Heat Check App

The concept for the heat check app is simple: we wanted to produce an app that would help users identify which players are seeing the biggest spike in fantasy production in recent games. Some players that are seeing boosts in minutes, usage, and/or fantasy production, perhaps due to injuries or trades are well known. However, some players manage to slip under the radar of the public and DraftKings and FanDuel pricing schemes. Furthermore, that task of developing an ordered hierarchy of players who have been the hottest in the last week can be overwhelming. Enter the Heat Check App.

App User Guide

  1. Site input: Select site scoring scheme to compare players over. DraftKings or FanDuel only, can't select multiple.
  2. "Heat Window" input: Select how many days back you want to consider as players' recent performance. The app aggregates per game averages of minutes, usage, true shot attempts, and fantasy points for "recent" games and compares that to per game average for games that aren't considered "recent". With this input, the user can define the window for which games are considered recent. The most appropriate window will vary from player to player and situation to situation. Some statistic increases happen suddenly and with much public attention, such that site pricing responds appropriately and quickly. To target these players before their pricing reaches efficiency, a shorter Heat Window might be warranted. However, there is an inherent tradeoff with this approach, as statistic reporting will be more sensitive to small sample sizes. However this is an unavoidable tradeoff in DFS trend recognition, balancing early recognition with small sample overreaction.
  3. Statistic columns input: In this input, you can select the statistics which margins you want to compare. The input is defaulted to compare marginal increase/decrease in minutes per game, usage per game, and fantasy points per game (either DK or FD as specified in (1)) as well as sample size variables like the number of games and overtimes each player has played in in recent and earlier games. You can also compare per game differences in true shot attempts ("TSA/Game"), which is an aggregation of 2-point and 3-point field goals and free throws, with appropriate weighting for the expected value each shot attempt provides.
  4. Teams input: This input is defaulted to show the teams that are playing each night. You can add specific teams by adding them to the input, or if you want to view all teams, you can add "All" to the input.
  5. Raw data download button: Clicking this button downloads a .csv file called "ASA Heat Check Data.csv"; this file can be opened in Microsoft Excel. The file contains our full NBA dataset, with an extra columns called "Heat Window", and values of either TRUE or FALSE, a logical variable of whether or not the specified row is in the "heat window" of recent games. This file could be useful when further examining the factors that might have contributed to certain players' statistical increases or decreases in recent games.
  6. Sample size columns: The "Games" and "OTs" columns provide context about the sample size of the specified heat window. In each column, the number to the left of the slash represents the number of games/overtimes the player has played in prior to the specified heat window and to the right of the slash the number of games/overtimes the player has played in within the recent heat window.
  7. Marginal per-game statistic columns: These columns are the information that should help users determine which players are hot and which players are cold. The numbers in each column represent the per-game difference in the specified statistic for recent games in the heat window and earlier games prior to the heat window. A positive value represents an increased per-game average in the heat window than in games prior. These columns can be sorted in increasing or decreasing order.