoff? Every build includes two weeks of hosted testing. Then take the keys — or have us keep running the tool on a flat monthly operations plan: hosting, maintenance, model and data operations, DevOps, and cloud cost management. Optional, cancel anytime, and it stays yours either way.
Working systems we can demo on request
These aren't products, and we don't sell them off the shelf. They're real proof-of-concept and MVP systems we've designed and built — shown here so you can see the kind of tool an engagement produces. Each one can be demonstrated live, and each one suggests what we could build around your organization's own data.
Prospect Board Prediction
An interactive what-if draft board: a deterministic model ranks prospects from internal, external, and ball-tracking evidence, and staff move the weights to see how the board reorders.
Learn more → Contracts & valuationPlayer Contract Modeling
An agentic contract advisor for front offices: plain-English questions, numbers-first answers on draft bonuses and pro deals, grounded in a normalized contract database with comps and evidence.
Learn more → Front officeEmployee Evaluation Analysis
An AI-guided 360 evaluation workspace for baseball-operations staff: rough notes go in, a fair, evidence-based, rubric-grounded evaluation comes out — reviewed by you, exported as PDF or DOCX.
Learn more → Game dayAdvance Scouting Intelligence
A nightly pipeline reads the public baseball web for every upcoming game and hands advance scouts an eight-section, source-cited brief — with a chat layer grounded in the evidence.
Learn more → Draft operationsWeb Draft Intelligence
An evidence-linked read on how the public web values a draft class: consensus, red flags, and disagreements for ~150 prospects, every claim cited — a calibration layer beside internal grades.
Learn more → Game intelligencePitch Tipping Analysis
Game intelligence from pitcher tendency detection — find the tells in your opponents and eliminate them in your own staff before someone else does.
Learn more → Scouting operationsScout Visit Optimization
An AI-first scouting-trip planner: plain-language requests become optimized itineraries — who to see, at which games, in what order — balancing coverage, travel, and reimbursement-grade cost.
Learn more → Scouting operationsScout Audio Insights
An offline-first, voice-first field platform: scouts tap-record notes at the park; the system transcribes, extracts entities and sentiment, and makes every observation searchable through grounded, cited AI chat.
Learn more → Game intelligenceABS Insight Modeling
Modeling and insight around Automated Ball-Strike data — how the machine zone and the challenge system shift pitching, receiving, and hitting value, and what to do about it.
Learn more →Your questions don't come off the shelf. Neither should your tools.
Platform vendors sell the same dashboards to every club in your league, on subscriptions and seat licenses that outlive their usefulness. Generic development shops will build whatever you ask — but they've never sat in a draft room, priced an extension, or routed a scout through three states in four days.
Fat Cat Data occupies the space between: sports-native decision science, built to order. We bring three decades of front-office analytics practice, we build against your data and your workflows, and we hand you the keys when we're done. No license fees. No black boxes. No tool your rivals can buy — and if we run it for you afterward, that's a service you can cancel, not a lock you can't open.
Quantitative AI is a discipline, not a buzzword
A sample of the techniques working inside our builds:
- Bayesian hierarchical modeling
- Survival & aging-curve analysis
- Mixed-effects regression
- Combinatorial route optimization
- Prize-collecting routing
- Signal processing
- Computer vision
- Natural language processing
- Speech recognition & diarization
- Named-entity extraction
- Sentiment analysis
- Embeddings & semantic search
- Agentic tool-use orchestration
- Evidence-cited retrieval