For the complete documentation index, see llms.txt. Markdown variants of the ranked boards (dynasty and fantasy rankings for NFL, NBA, MLB and CFB) are available by appending.mdto those URLs or by sendingAccept: text/markdown; other pages answer 404.
This page's interactive half has not started on this device. Nothing was read, and nothing was invented in its place. Reload the page.
This page's interactive half has not started on this device. Nothing was read, and nothing was invented in its place. Reload the page.
Basketball · Reference
Every metric a basketball screen prints, defined in one sentence, with a second line on how to read it. If a number on a board, a dossier or a trade verdict has ever needed explaining, its explanation is on this page.
The definitions here are the same words the tooltips use — hover a metric anywhere in basketball and the tooltip will bring you back to its entry.
346 terms · 10 groups
What a player is worth, who says so, and where the price disagrees with the play.
The single dynasty price we put on a player, built from published expert boards and weighted so a recent list counts for more than an old one. It is the number every trade tool, board row and portfolio total is denominated in.
It is a price, not a projection. A player can be worth a lot because the market believes in him and still be projected to score less than the man below him.
See also Market Value
The blend of published dynasty boards behind Master Value. Each list contributes a rank, ranks decay with age so a stale board fades rather than vanishes, and the blend is what the price is made of.
When only one or two boards are recent, the consensus is one or two opinions wearing a crowd's clothes — the board says so on the row.
Where a player sits when every player is sorted by Master Value, best first.
Rank compresses at the top and stretches in the middle: the gap between #2 and #6 is usually larger in value than the gap between #60 and #90.
Where our own engine puts a player, independent of what the boards say. It sits beside Market Rank so the two can disagree in public.
The gap is the point. A large positive gap is the engine saying the market is late; it is a lead to check, never a verdict on its own.
How fast a player is climbing the market measured against how much he has actually proven. The dynasty gap compares recent movement across expert boards with his production and projection percentile for his age and role.
Positive means the price is moving faster than the play. A player with no professional sample is unproven, never hyped — he has nothing to be measured against yet.
The rank change a player has made across published boards over the last two and six weeks, weighted by how many boards moved him.
One board moving a player ten spots is noise; six boards moving him three each is a trend.
The set of players with enough on-court sample to be ranked on production, used as the yardstick the Hype Gap measures a riser against.
Being outside the pool is a statement about sample, not about talent.
The two verdicts on the trade desk. Buy Low means our read of a player is above what the market is charging; Sell High means the market is paying more than our read supports.
Both are about PRICE, not about whether a player is good. A great player can be a sell and a flawed one a buy.
A band of players whose value ranges overlap enough that picking between them is a preference rather than an edge. Tiers are cut where the ranges stop overlapping, not every ten rows.
Trade across a tier break, not within one. Inside a tier you are swapping flavours of the same price.
How far apart the published boards are on one player.
Wide dispersion is disagreement, and disagreement is where a trade is available. It is not the same as risk.
The range of total value either side of a trade can be worth once every player's uncertainty is carried through. A trade whose two sides overlap inside the band is fair.
A band that overlaps is not a tie — it means the difference is smaller than what we can honestly measure.
The step-by-step account of how a trade verdict was reached: each asset's value, each adjustment, and the running total after each one.
It is the receipt. If a verdict surprises you, the step that moved the total most is the disagreement.
The history of one asset's value, with the trades that moved it marked on the line.
A flat ledger with a big jump is usually a role change, not a market mood.
Where a future draft pick is most likely to land, given the owning team's projected finish and the odds attached to it.
A pick is a distribution, not a slot. Two picks with the same expected landing spot can carry very different upside.
The chance a future pick lands in the lottery range rather than in the pack.
It is the upside half of a pick's price. A low expected landing spot with high lotto odds is a swing, not a certainty.
How many seasons a roster projects to be able to win, from the ages, roles and multi-year values of the players on it.
The window is the verdict a dynasty roster is graded against — a move is good or bad relative to when you intend to win.
How much of a player's ability his current situation actually lets him use — the distance between what he can do and what his role asks of him.
A large unused gap is the cheapest kind of upside, because it needs a role change rather than a leap.
What we think happens next, and how sure we are — every projection ships with its range.
Our forecast of a player's rate of production over a stated horizon, anchored on his multi-year career rather than on his most recent season.
A down or injured season is pulled back toward the career norm, never projected forward, unless the decline is a real age effect rather than a fluke.
The window a projection covers — rest of season, next season, or the season after.
Never compare two projections without checking they cover the same window.
The range a projection is most likely to land in: eight times out of ten the real number falls between the low and high edge.
A wide cone is not a bad projection — it is an honest one. Two players with the same middle and different cones are different bets.
How much history, role stability and sample stand behind a projection, stated so a thin one cannot pass as a firm one.
Low confidence widens the cone; it does not move the middle.
How much of a recent hot or cold stretch the underlying inputs support — shot quality, role, usage — rather than shooting luck.
Low sustainability on a hot run is the sell signal the box score will not give you for another month.
The minutes a player projects to play, built from role, depth ahead of him, usage and availability — not from age alone.
Minutes are the largest single lever on any counting projection: a rate that holds and five more minutes is a different player in your lineup.
The games a player projects to actually play across the season.
Availability is modelled from his own history, not assumed from last year's total, and it is the input most likely to decide a category league.
See also Games Played
What happens to everyone else's minutes and usage when a player is out — the production his absence hands to each teammate.
It is a redistribution, not a promotion: the touches go somewhere, and rarely all to one player.
A player's projected production one and two seasons out, which is what a dynasty roster is actually being priced on.
Read it as a shape, not a number: rising, flat and falling lines price very differently at the same value today.
The best season a player projects to have, and roughly when it arrives.
A peak two seasons out is an argument for holding; a peak already behind him is an argument for a price.
See also Peak Performance
The low and high edges of a player's plausible season — the bad outcome you can live with and the good one you are paying for.
They are the ends of the same cone. Quoting one without the other is how a bet gets mis-sold.
See also Performance Floor, Ceiling Projection
The chance a player finishes far above or far below where he is being drafted, from the same model that produced his cone.
A high breakout chance and a high bust chance usually travel together — that is one volatile player, not two signals.
Running an engine against a season that has already happened, with only what was known at the time, and scoring what it said against what occurred.
