Skip to content

JackLich10/nfl-draft-data

Repository files navigation

Notes

Please read the data dictionaries for descriptions of each dataset and descriptions of the variables present.

If you attempt to read in the .csv files in R with the commonly used readr::read_csv function, please use the parameter guess_max = 13000 to avoid parsing errors (this has to do with how readr guesses column types using only the first 1000 rows). Alternatively, use the read.csv function which will handle this without intervention.

If you are joining the datasets together, the value player_id is uniquely identified across all data. Be wary of pos_abbr, as the abbreviations sometimes (though rarely) differ across datasets. I have had no issues with school, school_name, and school_abbr, but player_id will always join the data correctly.

If you are joining in ESPN college QBR statistics from college_qbr.csv, join by guid and player_name.

Data Dictionaries

nfl_draft_prospects.csv

Information on previous NFL draft prospects dating back to 1967 (first year of the common draft). If the player has NA values for pick, overall, and round, it means he went undrafted or the draft has not occurred yet (for current year prospects).

variable class description
year integer year of draft
player_id character unique player ID
guid character guid
player_name character name of player
position character position
pos_abbr character position abbreviation
school character school
school_name character school name
school_abbr character school abbreviation
weight double weight (lbs)
height double height (inches)
link character player link
pick integer pick in round
overall integer overall pick in draft
round integer round in draft
traded logical pick traded?
trade_note character trade note
team character NFL team
team_abbr character NFL team abbreviation
team_logo_espn character NFL team logo
pos_rk double ESPN position rank
ovr_rk double ESPN overall rank
grade double ESPN player grade
player_image character player image

nfl_draft_profiles.csv

Information on NFL draft prospects including pre-draft text analysis in columns text* for * in 1, 2, 3, 4.

variable class description
player_id character unique player ID
guid character guid
player_name character player name
position character position
pos_abbr character position abbreviation
weight double weight (lbs)
height double height (inches)
player_image character player image
link character player link
school_logo character school logo
school character school
school_abbr character school abbreviation
school_name character school name
pos_rk double ESPN position rank
ovr_rk double ESPN overall rank
grade double ESPN player grade
text1 character prospect analysis 1
text2 character prospect analysis 2
text3 character prospect analysis 3
text4 character prospect analysis 4

college_qbr.csv

ESPN college QB QBR metrics for every quarterback in college football since 2004. Please join in this dataset using guid and player_name.

variable class description
season integer college season
guid character guid
player_name character player name
age double current age
total_qbr double Adjusted Total QB Rating, which values the QB on all play types on a 0-100 scale adjusted for the strength of opposing defenses faced.
points_added double Number of points contributed by a QB, accounting for QBR and how much he plays, above the level of an average quarterback.
qb_plays double Plays on which the QB has a non-zero expected points contribution. Includes most plays that are not handoffs.
total_epa double Total expected points added with low leverage plays, according to ESPN Win Probability model, down-weighted.
pass double Expected points added on pass attempts with low leverage plays down-weighted.
run double Clutch-weighted expected points added through rushes
exp_sack double Clutch-weighted expected points added (lost) from sacks (not fumbles that may occur because of sacks)
penalty double Expected points added on penalties with low leverage plays down-weighted.
raw_qbr double Raw Total QB Rating, which values QB on all play types on a 0-100 scale (not adjusted for opposing defenses faced)
sack double Expected points added on sacks with low leverage plays down-weighted.

ids.csv

Information for joining a player's ESPN ID (commonly referred to as player_id in many datasets) with their ID in the nflfastR package. At this time, it only has data for QB's. Might add in receivers and running backs later.

variable class description
espn_id character unique player ID from ESPN
player_name character player name
nflfastR_id character unique player ID in nflfastR

college_stats.csv

NFL draft prospect counting statistics in college. Includes counting statistics like interceptions, tackles, receiving touchdowns, passing touchdowns, etc. by a player's college season.

variable class description
player_id character unique player ID
alt_player_id character alternate player ID
player_name character player name
pos_abbr character position abbreviation
school character school
school_abbr character school abbreviation
school_primary_color character school primary color
school_alt_color character school alternate color
season integer college season
statistic character statistic
value double statistic value
active logical active player?
all_star logical all star in college?

About

NFL draft prospect data for text analysis

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published