How Much Data Does a Win-Probability Model Need?
It’s a lot; quite a heck of a lot
Posts
Inside the Simulation Model
We dig into the details of the model:
Is It Any Good? Evaluating the Model
How do we know whether a win probability is any good?
Simulating NFL Win Probability
I developed a simulation-based win probability model for football.
Watermarking runif
I breakdown how to (invisibly) watermark a generative process
Win Probability from Fully Synthetic Data
Real data can be so messy.. why not just use synthetic data!
Pokemon Individual Recourse
Or how to be the most effective Pokemon trainer
Assessing Distributions with SIMEX
Noisy observations bias a whole distribution’s tails; SIMEX is an approach for correcting this
Permutation Feature Importance Doesn't Make Sense
It gives arbitrary results when faced with dependent data
Neural Bradley Terry Models
I've got a fever! And the only prescription... is more neural networks.