A backtest that was not stored per player before the season is a memory, not a measurement.
How closely a predicted order matched the real finishing order, from 1.00 for a perfect match down to 0.00 for no relationship.
Compare it against the simple alternative — repeating last season — not against zero. Beating nothing is not an edge.
How far a projection missed by on average, in the stat's own units.
It treats a miss high and a miss low the same. A model can have a small average error and still be biased in one direction.
Category and points scoring, punt builds, draft prices and the shapes a league can take.
The standard category format: points, rebounds, assists, steals, blocks, threes, field-goal percentage, free-throw percentage and turnovers.
Turnovers count against you and the two percentages are weighted by volume — a great percentage on four attempts barely moves a week.
How many standard deviations above or below the league's average a player is in one category, so unlike stats can be added together.
Zero is average, not bad. A player at +1.0 in five categories beats a player at +3.0 in one for almost every roster.
The percentage categories scored by how much a player moves YOUR team's percentage, which depends on his attempts as much as his accuracy.
This is why a high-volume average shooter can hurt more than a low-volume poor one.
Conceding a category on purpose and spending the roster spots it would have cost on the categories you intend to win.
A punt only works once the remaining categories are re-scored without it — a player's value changes the moment a category stops counting.
How your roster projects in each category against the rest of the league.
Read it beside the schedule: a category you win narrowly is a category a thin week can lose.
Your record if you had played every other team every week instead of only your scheduled opponent.
It is the cleanest read on whether a record is deserved or scheduled.
The projected standings points a roster earns across every category in a rotisserie league at season's end.
Roto rewards the last unit in a category you are close in far more than the first unit in one you already own.
A format where every stat converts to points on the league's own scale and the weekly total wins.
The scale is the format. A league that pays three for a steal and minus one for a turnover ranks players differently to one that does not.
A player's production scaled to a full starter's workload of thirty-six minutes.
It answers what he does when he plays, never whether he will play. Pair it with the minutes projection or it will lie to you.
What the best freely available player in your league would give you — the bar every rostered player is actually competing with.
It moves with league size and roster depth; a twelve-team bar and a twenty-team bar are different players.
A player's projected production expressed as a share of an auction budget.
Values only balance against a stated budget and roster size; carried into a different auction they are a ranking, not a price.
Where a player is actually being drafted, averaged across real drafts.
ADP is a market, not an opinion — it tells you the price of a player, and each platform's drafters price differently.
What keeping a player costs next season, in the pick or the salary your league charges.
A keeper is only good at its cost — the same player is a bargain at one round and a trap at another.
The words your own league uses — weeks, waivers, keepers, eligibility, playoff windows.
The block of days one matchup is scored over, taken from your league's own calendar rather than from a generic Monday.
Platforms cut weeks differently, and a week that starts on the wrong day miscounts every game in it.
How many games each of your players is scheduled to play inside the scoring week.
In category and points leagues a four-game week from a lesser player often beats a two-game week from a better one.
A game played on the day after another game by the same team.
It is a rest risk, not a certainty — treat it as a reason to check a status, not as a benching on its own.
The scoring weeks your league's postseason is played over.
They are the only weeks a title is decided in — schedule strength in them is worth more than season-long strength.
The lineup slots a player may be started in, as your own platform rules them.
Eligibility is per platform, not universal. Multi-slot players are worth more than their ranking says because they unlock other players' starts.
The season-long budget you bid from to claim free agents.
A bid is only expensive relative to the weeks left to spend it in.
Cycling a roster spot through free agents to harvest extra games, usually chosen by schedule rather than by talent.
It is a games-played strategy. It wins categories that reward volume and does nothing for percentages.
Team strength ordered by projected production rather than by record.
It ships with a band; two teams whose bands overlap are not separated by their ranking.
The share of simulated seasons in which your roster wins the league, given its projections and the schedule ahead.
The useful reading is the CHANGE a move makes, not the level — a two-point swing is the same information at 8% as at 38%.
Plus-minus families, possession math and the rate stats behind every board.
A single number for a player's impact per 100 possessions, blending what the box score says with what the scoreboard did while he played.
Zero is a league-average player, not a bad one. It is a rate — a large number in few minutes is a smaller contribution than it looks.
A daily-updating estimate of impact per 100 possessions that responds to recent play faster than a season figure does.
It moves quickly by design, which makes it early on a role change and jumpy on a hot week.
See also Total DPM
Impact measured purely from the scoreboard across every stint, adjusted for teammates and opponents and pulled toward average where the sample is small.
It sees things the box score cannot and needs seasons, not weeks, to say them with confidence.
See also RAPM
Points scored minus points allowed per 100 possessions while a player or lineup is on the floor.
It is a team result attributed to whoever was standing there. Five-man samples are small — read the minutes beside it.
The difference between how a team performs with a player on the floor and without him.
It is as much about who replaces him as about him. A strong on/off with a weak backup is partly the backup's fault.
How a player performs with a given teammate on the floor compared with without him.
It is the pairing question, and it needs a real shared sample — a handful of games together says nothing.
How much better or worse a five-man unit plays than the sum of what its five players do individually.
A positive number is a fit story rather than a talent one, and it lives in small samples — treat a single unit's figure as a lead.
One team's turn with the ball, which is the unit almost every rate stat is divided by.
Per-possession stats let a fast team and a slow team be compared honestly; totals do not.
See also POSS+
Possessions used per 48 minutes — by a team across a season, or by a team while one player is on the floor, which is the figure a player card prints.
Pace inflates every counting stat without making anyone better. A high-pace team's totals need discounting before comparison.
See also Pace Impact
An unbroken run of floor time by one set of five players, ended by any substitution.
Stints are the raw material of lineup and rotation reads; a lineup's whole season can be a handful of them.
Floor time late in a decided game, when both teams' choices stop being about winning.
Removing it changes deep-bench numbers most, and barely moves a starter's.
The chance a team wins from the current score, time and possession.
It is a statement about the game state, not about how well anyone is playing.
The closing minutes of a close game, split out so late performance can be read on its own.
Clutch samples are tiny for everyone. A season of clutch minutes is a few games of ordinary ones.
A stretch in which one team scores heavily against the other without reply.
A run is a description of what happened; who was on the floor for it is the part worth keeping.
Who a defense makes life hard for, and how heavy the week ahead actually is.
