Earlier this week I revealed my final 2018 NCAA D1 men's college basketball TEAM rankings. Today I'm presenting you the final 2018 NCAA D1 men's college basketball PLAYER rankings.
The full 2018 Player Rankings can be found here.
Below is an excerpt of a few of the ratings for the Top 10 players ranked by Total Q Value for the 2017-18 season.
Let's start with what the different values mean.
Glossary
PP100: This stands for Points Produced per 100 estimated team possessions. This is a rate stat calculated by combining the SCR (Scoring Value per 100 possessions), OR (Offensive Rebounding Value per 100 possessions), A (Assist Value per 100 possessions), and T (Turnover Value per 100 possessions) stats into rPP (raw Points Produced per 100 possessions). rPP is then adjusted for dSOS (defensive opponent strength of schedule) and then adjusted again so that the total Points Produced number by all D1 players is equal to the total Points Scored by all D1 players. This number is an estimate of how many points an individual player would have produced against a D1 average opponent for every 100 team possessions played. It is important to remember that there is a lot of noise in this stat among players who haven't played very much, and that it is much harder to have a high PP100 when playing a huge number of minutes.
OV: This stands for Offensive Value. This is an estimate of the total offensive value a player produced over the season. It is calculated by dividing PP100 by 100 to get Points Produced per team possession, and then multiplying by the number of estimated team possessions for which the player was on the court.
PA100: This stands for Points Allowed per 100 estimated team possessions. This is a rate stat calculated by combining the TPA (Team Points Allowed per 100 possessions), B (Block Value per 100 possessions), S (Steal Value per 100 possessions), DR (Defensive Rebounding Value per 100 possessions), and F (Foul Value per 100 posessions) stats into rPA (raw Points Allowed per 100 possessions). rPA is then adjusted for oSOS (offensive opponent strength of schedule), and then adjusted again so that the total Points Allowed number by all D1 players is equal to the total points scored by the opponents of D1 teams. This number is an estimate of how many points an individual player would have allowed against a D1 average opponent for every 100 team possessions played. Like with PP100, this number is very noisy for players who didn't play very many minutes, and it is much harder for players who play a ton of minutes to maintain a low PA100.
DV: This stand for Defensive Value. This is an estimate of the total opponent offensive value a player was responsible for allowing over the season. It is calculated by dividing PA100 by 100 to get Points Allowed per team possession, and then multiplying by the number of estimated team possessions for which the player was on the court.
QVal: This is the Q Value, or overall total individual value of a player's on court performance. It is calculated simply by subtracting DV from OV. There is some noise, but adding up all of the QVal numbers for all of the teams produces a team ranking very close to those calculated only by looking at team stats.
The top 10 teams by total QVal for the 2018 season are:
1. Villanova
2. Duke
3. Kansas
4. Purdue
5. Michigan St.
6. North Carolina
7. West Virginia
8. Michigan
9. Cincinnati
10. Virginia
QV100: This is the Q Value per 100 team possessions. It is calculated simply by subtracting PA100 from PP100.
Component Stats:
SCR: Scoring Value per 100 possessions. This is calculated by finding the total points scored by a player per 100 team possessions for which he is on the court. And then subtracting that number by the opportunity cost value for each missed 2 point field goal, missed 3 point field goal, and missed Free Throw.
OR: Offensive Rebounding Value per 100 possessions. This is calculated by taking the number of offensive rebounds above average a player grabbed per 100 team possessions, and then multiplying that by an estimated fractional scoring value of the extra offensive possession those offensive rebounds created.
A: Assist Value per 100 possessions. This is calculated by finding the difference between the average number of points a D1 average team scores on assisted offensive scoring possession and the average number of points a D1 average team scores on all other offensive possessions. That difference is then multiplied by the number of assists per 100 team possessions a player is credited with.
T: Turnover Value per 100 possessions. This is calculated by finding the average value of an offensive possession that does not end with a turnover and adding that to the estimated average value of an offensive possession that begins with a turnover. This total is then multiplied by the number of turnovers per 100 team possessions a player commits. This component is subtracted from the total value of all of the other offensive components when calculating rPP.
TPA: Team Points Allowed per 100 possessions. This component is simply 20% of a team's adjusted defensive efficiency (Opponent Points Scored per 100 possessions adjusted for strength of schedule). All players on the same team will have the same TPA. The individual differences come into play in the other defensive stats measured.
B: Block Value per 100 possessions. This is calculated by finding the difference between the average number of points a D1 average team scores on an offensive possession and the average number of points a D1 average team scores on an offensive possession that does not end in a block. This is than added to the estimated average extra number of points an average D1 team scores on an offensive possession that begins with a block as opposed to a possession that begins any other way. This total is then multiplied by the number of blocks per 100 team possessions a player is credited with.
S: Steal Value per 100 possessions. This is calculated like block value except it looks at the difference between defensive possessions that end and do not end in steals, and offensive possessions that do and do not begin with steals.
DR: Defensive Rebounding Value per 100 possessions. This is calculated by taking the number defensive rebounds above average a player grabbed per 100 team possessions, and then multiplying that by an estimated fractional scoring value of an extra offensive possession that otherwise would have been created by an offensive rebound.
F: Foul Value per 100 possessions. This is calculated by finding the difference between the average number of points a D1 average opponent scores on possession that ends in a foul and one that does not. It is important to note that this stat definitely interjects some noise, because there are definitely times when committing a foul is a better defensive strategic decision than not committing a foul.
B, S, and DR are positive components that subtract from the TPA, while F is a negative component that adds to it. (Things are reversed on the defensive side, since you want to have fewer points allowed, not more).
