SImulations to answer: what is the likelihood of a given % of entrants making it through a week, given the pick distribution and win probabilities?
R
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
2015 survivor week 2 analysis.R
LICENSE
README.md
simulation results.png

README.md

NFL Survivor Pool Win Total Simulations

Simulations to answer: what is the likelihood of a given % of entrants making it through a week, given the pick distribution and win probabilities?

See the full context & writeup at my blog: http://samfirke.com/2015/09/28/calculating-likelihood-of-x-of-entrants-advancing-in-an-nfl-survivor-pool/

The code lives here in the file 2015 survivor week 2 analysis.R. It takes the pick distributions and win probability for a given NFL football week and then runs 1 million simulations of that week, generating a distribution of the % of survivor pool entrants who win.

The particular inputs used in this file are from the 2015 NFL Season Week 2, which (my analysis shows) produced a staggering share of losses, a 1-in-1461 event. But you can swap in new values and re-run for any particular week.

Results

Here are a million simulations of week 2 of the 2015 NFL season. The blue line is the median expected result, the yellow line is the 1st percentile expected result, and the red line is the actual result. ![alt text](simulation results.png)