• Used regression analysis to improve Oversee's domain buying profit by over 100% and optimized renewal strategies to save about $30k per month.
  • Redesigned the statistics for A/B testing and the "auto-tester" used for optimization experiments at Oversee.net, one year leading to a lift in revenue per visitor of over 40%.
  • Mathematics degree from Pomona College

My background

Personal Background

  • Led an effort to objectively rank  MMA fighters for the first time at Sherdog.com.
​​Education...

UC Berkeley's MIDS Program (Master of Information and Data Science)

​




































(Video-only and bridge courses at UC Berkeley)

  MIDS 1a - Fundamentals of Linear Algebra

  MIDS 1b - Fundamentals of Data Structures and Algorithms

  INFO W18 - Python Bridge

​

  • Master of Information and Data Science degree from UC Berkeley

​


Computer Science


     MITx: 15.071x The Analytics Edge​

     MITx: 6.00.1x Introduction to Computer Science and Programming Using Python

     HarvardX: PH525.1x Statistics and R for the Life Sciences
     HarvardX: PH525.2x Matrix Algebra and Linear Models
     HarvardX: PH525.3x Advanced Statistics for the Life Sciences
     DavidsonX: D003x.2 Applications of Linear Algebra (Part 2)

     Johns Hopkins: The Data Scientist’s Toolbox

     Johns Hopkins: R Programming 


​


  • Scored a perfect 10 on the probability / statistics actuarial exam.




Only One Move

 

My most memorable instance of data-driven decision-making was the development of a simple and profitable strategy for online poker.  It also provided content for my guest lectures for Harvey Mudd College’s popular Mathematics of Games course on how the combination of data and mathematics can result in successful strategies that defy conventional wisdom.

It all started when I saw a “poker corner” segment on TV stating that a player who is short-stacked (has few chips remaining) has only one move: all-in.  This was presented as a bad situation, but in my mind it was a great opportunity to make the game tractable.  Some poker sites allowed you to start with a short-stack, so if my hypothesis was correct, I could actually make a profit.  Being somewhat risk-averse, I only ever deposited $50 into my online poker account. 

After utilizing an initial all-in or fold strategy that allowed me to gather hand history files on my opponents, I engineered an exploitive strategy by calculating the expected call equity (value when my bet is called), fold equity (value when everyone folds), and the cost of patience (the blinds).  Conventional wisdom states that repetitive strategies can’t work, and that your specific opponents and position at the table are the most important things to consider.  However, my data was telling me that all of this was incorrect and that a handy profit could be made. 

While the strategy was simple, the analysis was not.  In addition to creating a predictive model to evaluate potential strategies, I also had to estimate my precise edge in the game, in order to use the Kelly Criterion to minimize exposure to bad luck while maximizing hourly winnings.
 
In the end, my $50 became $30,000, and after sharing the strategy with friends, we collected some crazy stories to tell disbelieving family members.

MOOCs I recommend


Math and Statistics


  Stanford: StatLearning Statistical Learning
  Duke:  Data Analysis and Statistical Inference 

  The Caltech-JPL Summer School on Big Data Analytics

  DavidsonX: D003x.1 Applications of Linear Algebra (Part 1)
  Khan Academy: Linear Algebra (140 videos)
​​