Daniel Brooks
Data Scientist
Actions
Daniel Brooks is a games enthusiast interested in AI applications in PyTorch. He is the original developer of the popular draftsim.com draft simulator, as well the as the co-author of the paper "AI drafting solutions for Magic: the Gathering". He currently works as a lead data scientist at Disney and previously worked at Meta. He holds a PhD in Applied Physics from Caltech and a BS in Engineering Physics from Cornell University.
Area of Expertise
Learning to Draft: AI Models for Magic: The Gathering
Drafting in Magic: the Gathering is a popular game mode in which players select cards from rotating packs. Historically, developing up-to-date drafting models is challenging as new sets are frequently released. To address this, we developed a pipeline for training drafting models (MLPs) using data from winning players via the 17lands data aggregator. These models are used to produce a list of candidate picks given a collection. Quantitative and qualitative results are provided for a total of 40 recent sets. Notable quirks in human pick behavior—such as the tendency to overvalue high-rarity cards—are analyzed and mitigated through an additional term in the loss function. The project code and a ready-to-use Colab draft assistant are publicly available at: https://github.com/danieljbrooks/statistical-drafting
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top