Please enable javascript in your browser to view this site!

 

Our application user guides are intended to help you a) use our applications more efficiently and b) understand their underlying methodology. Use the menu to the left to navigate to each application's user guide in this page.

PLAYER DISTRIBUTIONS APP

The motivation behind this app is to create a dynamic interface for comparing the historical fantasy production of NFL players. We’ve aggregated data across DraftKings, FanDuel, and Yahoo! scoring formats, and have allowed for date filtering so you can compare players across long-term  averages or restrict distributions to recent performances.

Mockups-18.jpg

1.   Player(s) Input: Select the player(s) whose distribution(s) you would like to compare. Distributions are of fantasy points accrued per game. Offensive players and kickers can be searched by name, D/STs can be searched by the teams’ abbreviations plus “DST” (i.e. Chicago Bears D/ST = “CHI DST”).

2.   Date Range Input: Select the date range over which you would like to compare distributions.

3.   Site Input: Select the site site scoring scheme for which you would like to compare player(s).

4.   Show Quantiles Input: Specify whether or not you would like for the output to show player distribution quantiles (see 7).

5.   Raw Data Download Button: Download the game-level fantasy data for the players that are specified in (1) during the date range specified in (2).

6.   Distribution Quantiles: This output shows the 5th, 10th, 25th, 50th (median), 75th, 90th, and 95th percentile historical outcomes of players specified in (1) within the date range specified in (2). The solid line represents the 50th percentile/median outcome, the dashed lines represent the 25th and 75th percentile outcome, the dashed-and-dotted lines represent the 10th and 90th percentile outcome, and the dotted lines represent the 5th and 95th percentile outcomes.

7.   Player Legend: Legend for discerning between multiple players’ distributions.

Team Point Total Projection App

Mockups-29.jpg

  1. Team Input: Select the teams whose run projection distributions and probabilities you want to compare. Selecting “All” will compare the distributions and probabilities of all teams, you can select a set of teams by slate, i.e. “Sun. Early” and/or “Sun. Afternoon”.

  2. Custom Point Total Input: Optional input for estimating the probability that each team exceeds (or scores equal to) the point total specified in this input. Output is shown in the “Max. Point Probabilities” tab (4).

  3. Show Quantiles Input: Yes/no input for specifying whether or not you would like the distribution output to show teams’ 5th, 10th, 25th, 50th (median), 75th, 90th, and 95th percentile point total projections.

  4. Tab Select: Toggle between distributional (“Point Total Projections”) and probabilistic (“Max. Point Total Probabilities”) point total projections.

  5. Team Legend: Legend for identifying teams’ distributions in the output plot.

  6. Highest Total Probability Output: Probability of each team scoring the most points of the team set specified in (1). Not necessarily the probability of each team being the highest scoring team of the week, unless you have “All” selected.

  7. Custom Run Probability Output: Probability of each team scoring at least as many points as specified in (2).

Play Count Projection App

Mockups-32.jpg
Mockups-31.jpg



  1. Team Input: Select the teams whose play count projections you want to compare. Selecting “All” will compare projections of all teams, you can select a set of teams by slate, i.e. “Sun. Early” and/or “Sun. Afternoon”.

  2. Staff Position Input: In the “Coach Historic Play Counts” tab (3), you can view the historic total, run, and pass play counts run by head coaches (as either a head coach or offensive coordinator), offensive coordinators (as either head coaches or offensive coordinators), or defensive coordinators (total, run, and pass plays run against as DC) of each team. In this input, select whether you want to view historic data of head coaches (HC), offensive coordinators (OC), or defensive coordinators (DC).

  3. Tab Select: Select whether you would like to view team play count projections (“Play Projections”) or historic coach play count data. The play count projections are produced by a predictive machine learning model that considers the over/under and spread of a game, historic coach play count data, and teams’ play count history in the present year.

Volume X Efficiency App

Mockups-34.jpg
Mockups-35.jpg


  1. Player(s) Input: Select the player(s) whose rushing/receiving volume and efficiency you would like to compare. You can select team positional units, i.e.  “KC - WR” and/or “KC - FLEX”. You can also select a set of players from an entire slate.

  2. Date Range Input: Select the date range over which you would like to compare volume and efficiency.

  3. Site Input: Select the site site scoring scheme for which you would like to compare player(s).

  4. Volume/Efficiency Columns: Select which volume and/or efficiency columns you would like to view in the application tabe output. “MS” stands for market share, it is the  proportion of the team total that a player accounts for (i.e. a “Target MS” of 0.25 means a player receives a quarter of his team’s total targets). “RZ” stands for redzone, “i-10” stands for inside-10-yardline, they represent players’ volume/efficiency near the goal line. “FP/Tch” stands for fantasy points as specified in (3) per touch attempt (rush attempts + targets) - this is an efficiency metric, while MS columns are volume metrics. The columns selected in this input will appear in the above table, and can be sorted by clicking the arrows next to each column name.

