Method
Player projections
- Method
m4-v0- Version
m4-v0- As of
- 2026-10-08
- Family
- Models (player)
- Grain
- player x projected season (four seasons ahead of the last season on record)
Cohort rule
No cohort rule applies to these stats.
Stats this method computes
| Key | Name | Definition | Unit | Transform |
|---|---|---|---|---|
proj_pts_p75 | Projected points per 75 possessions (Proj PTS/75) | Points per 75 possessions, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 75 possessions | per_75 |
proj_reb_p75 | Projected rebounds per 75 possessions (Proj REB/75) | Rebounds per 75 possessions, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | rebounds per 75 possessions | per_75 |
proj_ast_p75 | Projected assists per 75 possessions (Proj AST/75) | Assists per 75 possessions, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | assists per 75 possessions | per_75 |
proj_stl_p75 | Projected steals per 75 possessions (Proj STL/75) | Steals per 75 possessions, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | steals per 75 possessions | per_75 |
proj_blk_p75 | Projected blocks per 75 possessions (Proj BLK/75) | Blocks per 75 possessions, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | blocks per 75 possessions | per_75 |
proj_tov_p75 | Projected turnovers per 75 possessions (Proj TOV/75) | Turnovers per 75 possessions, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | turnovers per 75 possessions | per_75 |
proj_fg2_pct | Projected two-point percentage (Proj 2P%) | Two-point makes over attempts, projected: the 5/4/3 weighted base shrunk by attempts toward the age cohort and the correlated shooting factor, then aged. Projected for the season named by the row, from data through the as-of season; carries as_of. | share (0-1) | |
proj_fg3_pct | Projected three-point percentage (Proj 3P%) | Three-point makes over attempts, projected: the 5/4/3 weighted base shrunk by attempts toward the age cohort plus the correlated shooting factor (a strong free-throw and two-point shooter is projected to shoot threes better than few attempts say), then aged. Projected for the season named by the row, from data through the as-of season; carries as_of. | share (0-1) | |
proj_ft_pct | Projected free-throw percentage (Proj FT%) | Free-throw makes over attempts, projected: the 5/4/3 weighted base shrunk by attempts toward the age cohort and the correlated shooting factor, then aged. Projected for the season named by the row, from data through the as-of season; carries as_of. | share (0-1) | |
proj_mpg | Projected minutes per game (Proj MPG) | Minutes per game, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | minutes per game | per_game |
proj_bpm | Projected box plus/minus (Proj BPM) | Box plus/minus, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 100 possessions vs average | |
proj_obpm | Projected offensive box plus/minus (Proj OBPM) | Offensive box plus/minus, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 100 possessions vs average | |
proj_dbpm | Projected defensive box plus/minus (Proj DBPM) | Defensive box plus/minus, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 100 possessions vs average | |
proj_ws_per_48 | Projected win shares per 48 minutes (Proj WS/48) | Win shares per 48 minutes, projected: the last three seasons weighted 5/4/3 by exposure, shrunk toward the player's age cohort (career-year cohort without a birth date) by reliability, then moved along the aging curve. Projected for the season named by the row, from data through the as-of season; carries as_of. | wins per 48 minutes | |
proj_gp | Projected games played (Proj GP) | Projected games on an 82-game schedule: the weighted share of the season's games played, shrunk toward the cohort; not aged. Projected for the season named by the row, from data through the as-of season; carries as_of. | games (82-game season) | |
proj_min | Projected minutes (Proj MIN) | Projected minutes: projected minutes per game times projected games. Projected for the season named by the row, from data through the as-of season; carries as_of. | minutes (82-game season) | total |
proj_pts_pg | Projected points per game (Proj PTS) | Projected points per game: projected points per 75 possessions at the as-of season's league pace, times projected minutes per game over 48. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per game | per_game |
proj_reb_pg | Projected rebounds per game (Proj REB) | Projected rebounds per game: projected rebounds per 75 possessions at the as-of season's league pace, times projected minutes per game over 48. Projected for the season named by the row, from data through the as-of season; carries as_of. | rebounds per game | per_game |
proj_ast_pg | Projected assists per game (Proj AST) | Projected assists per game: projected assists per 75 possessions at the as-of season's league pace, times projected minutes per game over 48. Projected for the season named by the row, from data through the as-of season; carries as_of. | assists per game | per_game |
proj_vorp | Projected value over replacement player (Proj VORP) | Projected VORP: (projected BPM + 2.0) times projected minutes over 3,936, that is 48 x 82 (vorp-v0's scale: a fifth of an 82-game team's minutes). Projected for the season named by the row, from data through the as-of season; carries as_of. | replacement-level points per 100 team possessions, season | |
proj_ws | Projected win shares (Proj WS) | Projected win shares: projected win shares per 48 minutes times projected minutes over 48. Projected for the season named by the row, from data through the as-of season; carries as_of. | wins | |
proj_pts_p75_lo | Projected points per 75 possessions, 80% interval low end (Proj PTS/75 lo) | The low end of the 80 percent interval around the projection: the central estimate minus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 75 possessions | per_75 |
proj_pts_p75_hi | Projected points per 75 possessions, 80% interval high end (Proj PTS/75 hi) | The high end of the 80 percent interval around the projection: the central estimate plus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 75 possessions | per_75 |
proj_bpm_lo | Projected box plus/minus, 80% interval low end (Proj BPM lo) | The low end of the 80 percent interval around the projection: the central estimate minus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 100 possessions vs average | |
proj_bpm_hi | Projected box plus/minus, 80% interval high end (Proj BPM hi) | The high end of the 80 percent interval around the projection: the central estimate plus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | points per 100 possessions vs average | |
proj_vorp_lo | Projected value over replacement player, 80% interval low end (Proj VORP lo) | The low end of the 80 percent interval around the projection: the central estimate minus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | replacement-level points per 100 team possessions, season | |
proj_vorp_hi | Projected value over replacement player, 80% interval high end (Proj VORP hi) | The high end of the 80 percent interval around the projection: the central estimate plus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | replacement-level points per 100 team possessions, season | |
proj_ws_lo | Projected win shares, 80% interval low end (Proj WS lo) | The low end of the 80 percent interval around the projection: the central estimate minus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | wins | |
proj_ws_hi | Projected win shares, 80% interval high end (Proj WS hi) | The high end of the 80 percent interval around the projection: the central estimate plus 1.2816 times the one-season-ahead residual sd on the training seasons, times the square root of the seasons ahead. The backtest reports how often the actual fell inside. Projected for the season named by the row, from data through the as-of season; carries as_of. | wins | |
shooting_factor | Shooting factor (Shoot F) | The correlated shooting factor: the reliability-weighted mean of the player's 2P%, 3P% and FT% standing (z within the as-of season), the latent shooting skill the three percentages share. It moves each projected percentage toward what the other two say. | standard deviations from the season's league mean | |
reliability | Projection reliability (Rel) | How much of the projection is the player's own record: weighted minutes over weighted minutes plus 1,000. Near 0 the projection is mostly the cohort; near 1 mostly the player. | share (0-1) | |
base_min | Projection base minutes (Base MIN) | Reliability input: minutes over the seasons the projection is built from (the as-of season and up to two before). | minutes | |
base_gp | Projection base games (Base GP) | Reliability input: games over the seasons the projection is built from (the as-of season and up to two before). | games | |
base_seasons | Projection base seasons (Base Seasons) | Reliability input: how many of the three base seasons (the as-of season and the two before) the player played. | seasons |
This page carries the definitions only. The full method note, its coverage and its calibration record publish with the data pages. Every value a data page shows will name this method and its version. All methods