I have a lot of ideas for articles but don’t have time to work on them all. Here are all the articles I want to write but haven’t had the time yet
- Analysis of Winning Fantasy Football Teams
- Acquiring data as we speak, preliminary results show
- I want to expand on this and get data from 2007-2016 ready by 8/20
- Also, I want to determine if autopicker is better than most people just like how etf’s outperform most money managers
- Acquiring data as we speak, preliminary results show
- Replicating Youtube’s recommender engine and making it better through page rank
- Interactive graph visualization to help you explore new domains like music in spanish
- Pure analysis using the recommender engine I develop
- May lead to a new browser control system (chromium fork) that allows you to explore the web in the way I think is proper
- Data Scraping Guide
- Sequence of web posts that I compile into a guide and eventual e-book
- FiveThirtyEight data dump
- Show how to scrape interactive visualizations
- Backtest Nate Silver against the spread and publish an imperative definition of his algorithms like I did for the NBA.
- Conditional Probability
- Basic explanation of the concept in how it relates to decision making
- Itemset analysis of Daily Fantasy Lineups
- Interactive visualization of player ownership itemsets for daily fantasy
- Personal Notification Service
- I’ve created a chatbot that tells me when the train is and sends my my picks for different games.
- I want to extend this into a library/howto that shows you how to send emails and texts programmatically
- Evaluate AWS Lex
- Analysis of mean reversion and how it affects ownership of fantasy players
- Article on gambler’s ruin and different betting/position sizing strategies
- Publish an NBA Strength of Schedule + stats model for picking teams
- Potentially publish against the spread version as well
- The leading case against the Standard Model of Physics and the Copenhagen interpretation– Bohmian mechanics
- Would inspire another article detailing the problem of the status quo incentives in physics research.
Let me know if you want me to work on something
may I suggest regarding draft position:
– start with 10 teams, in snake format and 20 players with value descending in equal steps
– then model the 20 players as a sharp drop over the first 5 or so and then a steady drop or even plateau on the remaining
– now make a model for another position with different steepnesses for the first players and run your simulation
– you might discover that the rapidness of the drop of good RBs or WRs, or star QBs in various combinations means you need to alter your strategy for your position to beat the odds or a large data set
– there is actually a winning strategy for each position in a snake draft based on positional ranking and the quickness to least value
– I stumbled on this after a 4/10 year winner in my league retired and he told me about it. In our league the most winners were from positions 7&8, then 4, then 2. In head to head it’s important to also stagger the division membership on draft picks
hope there is enough nonsense in here to get you back into coding! (BTW I do this stuff but for startups at my VC Fund – https://preference.vc)
Very interesting thanks!