How much fantasy production a defense gives up to each position relative to the league.
It is a team result, not an assignment — the defender who actually guards your player matters more than the team average.
Defense vs Position corrected for who a defense has actually faced, so a soft schedule cannot masquerade as a strong defense.
Use the adjusted figure early in a season, when unbalanced schedules distort the raw one most.
How the specific player who guards your player has performed against that kind of scorer.
This is the read DvP is a rough proxy for, and it is the one that survives a coaching change.
How hard the week ahead is for one player, from the defenses he is scheduled to face.
It is a nudge, not a verdict — a difficult week for a starter is still a start.
The words a screen uses when it will not print a number, and what each one promises.
A number computed on too little play to be stable. The screen prints the value with its sample in plain words rather than hiding it.
A thin number is directional. It is labelled, never withheld — the only value we withhold is one that does not exist yet.
The state a screen shows when a number genuinely cannot be computed — nothing has happened yet to compute it from.
Withheld is never a zero and never a dash that reads as one. The sentence beside it says what is missing.
A player with no professional sample to measure against.
Unproven is a statement about evidence, not about talent, and it is the honest label for a rookie the market already prices highly.
A value that was correct when it last landed but has not been refreshed since it should have been.
A stale screen shows you the clock rather than pretending the number is current.
The clock on a number — when it last landed and what it is current through.
It is always a real timestamp from the data. A clock that reads the current time is a defect, not a freshness stamp.
The share of the comparison group a player is above in one metric, from 0 to 100.
A percentile is meaningless without its group — among all players, among starters and among his position are three different answers.
The share of the players on a screen for whom a given metric actually exists.
Partial coverage is stated rather than filled in — a missing player is left missing, never averaged into place.
The counting and rate stats every table is built from, per game, per 100 and per 75.
Player's current age. Important for evaluating development trajectory and remaining prime years.
NBA prime typically 25-30. Under 24 = still developing. Over 32 = potential decline. Scale 19-42.
Total games played this season. Important for sample size and availability.
More games = more reliable stats. 50+ games is a solid sample. Scale 0-82.
See also Availability
Average minutes played per game. Indicates role and importance to the team.
30+ = starter/heavy minutes. 20-30 = rotation. Under 20 = limited role. Scale 0-40.
Average points scored per game. Traditional scoring output measure.
25+ is elite scorer. 15-25 is good. Under 10 is role player. Scale 0-40.
Average assists per game. Measures playmaking output.
8+ is elite playmaker. 5-8 is good. Under 3 is off-ball player. Scale 0-15.
Average total rebounds per game.
10+ is elite rebounder. 6-10 is good for a big. Guards typically 3-5. Scale 0-18.
Average steals per game. Indicates perimeter disruption.
1.5+ is elite. 1.0+ is good. Shows active hands and anticipation. Scale 0-3.
Average blocks per game. Indicates rim protection ability.
2.0+ is elite shot blocker. 1.0+ is solid rim presence. Scale 0-4.
Points scored per 100 possessions. Pace-adjusted scoring rate.
30+ is elite scoring rate. 20-30 is good. Removes pace bias from PPG. Scale 0-50.
Assists per 100 possessions. Pace-adjusted playmaking rate.
10+ is elite playmaking. Removes pace bias from raw assist numbers. Scale 0-20.
Offensive rebounds per 100 possessions.
4+ is elite offensive rebounder. Shows effort and positioning. Scale 0-10.
Defensive rebounds per 100 possessions.
10+ is elite defensive rebounder. Scale 0-20.
Total rebounds per 100 possessions.
15+ is elite total rebounding. Scale 0-25.
Steals per 100 possessions. Pace-adjusted steal rate.
2.5+ is elite ball hawk. Scale 0-5.
Blocks per 100 possessions. Pace-adjusted shot blocking.
3+ is elite rim protector. Scale 0-8.
Turnovers per 100 possessions.
Lower is better. Under 2 is excellent. Over 3 is concerning. Scale 0-6.
Personal fouls per 100 possessions.
Lower is better. Under 3 is disciplined. Over 5 is foul prone. Scale 0-8.
Field goal attempts per 100 possessions. Shows shot volume.
20+ is high-volume shooter. Context matters - need efficiency too. Scale 0-30.
Free throw attempts per 100 possessions.
8+ means getting to the line consistently. Scale 0-15.
3-point attempts per 100 possessions.
10+ is high-volume 3PT shooter. Scale 0-18.
Rim attempts per 100 possessions. Shows paint presence.
8+ is rim attacker. Shows ability to get to the basket. Scale 0-15.
Percentage of team possessions used while on court. How much the offense runs through this player.
30%+ is high-usage star. 20-30% is starter. Under 20% is complementary. Scale 0-40%.
Percentage of teammate FGM assisted while on court.
35%+ is primary playmaker. 20-35% is secondary. Under 20% is off-ball. Scale 0-55%.
Turnovers per 100 plays (FGA + 0.44×FTA + AST + TOV).
Lower is better. Under 10% is excellent. Over 15% is concerning. Scale 0-25%.
Percentage of available offensive rebounds grabbed while on court.
10%+ is elite offensive rebounder. Shows effort and positioning. Scale 0-20%.
Percentage of available defensive rebounds grabbed while on court.
25%+ is elite defensive rebounder. Scale 0-40%.
Percentage of opponent possessions that end with a steal while on court.
2%+ is elite ball hawk. Scale 0-4%.
Percentage of opponent 2PT attempts blocked while on court.
5%+ is elite rim protector. Scale 0-12%.
Field goal percentage. Basic shooting efficiency.
50%+ is good. Context matters - shot selection affects this. Scale 0-70%.
2-point field goal percentage.
55%+ is good. 60%+ is elite inside scorer. Scale 0-75%.
3-point field goal percentage.
38%+ is good. 40%+ is elite. League avg ~36%. Scale 0-50%.
Free throw percentage.
80%+ is good. 85%+ is elite. Below 70% is poor. Scale 0-100%.
Field goal percentage at the rim (within 4 feet).
65%+ is good finisher. 70%+ is elite. Scale 0-85%.
True shooting percentage. Accounts for 2s, 3s, and FTs in one number.
60%+ is efficient. 65%+ is elite. League avg ~58%. Scale 0-75%.
Effective field goal percentage. Adjusts FG% for 3-pointers being worth more.