  5. Current Team/Include Preseason Conditions: If you want to view players’ volume/efficiency only in games when they are on their current team, check this box. If you want to include players’ preseason volume/efficiency metrics, check this box. These conditions carry over to the “Time Series” plot tab (see 8).

  6. Sample Size Conditions: If you want to only view players who have played in at least x games (particularly for MS metrics) and/or received at least y touch attempts (particularly for FP/Tch metrics), specify minimum thresholds here. These conditions carry over to the “Time Series” plot tab.

  7. Player Search: You can search for specific players with this input; this could be particularly useful if you’re looking at many players.

  8. Tab Select: Select whether you would like to view player metrics as means in tabular form or through a week-by-week time series plot.

  9. Plot Y-Variable Input: The “Plot Series” tab can only chart a single metric (compared to multiple metrics in the tabular output). Select which metric you would like to plot over time for the players specified in (1).

  10. Raw Data Download: Download the game-level data of the players and columns specified in (1) and (4), in the time frame specified in (2).

  11. Data Point Info: These labels show extra information about the data point in question. You can pull up the data point info by hovering your mouse over a certain data point. The info labels show player name, the numeric y-value, the metric “base” or denominator (so for “MS Air Yards”, the base of each data point is the total team air yards for the date in question, for “RZ Rush Att” the base would be the team total redzone rush attempts), and the players’ teams’ opponent and game score of the game in question.

  12. Plot Activity Menu: This menu allows you to interact and edit the plot in the application output. Hover over each icon to see an icon description.

QB Dropback App

Mockups-37.jpg

  1. Player(s) Input: Select the quarterback(s) whose dropback frequency and efficiency you would like to compare. You can also select a set of players from an entire slate.

  2. Date Range Input: Select the date range over which you would like to compare dropback frequency and efficiency.

  3. Site Input: Select the site site scoring scheme for which you would like to compare player(s).

  4. Dropback Frequency/Efficiency Columns: Select which frequency and/or efficiency columns you would like to view in the application tabe output. Any column with “%” is the number of plays where a QB drops back, scrambles, or is sacked divided by the number of plays the team runs (so QBs who play only parts of games will have deflated frequency numbers). “RZ” stands for redzone, “i-10” stands for inside-10-yardline, they represent players’ frequency/efficiency near the goal line. “FP/Dropback” stands for fantasy points as specified in (3) per dropback - this is an efficiency metric, while % columns are frequency metrics. The columns selected in this input will appear in the above table, and can be sorted by clicking the arrows next to each column name.

  5. Current Team/Include Preseason Conditions: If you want to view players’ frequency/efficiency only in games when they are on their current team, check this box. If you want to include players’ preseason frequency/efficiency metrics, check this box. These conditions carry over to the “Time Series” plot tab (see 8).

  6. Sample Size Conditions: If you want to only view players who have played in at least x games (particularly for MS metrics) and/or executed at least y dropbacks (particularly for FP/Dropback metrics), specify minimum thresholds here. These conditions carry over to the “Time Series” plot tab.

  7. Player Search: You can search for specific players with this input; this could be particularly useful if you’re looking at many players.

  8. Tab Select: Select whether you would like to view player metrics as means in tabular form or through a week-by-week time series plot.

  9. Plot Y-Variable Input: The “Plot Series” tab can only chart a single metric (compared to multiple metrics in the tabular output). Select which metric you would like to plot over time for the players specified in (1).

  10. Raw Data Download: Download the game-level data of the players and columns specified in (1) and (4), in the time frame specified in (2).

  11. Data Point Info: These labels show extra information about the data point in question. You can pull up the data point info by hovering your mouse over a certain data point. The info labels show player name, the numeric y-value, the metric “base” or denominator (so for “Dropback %”, the base of each data point is the team total plays run for the date in question, for “FP/non-RZ Dropback” the base would be the the total non-redzone dropbacks executed by each player), and the players’ teams’ opponent and game score of the game in question.

  12. Plot Activity Menu: This menu allows you to interact and edit the plot in the application output. Hover over each icon to see an icon description.

RAW DATA DOWNLOAD

Mockups-21.jpg

1.   Seasons Input: Select which seasons of data you would like to download. You can select a single season, or a range of seasons using the numeric slider.

2.   Box Score Type Input: Select which type of box scores you’d like to download. Different box score types have different columns based on the type. The “All” box score type has all offensive players, kickers, and D/STs as rows with offensive, kicking, and D/ST columns as “NA” for mismatched row and column types.

3.   Data Download Button: Press this button to download the data that you have specified in (1) & (2) as a csv file.