Juan Lomeli

Data scientist · Dallas, TX

Juan Lomeli

Read the build log GitHub LinkedIn
2025 ADP board · rounds 1–3 12-team PPR · weeks 1–17
  1. 1.0149
  2. 1.0256
  3. 1.0346
  4. 1.0442
  5. 1.0544
  6. 1.0655
  7. 1.0740
  8. 1.0861
  9. 1.0957
  10. 1.1048
  11. 1.1152
  12. 1.1253
  13. 2.1253
  14. 2.1145
  15. 2.1047
  16. 2.0952
  17. 2.0846
  18. 2.0759
  19. 2.0649
  20. 2.0551
  21. 2.0449
  22. 2.0344
  23. 2.0257
  24. 2.0150
  25. 3.0147
  26. 3.0253
  27. 3.0344
  28. 3.0450
  29. 3.0552
  30. 3.0646
  31. 3.0745
  32. 3.0845
  33. 3.0945
  34. 3.1059
  35. 3.1160
  36. 3.1258
≤40%≥60% Hover a pick: win % of the teams that drafted him
Every square is a real pick from the 2,000 simulated leagues behind the ADP heat map, scored on actual 2025 NFL points. Green picks helped their teams win; red ones did not.

Featured project

Built for one league, honest about the numbers

2024 – present

ESPN Fantasy Analytics

A full-stack analytics app for a 12-team ESPN fantasy football league. It reads the league through ESPN's fantasy API, adds eleven seasons of public NFL box scores, and answers the arguments a league actually has, with the uncertainty attached.

  • Luck & all-play. Every team against every score, every week.
  • Season audit. The exact distribution of records a season's scores allowed.
  • Draft tools. Thirteen ranking lists reconciled player by player, a strategy lab, and an ADP heat map.
  • Trade analyzer. Graded on both lineups and playoff odds, not points traded.
  • Weekly projections. Scored against ESPN's every week, losses included.
  • Stats lab. ANOVA, permutation tests and shrinkage on the league's own scores.
What I learned building it

Writing

Notes and case studies

All writing
  1. Oct 9, 2026

    What I learned building an analytics app for my ESPN fantasy leagueNew

    Luck, bench points, draft grades and trade values for a 12-team league, and the lesson I relearned every week: the dangerous bugs don't crash, they return a confident wrong number.

    PythonReactStatisticsFantasy football
  2. Oct 8, 2020

    Scatter Plot Matrix to Visualize Regional Data

    Ten thousand retail sales, four regions and every product category in one chart, by encoding the extra dimensions into a scatter plot matrix.

    VisualizationRetail
  3. Sep 13, 2020

    Parallel Coordinates for Resident Activity Data

    Applications, move-ins, move-outs and renewals across 100+ apartment properties, read in one parallel coordinates plot, with brushing to pull out the outliers.

    VisualizationReal estate
  4. Sep 7, 2020

    Property Management Analytics Using the Pareto Principle

    The 80/20 rule applied to 176 properties: where delinquency concentrates, and what average resident stay says about turnover.

    Pareto analysisReal estate
  5. Jul 5, 2020

    Apache Hive vs. Spark: A Speed Comparison with a SQL Query

    How Hadoop's pieces fit together, and a timed SQL query on an EC2 cluster showing why Spark SQL outruns Hive.

    Big dataSQL

About

Hi, I'm Juan.

Analysis

  • Python
  • pandas
  • NumPy
  • SQL
  • Statistics
  • Simulation
  • Machine learning

Building

  • Flask
  • React
  • TypeScript
  • REST APIs
  • Automated testing

Data platforms

  • Microsoft Fabric

Education

M.S. Advanced Data Analytics and B.S. Mathematics,
University of North Texas