55%+ is efficient. 60%+ is elite. Scale 0-70%.
Percentage of field goal attempts that are 3-pointers.
50%+ is 3PT specialist. Shows shot profile. Scale 0-100%.
Free throw attempts per field goal attempt. Shows ability to draw fouls.
0.35+ is elite at getting to the line. Scale 0-0.6.
Points per 75 possessions (EPM's pace-adjusted rate).
20+ is elite scoring. ~20 possessions = starter minutes. Scale 0-35.
Assists per 75 possessions.
7+ is elite playmaker. Shows passing rate in starter-like minutes. Scale 0-15.
Rebounds per 75 possessions.
10+ is elite rebounder. Scale 0-18.
Steals per 75 possessions.
1.5+ is elite ball hawk. Scale 0-4.
Blocks per 75 possessions.
2+ is elite shot blocker. Scale 0-5.
Turnovers per 75 possessions.
Lower is better. Under 2 is excellent ball security. Scale 0-5.
Total touches per game. How often the ball is in their hands.
70+ is high-touch player. Shows involvement in offense. Scale 0-120.
Drives to the basket per game.
10+ is aggressive driver. Shows ability to attack the rim. Scale 0-25.
Points scored from drives per game.
8+ is elite scorer off drives. Scale 0-15.
Total passes made per game.
40+ is heavy ball mover. Shows offensive involvement. Scale 0-80.
Assists on baskets where the scorer was assisted.
1.5+ shows good ball movement contribution. Scale 0-4.
Passes that would be assists if converted.
10+ is elite passer. Shows playmaking ability regardless of teammates. Scale 0-20.
Assists adjusted for teammate shooting. Better for bad teams.
More accurate than raw assists for cross-team comparison. Scale 0-15.
Deflections per game. Tipping passes or dribbles.
3+ is elite disruptor. Shows defensive activity. Scale 0-6.
Loose balls recovered per game.
1+ shows hustle and motor. Scale 0-3.
Charges drawn per game.
0.3+ shows good positioning and sacrifice. Scale 0-1.
Screen assists per game (setting screens that lead to scores).
2+ is valuable screener. Important for role players. Scale 0-6.
Shots contested per game.
8+ shows defensive effort. More = more active defender. Scale 0-18.
Estimated wins added above replacement level, the EPM wins figure.
8+ is All-Star level. 15+ is MVP level. Scale 0-20.
Total field goal attempts defended this season.
More = larger sample size. 400+ is meaningful sample. Scale 0-1000.
Opponent FG% when defended by this player.
Lower is better. Under 42% is elite overall defender. Scale 30-60%.
Difference between opponent FG% when defended vs their average.
Negative is better. -5% means holding opponents 5% below average. Scale -15% to +15%.
Rim attempts (within 6 feet) defended this season.
More = primary rim protector. 200+ is meaningful sample. Scale 0-500.
Opponent FG% at the rim when defended by this player.
Lower is better. Under 55% is elite rim protector. Scale 40-75%.
Difference between opponent rim FG% vs their average.
Negative is better. -8% is elite. Indicates paint deterrence. Scale -20% to +20%.
Percentage of defense played at the rim.
30%+ means primary rim protector. Shows where they defend. Scale 0-50%.
Non-rim paint attempts defended this season.
Paint protection volume (floaters, short midrange). Scale 0-400.
Opponent FG% in the paint when defended.
Lower is better. Under 40% is excellent paint defender. Scale 30-55%.
Difference between opponent paint FG% vs their average.
Negative is better. Shows floater/short shot deterrence. Scale -15% to +15%.
Percentage of defense played in the paint.
Shows interior defensive responsibility. Scale 0-40%.
Perimeter (midrange) attempts defended this season.
Perimeter defense volume. Scale 0-400.
Opponent FG% on perimeter shots when defended.
Lower is better. Under 38% is strong perimeter defender. Scale 30-50%.
Difference between opponent perimeter FG% vs their average.
Negative is better. Shows midrange deterrence. Scale -15% to +15%.
Percentage of defense played on perimeter.
Shows where player defends most. Scale 0-50%.
3-point attempts defended this season.
3PT defense volume. 200+ is meaningful sample. Scale 0-500.
Opponent 3PT% when defended by this player.
Lower is better. Under 33% is elite perimeter stopper. Scale 25-45%.
Difference between opponent 3PT% vs their average.
Negative is better. -5% is elite. Note: 3PT defense has high variance. Scale -15% to +15%.
Percentage of defense played on 3PT shooters.
40%+ means defending primarily on the perimeter. Scale 0-60%.
Average expected eFG% of shots taken.
55%+ means taking good shots. Shows shot selection quality. Scale 40-65%.
FG% on shots with <4 seconds on shot clock.
40%+ is clutch shot maker in difficult situations. Scale 20-50%.
And-1 plays per 100 rim attempts.
10+ shows ability to finish through contact. Scale 0-20.
Percentage of steals from intercepting passes.
Higher = better anticipation. Shows reading the game. Scale 0-100%.
Percentage of steals from stripping ball handlers.
Higher = more aggressive/gambling defense. Scale 0-100%.
Percentage of turnovers that are live ball (lead to fastbreaks).
Lower is better. Live ball TOs are more costly. Scale 0-100%.
Percentage of blocks where team recovers possession.
45%+ means valuable blocks. Blocks that go out of bounds hurt value. Scale 0-70%.
Assists on shots at the rim (highest value assists).
40%+ of assists at rim is elite lob/cutting creation. Scale 0-100%.
Points from putbacks per game.
2+ shows motor and positioning for second chances. Scale 0-6.
Own missed shots rebounded per game.
0.5+ shows effort to follow shots. Scale 0-2.
Points on second chance opportunities per game.
3+ is valuable second-chance scorer. Scale 0-8.
Raw contested offensive rebounds per game. Boards grabbed with defenders nearby.
2+ is excellent hustle. 1+ is solid. Shows second-chance effort. Scale 0-5.
Raw contested defensive rebounds per game. Boards won in traffic.
4+ is strong. Shows ability to secure boards vs opposing crashers. Scale 0-10.
Percentage of rebounds that were contested vs uncontested.
Higher = wins more battles. Very high can indicate role (garbage time = more uncontested). Scale 0-100%.