The full 2018 Player Rankings can be found here.
Below is an excerpt of a few of the ratings for the Top 10 players ranked by Total Q Value for the 2017-18 season.
Let's start with what the different values mean.
Glossary
PP100: This stands for Points Produced per 100 estimated team possessions. This is a rate stat calculated by combining the SCR (Scoring Value per 100 possessions), OR (Offensive Rebounding Value per 100 possessions), A (Assist Value per 100 possessions), and T (Turnover Value per 100 possessions) stats into rPP (raw Points Produced per 100 possessions). rPP is then adjusted for dSOS (defensive opponent strength of schedule) and then adjusted again so that the total Points Produced number by all D1 players is equal to the total Points Scored by all D1 players. This number is an estimate of how many points an individual player would have produced against a D1 average opponent for every 100 team possessions played. It is important to remember that there is a lot of noise in this stat among players who haven't played very much, and that it is much harder to have a high PP100 when playing a huge number of minutes.
OV: This stands for Offensive Value. This is an estimate of the total offensive value a player produced over the season. It is calculated by dividing PP100 by 100 to get Points Produced per team possession, and then multiplying by the number of estimated team possessions for which the player was on the court.
PA100: This stands for Points Allowed per 100 estimated team possessions. This is a rate stat calculated by combining the TPA (Team Points Allowed per 100 possessions), B (Block Value per 100 possessions), S (Steal Value per 100 possessions), DR (Defensive Rebounding Value per 100 possessions), and F (Foul Value per 100 posessions) stats into rPA (raw Points Allowed per 100 possessions). rPA is then adjusted for oSOS (offensive opponent strength of schedule), and then adjusted again so that the total Points Allowed number by all D1 players is equal to the total points scored by the opponents of D1 teams. This number is an estimate of how many points an individual player would have allowed against a D1 average opponent for every 100 team possessions played. Like with PP100, this number is very noisy for players who didn't play very many minutes, and it is much harder for players who play a ton of minutes to maintain a low PA100.
DV: This stand for Defensive Value. This is an estimate of the total opponent offensive value a player was responsible for allowing over the season. It is calculated by dividing PA100 by 100 to get Points Allowed per team possession, and then multiplying by the number of estimated team possessions for which the player was on the court.
QVal: This is the Q Value, or overall total individual value of a player's on court performance. It is calculated simply by subtracting DV from OV. There is some noise, but adding up all of the QVal numbers for all of the teams produces a team ranking very close to those calculated only by looking at team stats.
The top 10 teams by total QVal for the 2018 season are:
1. Villanova
2. Duke
3. Kansas
4. Purdue
5. Michigan St.
6. North Carolina
7. West Virginia
8. Michigan
9. Cincinnati
10. Virginia
QV100: This is the Q Value per 100 team possessions. It is calculated simply by subtracting PA100 from PP100.
Component Stats:
SCR: Scoring Value per 100 possessions. This is calculated by finding the total points scored by a player per 100 team possessions for which he is on the court. And then subtracting that number by the opportunity cost value for each missed 2 point field goal, missed 3 point field goal, and missed Free Throw.
OR: Offensive Rebounding Value per 100 possessions. This is calculated by taking the number of offensive rebounds above average a player grabbed per 100 team possessions, and then multiplying that by an estimated fractional scoring value of the extra offensive possession those offensive rebounds created.
A: Assist Value per 100 possessions. This is calculated by finding the difference between the average number of points a D1 average team scores on assisted offensive scoring possession and the average number of points a D1 average team scores on all other offensive possessions. That difference is then multiplied by the number of assists per 100 team possessions a player is credited with.
T: Turnover Value per 100 possessions. This is calculated by finding the average value of an offensive possession that does not end with a turnover and adding that to the estimated average value of an offensive possession that begins with a turnover. This total is then multiplied by the number of turnovers per 100 team possessions a player commits. This component is subtracted from the total value of all of the other offensive components when calculating rPP.
TPA: Team Points Allowed per 100 possessions. This component is simply 20% of a team's adjusted defensive efficiency (Opponent Points Scored per 100 possessions adjusted for strength of schedule). All players on the same team will have the same TPA. The individual differences come into play in the other defensive stats measured.
B: Block Value per 100 possessions. This is calculated by finding the difference between the average number of points a D1 average team scores on an offensive possession and the average number of points a D1 average team scores on an offensive possession that does not end in a block. This is than added to the estimated average extra number of points an average D1 team scores on an offensive possession that begins with a block as opposed to a possession that begins any other way. This total is then multiplied by the number of blocks per 100 team possessions a player is credited with.
S: Steal Value per 100 possessions. This is calculated like block value except it looks at the difference between defensive possessions that end and do not end in steals, and offensive possessions that do and do not begin with steals.
DR: Defensive Rebounding Value per 100 possessions. This is calculated by taking the number defensive rebounds above average a player grabbed per 100 team possessions, and then multiplying that by an estimated fractional scoring value of an extra offensive possession that otherwise would have been created by an offensive rebound.
F: Foul Value per 100 possessions. This is calculated by finding the difference between the average number of points a D1 average opponent scores on possession that ends in a foul and one that does not. It is important to note that this stat definitely interjects some noise, because there are definitely times when committing a foul is a better defensive strategic decision than not committing a foul.
B, S, and DR are positive components that subtract from the TPA, while F is a negative component that adds to it. (Things are reversed on the defensive side, since you want to have fewer points allowed, not more).
Comments
Post a Comment
Practice Respect!