Average distance traveled to grab rebounds (in feet). Effort indicator.
Higher = crashes from distance (effort). Very low = only grabs easy ones. Scale 0-15 ft.
Rebound chance conversion - boards grabbed divided by rebound opportunities.
Higher = efficient at converting chances. Low = poor positioning or awareness. Scale 0-100%.
Total rebound chances (opportunities to grab a board) per game.
Shows opportunity volume. Context for raw rebound numbers. Scale 0-20.
The same box and rate words measured on a college season, where the bar for elite sits somewhere else entirely.
Box score estimate of points contributed per 100 possessions above a replacement-level player.
+8 is elite (NBA lottery pick level). +5 is All-Conference. 0 is average D1 player. Scale -10 to +15.
Points produced over a replacement-level player per adjusted game. The primary all-in-one college value metric.
+6 is elite. +3 is very good starter. 0 is replacement level. Scale -5 to +12.
Offensive box plus/minus - points contributed on offense per 100 possessions.
+6 is elite offensive player. +3 is very good. 0 is average. Scale -5 to +12.
Defensive box plus/minus - points prevented on defense per 100 possessions.
+3 is elite defender. +1 is above average. Negative = defensive liability. Scale -5 to +8.
Average points scored per game.
20+ is elite scorer. 15+ is very good. 10+ is solid contributor. Scale 0-30.
Percentage of team plays used while on floor (shots, assists, turnovers). Measures offensive role/burden.
30%+ is primary option. 25% is secondary. 20% is role player. Scale 10-40%.
Shooting efficiency accounting for 2-pointers, 3-pointers, and free throws.
60%+ is elite. 55%+ is good. Below 50% is inefficient. Scale 40-70%.
Field goal percentage adjusted for 3-pointers being worth more. (FGM + 0.5 × 3PM) / FGA.
55%+ is elite. 50%+ is good. Below 45% needs improvement. Scale 35-70%.
Two-point field goal percentage (rim + mid-range shots combined).
55%+ is elite. 50%+ is good. Includes both rim finishing and mid-range shooting. Scale 35-70%.
Three-point field goal percentage.
38%+ is elite. 35%+ is good. Below 30% is a weakness. Scale 25-45%.
Free throw percentage.
80%+ is good. 85%+ is elite. Below 65% is concerning for late-game situations. Scale 50-95%.
Field goal percentage on shots at the rim (within ~4 feet).
65%+ is elite finisher. 60%+ is good. Shows athleticism and touch. Scale 40-80%.
Average assists per game.
6+ is elite playmaker. 4+ is very good. 2+ is solid secondary playmaker. Scale 0-10.
Percentage of teammate field goals assisted while on floor.
30%+ is elite playmaker. 20%+ is good passer. Below 10% is off-ball player. Scale 5-45%.
Ratio of assists to turnovers. Measures ball security and playmaking efficiency.
2.5+ is elite decision maker. 1.5+ is good. Below 1.0 is careless. Scale 0.5-4.0.
Turnovers per 100 plays. How often possessions end in turnovers.
Below 12% is excellent. 15% is average. Above 20% is problematic. Scale 5-25%.
Average total rebounds per game (offensive + defensive).
10+ is elite rebounder. 7+ is very good. 5+ is solid. Scale 0-15.
Percentage of available offensive rebounds grabbed while on floor.
12%+ is elite. 8%+ is good. Shows hustle and positioning. Scale 0-20%.
Percentage of available defensive rebounds grabbed while on floor.
25%+ is elite. 18%+ is good. Important for ending opponent possessions. Scale 5-35%.
Average steals per game.
2+ is elite ball hawk. 1.5+ is very active. Shows defensive instincts. Scale 0-3.
Average blocks per game.
2+ is elite rim protector. 1+ is good. Rare outside of bigs. Scale 0-4.
Percentage of opponent possessions ending in a steal by this player.
3%+ is elite. 2%+ is very good. Shows anticipation and quick hands. Scale 0-5%.
Percentage of opponent 2-point attempts blocked while on floor.
8%+ is elite rim protector. 4%+ is good. Mainly a big man stat. Scale 0-15%.
Points produced per 100 possessions used. Individual offensive efficiency.
125+ is elite. 115+ is good. Below 100 is inefficient. Scale 80-150.
Points allowed per 100 possessions while on floor.
Below 95 is elite. Below 100 is good. Above 105 is concerning. Scale 70-120.
Free throw attempts per field goal attempt. Measures ability to get to the line.
40%+ is elite at drawing fouls. 30%+ is good. Below 20% avoids contact. Scale 10-60%.
The scored traits behind the player DNA tables — each one a 0–100 read on a single skill.
Overall player impact combining box score production, plus-minus impact, and two-way contribution. The single best measure of total value.
Higher = more impactful player. 90+ is elite, 70+ is starter-quality, 50+ is rotation player. Scale 0-100.
Measures primary scorer/leader traits: high usage combined with on-court impact. Identifies players who can be the focal point of an offense.
High Alpha = can be "the guy" in crunch time. Low Alpha = better as complementary piece. Scale 0-100.
Balance between offensive and defensive impact. Rewards players who contribute meaningfully on both ends rather than being one-dimensional.
High Two-Way = complete player. Low = specialist (either offense or defense focused). Scale 0-100.
How unique is this player's skill combination? Measures statistical rarity compared to peers at the same position.
High = rare skill set (hard to replace). Low = replaceable skillset. Scale 0-100.
Offensive engine capability: playmaking, usage, and pace impact combined. Identifies players who drive the offense.
High = can run an offense. Low = plays off others. Scale 0-100.
Offensive Daily Plus-Minus. Points per 100 possessions the team scores better with this player on court (vs replacement).
+3.0 is elite offense. +0.0 is average. Negative = hurts the offense. Scale -10 to +10.
Defensive Daily Plus-Minus. Points per 100 possessions the team allows fewer with this player on court.
+2.0 is elite defense. +0.0 is average. Negative = defensive liability. Scale -10 to +10.
Combined offensive and defensive plus-minus. Net point differential per 100 possessions with player on court.
+5.0 is MVP-level. +2.0 is All-Star. +0.0 is average. Negative = net negative. Scale -10 to +10.
See also Daily Plus-Minus
Regularized Adjusted Plus-Minus. Team point differential when player is on court vs off, adjusted for teammates and opponents.
The gold standard for "real impact." Captures things box scores miss. Scale -10 to +10.
See also Regularized Adjusted Plus-Minus
How this player affects game tempo. Positive = games are faster, negative = games are slower.
Not good or bad—just style. Some players naturally speed up or slow down the game. Scale -1 to +1.
See also Pace
Hidden impact not captured by box scores. Calculated as On/Off impact minus Box score contribution. Shows "intangible" value.
High = does things that don't show up in stats (screens, gravity, communication). Low/negative = box stats overstate impact. Scale -5 to +5.
How much does the team improve when this player is on court? Pure on/off differential.
High = teammates play better around them. Low = may be stat-padding in bad lineups. Scale -15 to +15.
Offensive attention this player commands. Draws defenders and creates space for teammates.
High = defense collapses on them (creates open shots for others). Low = can be ignored. Scale 0-100.
Anchors the defense. Combines on-court defensive impact with rim protection activity.
High = defense is built around them. Low = needs help defensively. Scale 0-100.
How this player affects game pace. Positive = games are faster when they play. Negative = games slow down.
Not good or bad—just style. Fast = transition-heavy. Slow = half-court focused. Scale -1 to +1.
Offensive pace impact. Positive = pushes the ball, plays fast. Negative = slows down, runs half-court sets.
High = transition player. Negative = methodical, deliberate offensive player. Scale -1 to +1.
Ability to slow down opposing offenses. Forces opponents into half-court sets.
High = slows opponents down. Not inherently good or bad—depends on team style. Scale -1 to +1.
Offensive pace style. Positive = plays fast, pushes tempo. Negative = methodical, half-court player.
Not good or bad—just style. Fast suits some systems, slow suits others. Scale -5 to +5.
Transition offense effectiveness. Combines points per possession in transition with frequency.
High = lethal in the open court. Low = half-court player. Scale 0-100.
How much of the offense goes through this player. Combines usage rate, shot attempts, and minutes.
High = primary option. Low = role player/specialist. Scale 0-100.
True shooting percentage relative to league average. How efficiently they score compared to peers.
+5 is very efficient. 0 is average. Negative = below-average efficiency. Scale -20 to +20.
Ability to attack and finish at the rim. Combines rim attempt rate with finish percentage.
High = constant paint threat. Low = perimeter-oriented or finesse player. Scale 0-100.
Three-point threat level. Combines 3PT volume with efficiency to measure spacing value.
High = must be guarded at the 3PT line. Low = defenders can sag off. Scale 0-100.
Total rebounding per 100 possessions. Weighted 2x for offensive rebounds (harder to get).
High = dominant rebounder. Controls the glass on both ends. Scale 0-20.
Steals + blocks per 100 possessions. Combined defensive playmaking.
High = creates turnovers and blocks shots. Active defensive presence. Scale 0-8.
Steal rate per 100 possessions. Ability to read passing lanes and create turnovers.
High = disruptive, forces bad passes. Elite anticipation. Scale 0-5.
Block rate per 100 possessions. Rim protection and shot-altering ability.
High = changes shots at the rim. Opponents think twice about attacking. Scale 0-5.
Pure passing ability. Assist percentage adjusted for turnover tendency.
High = elite passer, creates for others. Low = scorer or off-ball player. Scale 0-50.
Turnover rate adjusted for usage. How often possessions end in turnovers relative to their offensive load.
High = careless with the ball. Low = protects the rock. Scale 0-100.
Free throw drawing ability. Combines free throw rate with attempts per 100 possessions.
High = gets to the line often (valuable in close games). Low = avoids contact. Scale 0-100.
Midrange shooting prowess. Two-point percentage excluding rim attempts.
High = reliable midrange scorer. Valuable in playoffs when 3s aren't falling. Scale 0-100.
Average of the last 7 snapshot days minus the season-to-date average of the same daily ratings. Positive = playing above their norm recently.
High positive = hot streak. Negative = cold streak or returning from injury. A dash means fewer than 7 snapshot days -- too thin to call. Scale -5 to +5.
Currently on a hot streak: the last 7 snapshot days average more than 1.5 rating points above the season-to-date average.
True = ride the hot hand. False = normal/cold performance. Scale Yes/No.
Season-to-date direction: change in daily rating from first snapshot day to latest, divided by the number of snapshot days. Positive = improving.
Positive = trending up (breakout candidate). Negative = slumping. Scale -10 to +10.
How much performance fluctuates game-to-game. Standard deviation of daily ratings.
High = inconsistent (might dominate or disappear). Low = reliable every night. Scale 0-100.
Opposite of variance. How reliable is performance night-to-night?
High = you know what you're getting. Low = volatile performer. Scale 0-100.
Highest single-game performance vs season average. Shows ceiling potential.
High = has shown dominant upside. Low = limited ceiling. Scale 0-10.
See also Peak
Worst game performance below season average. Shows how bad they can be.
High = significant dropoffs possible. Low = doesn't have terrible games. Scale 0-10.
See also Floor and Ceiling
Overall scoring ability. Points per 100 possessions adjusted for true shooting efficiency.
High = elite scorer. Combines volume with efficiency. Scale 0-100.
Ability to create shots for self and others. Combines assists with unassisted scoring.
High = can generate offense from nothing. Low = needs plays run for them. Scale 0-100.
Can score from all three zones: rim, midrange, and 3-point. Average efficiency across all areas.
High = unguardable, can score from anywhere. Low = predictable, has weak spots. Scale 0-70%.
Finishing ability at the rim. Combines rim FG% with rim attempt volume.
High = automatic at the rim. Low = struggles in traffic. Scale 0-100.
Three-point shooting prowess. 3PT% weighted by volume attempted.
High = elite shooter, stretches the floor. Low = can be left open. Scale 0-100.
How much offensive burden this player carries. Usage rate times minutes played.
High = heavy workload (primary option). Low = limited role. Scale 0-100.
Rim protection ability. Block rate combined with defensive impact at the rim.
High = paint presence, deters drives. Low = can be attacked inside. Scale 0-100.
Perimeter defensive disruption. Steal rate combined with defensive impact rating.
High = harasses ball handlers, creates turnovers. Low = can be exploited on the perimeter. Scale 0-100.
Point-of-attack defense. Ability to defend primary ball handlers one-on-one.
High = can guard stars. Low = needs help on ball handlers. Scale 0-100.
Pure rim protection. Block rate and defensive impact at the basket.
High = opponents avoid the paint. Low = rim is exposed. Scale 0-100.
Percentage of possessions ending without opponent scoring when defending. Steals + blocks + forced misses.
High = ends possessions on defense. Low = opponents score efficiently. Scale 0-10%.
Contested shots per 36 minutes. Measures defensive activity and effort.
High = active, contests everything. Low = lazy or poor positioning. Scale 0-15.
Net possessions created per 75 possessions. Steals + offensive rebounds + drawn fouls - turnovers.
Positive = creates extra possessions. Negative = loses possessions for the team. Scale -2 to +3.
Second-chance opportunities created. Offensive rebounds plus steals minus turnovers.
High = generates extra possessions. Vital for winning close games. Scale -5 to +8.
Turnover cost of shot creation. Turnovers per assist and unassisted field goal.
Low = efficient creator (few mistakes). High = turnover-prone creator. Scale 0-1.
Inverse of Creation Tax. How many shots created (assists + unassisted FGM) per turnover.
High = creates offense without mistakes. Low = careless with the ball. Scale 0-10.
Turnover rate adjusted for offensive burden. Accounts for high-usage players naturally having more turnovers.
High = careless relative to their role. Low = handles responsibility well. Scale 0-30.
How much of the offensive load this player carries. Usage times minutes.
High = primary option, heavy workload. Low = role player. Scale 0-15.
Estimated wins contributed above replacement level. Based on RAPM and minutes played.
10+ is All-Star level. 5+ is quality starter. 0 is replacement level. Scale -5 to +15.
Total value above a replacement-level player. Cumulative impact measure.
5+ is star level. 2+ is solid starter. 0 is replacement level. Scale -5 to +10.
Ben Taylor's creation metric. Measures shot creation combining assists, scoring, and turnover avoidance.
8+ is elite creator. 5+ is quality playmaker. Under 3 is off-ball player. Scale 0-15.
Estimated market value based on impact and age. Combines Crown score with contract value factors.
High = worth a max contract. Low = minimum-level value. Scale 0-200.
See also Master Value
Projected peak performance. Based on best 7-day stretch with age adjustment.
High = high upside (development candidate). Low = what you see is what you get. Scale 0-150.
See also Floor and Ceiling
Estimated wins, the EPM (Estimated Plus-Minus) wins figure. Similar to Win Shares.
Higher = more valuable player. Scale 0-20.
Box score estimate of plus-minus. Points per 100 possessions better than average.
+5 is elite. +2 is very good. 0 is average. Negative = below average. Scale -10 to +15.
Per-minute rating that summarizes all box score contributions. League average is 15.
25+ is elite. 20+ is All-Star. 15 is average. Under 12 is poor. Scale 0-35.
Assists divided by turnovers. Pure ball security measure for playmakers.
3+ is elite. 2+ is good. Under 1.5 is turnover-prone. Scale 0-5.
Points scored per true shot attempt (FGA + 0.44*FTA). Pure scoring efficiency.
1.2+ is efficient. 1.0 is average. Under 1.0 is inefficient. Scale 0-2.
Points per possession on isolation plays. How efficiently they score 1-on-1.
1.0+ is elite. 0.85-1.0 is good. Under 0.85 = should not be isolating. Scale 0-2.
Points per possession as pick & roll ball handler. Core skill for modern guards.
1.0+ is elite. The best guards dominate in P&R. Scale 0-2.
Points per possession as roll man in pick & roll. How well they finish after screening.
1.3+ is elite. Key skill for modern bigs. Scale 0-2.
Points per possession on catch-and-shoot opportunities. Spot-up shooting efficiency.
1.1+ is elite. Must be respected as a shooter. Scale 0-2.
Points per possession on post-up plays. Back-to-basket scoring efficiency.
1.0+ is elite. Rare skill in modern NBA. Scale 0-2.
Points per possession on cuts to the basket. Off-ball movement efficiency.
1.3+ is elite (easy baskets). Good cutters are valuable role players. Scale 0-2.
Points per possession in transition. Fast break scoring efficiency.
1.2+ is elite. Transition is the most efficient offense. Scale 0-2.
How often this player touches the ball. Measures offensive involvement.
High = focal point of offense. Low = off-ball player. Scale 0-100.
Points generated per touch. How efficiently they use possessions.
Higher = more productive with the ball. Quality over quantity. Scale 0-0.5.
How often this player attacks the basket off the dribble.
High = constant paint threat. Low = perimeter-oriented. Scale 0-20.
Field goal percentage on drives to the basket.
50%+ is efficient. Shows ability to finish through contact. Scale 0-100%.
Three-point percentage on catch-and-shoot attempts. Pure shooting ability.
40%+ is elite. Shows value as a floor spacer. Scale 0-100%.
Three-point percentage on pull-up jumpers. Shot creation off the dribble.
37%+ is elite. Rare and valuable skill. Scale 0-100%.
Passes deflected per game. Active hands and defensive awareness.
3+ is elite. Shows defensive engagement. Scale 0-10.
Points generated from screen assists. Value as a screener.
High = quality screen setter. Important role player skill. Scale 0-100.
Box outs per 100 possessions. Effort and positioning for rebounds.
High = fights for position. Low = watches the ball. Scale 0-10.
Loose balls recovered per game. Hustle and effort metric.
High = dives for every ball. Shows motor. Scale 0-5.
Hustle stats combined: deflections, loose balls, charges, screen assists per 36 minutes.
High = does the little things. "Glue guy" indicator. Scale 0-15.
Total EPM impact weighted by playing time. Combines offensive and defensive EPM with games played and minutes.
+5 is elite. +2 is very good. 0 is average. Measures overall value considering playing time. Scale -10 to +15.
Offensive EPM weighted by usage rate. High usage + high offensive EPM = elite offensive player.
+10 is elite scorer/creator. +5 is very good. Shows offensive value adjusted for role. Scale -10 to +15.
Defensive EPM weighted by defensive activity (blocks + steals). Active defenders get more credit.
+3 is elite defender. +1 is above average. Rewards impactful defensive contributors. Scale -10 to +10.
Net possessions created per 75 possessions. Counts steals (weighted by type), offensive rebounds, foul drawing value, minus turnover cost (live ball turnovers cost more than dead ball).
+1.5 is elite possession creator. Negative = gives away more than creates. Best players hover around +2. Scale -5 to +5.
See also Possession
Weighted turnover cost. Live ball turnovers (bad passes, lost dribbles in open court) cost 1.25× more than dead ball turnovers because they lead to fastbreaks.
Lower is better. Under 2.0 is excellent ball security. Over 3.5 is problematic. Scale 0-5.
Net possession advantage. How many more possessions a player creates than they cost the team.
Positive = net positive. +2 means creates 2 more possessions than loses per 75. Scale -5 to +5.
True defensive stops per 100 possessions. Includes steals, blocks (weighted by whether team retained possession), charges drawn, and deflections (partial credit).
4+ is elite disruptor. Accounts for block recovery - a block that goes out of bounds is only ~40% as valuable as one your team recovers. Scale 0-10.
Percentage of steals from anticipation (intercepting passes) vs activity (stripping the ball handler). High = reading the game. Low = gambling/activity.
70%+ indicates elite anticipation. 40% or below is mostly strip steals which can lead to fouls. Scale 0-100%.
Points above expected per 75 possessions based on shot quality. Compares actual eFG% vs expected eFG% from shot context (difficulty, defense, location) and multiplies by volume.
+2 means adding 2 points per 75 above what an average shooter would score on those shots. Elite shooters: +3 or higher. Scale -5 to +5.
Shot making skill: actual eFG% minus expected eFG% based on shot quality. Positive means beating expectations.
+5 is elite shot maker. 0 means shooting as expected. -5 means significantly underperforming shot quality. Scale -10 to +10.
Average points created per assist based on destination. Rim assists = 1.24 pts (62% make rate × 2). 3PT assists = 1.08 pts. Midrange = 0.80 pts.
1.15+ means creating mostly high-value looks (lobs, 3s). Below 1.0 means mostly midrange assists. Scale 0.8-1.3.
Percentage of assists that go to rim attempts (highest value assists). Lob throwers and drive-and-kick players have high values.
40%+ is elite at creating rim looks. Shows ability to collapse defenses. Scale 0-100%.
Percentage of assists on 3-point attempts. High for playmakers who create from the perimeter.
50%+ means primarily creating 3PT looks. Shows spacing creation ability. Scale 0-100%.
Expected points from foul drawing per game. 2PT fouls = 1.70 pts (2 FTs × 85%). 3PT fouls = 2.55 pts. And-1s = 0.85 pts (1 FT).
4+ is elite at getting to the line. Shows ability to create contact and finish through it. Scale 0-10.
And-1 trips per 100 rim attempts. Shows ability to finish through contact at the rim.
10+ is elite at finishing through fouls. Giannis/Embiid territory. Scale 0-20.
Total value from drives: direct scoring points + kickout creation value (passes × conversion rate × point value).
8+ is elite driver who scores AND creates. Combines finishing with playmaking. Scale 0-15.
Gravity created from drives. Combines drive volume with pass rate to measure how much attention defenses must pay.
60+ means defense collapses consistently. Creates open looks for teammates. Scale 0-100.
Performance on late shot clock situations (<4 seconds). Volume × efficiency on these difficult possessions.
20+ shows clutch shot-making ability when play breaks down. Who you want with the ball at end of shot clock. Scale 0-50.
Putback scoring value: direct putback points + self-offensive rebound value (getting your own misses).
3+ is elite second-chance scorer. Shows motor and positioning. Scale 0-10.
Assists on 3-point shots.
Shows ability to create open 3PT looks. Scale 0-100%.
Field goal percentage on shots where the closest defender is 4+ feet away. Measures pure shooting ability without defensive pressure.
50%+ is elite. Below 40% suggests shooting mechanics issues even when open. Scale 0-100%.
Field goal percentage on shots with a defender within 2-4 feet. Shows ability to convert contested shots.
45%+ is elite shot-making. Below 35% = struggles vs tight defense. Scale 0-100%.
Percentage of shots taken with the defender 6+ feet away. Indicates ability to get clean looks.
Higher = better at finding open shots or teammates create space for them. Scale 0-100%.
Percentage of shots taken with a defender within 2-4 feet. Shows shot selection difficulty.
High = takes tough shots (good if efficient, bad if not). Low = patient shot selection. Scale 0-100%.
Three-point percentage on open shots (defender 4+ feet away). True measure of 3-point shooting ability.
40%+ is elite. Below 33% = unreliable even when open. Scale 0-100%.
Three-point percentage on contested shots (defender 2-4 feet). Tests shot-making under pressure.
35%+ is very good contested shooting. Below 28% = struggles vs closeouts. Scale 0-100%.
Percentage of 3-point attempts from 28+ feet (beyond normal NBA 3-point range).
High = logo-range threat like Curry/Dame. Creates extra floor spacing. Scale 0-100%.
Three-point percentage on attempts from 28+ feet. True deep-range shooting ability.
35%+ is elite deep range. Only a few players can sustain this. Scale 0-100%.
Composite score measuring average defender distance on shot attempts. Higher = consistently gets open looks.
High = great at finding space or teammates create for them. Low = takes contested shots. Scale 0-100.
Average difficulty of shots attempted based on defender distance, shot type, and location.
Higher = takes harder shots (valuable if efficient). Lower = easier shot diet. Scale 0-100.
Efficiency on contested shots relative to expected. Positive = beats expectations on tough shots.
+5% is elite shot-maker. Negative = should avoid contested shots. Scale -20 to +20%.
Three-point percentage from corner spots (left and right). The most efficient 3-point shot location.
42%+ is elite corner shooting. Below 35% = avoid corner spots. Scale 0-100%.
Percentage of 3-point attempts from corners. Shows shot diet and offensive scheme fit.
High = 3&D role, stationed in corners. Low = more above-break or off-dribble 3s. Scale 0-100%.
Three-point percentage from above the break (top of key, wings). Longer distance than corners.
38%+ is strong above-break shooting. Below 33% = below average. Scale 0-100%.
Percentage of 3-point attempts from above the break. Most common 3-point location.
High = takes 3s from anywhere. Low = corner specialist or non-shooter. Scale 0-100%.
Field goal percentage on shots in the painted area (excluding rim). Includes floaters and short shots.
50%+ is excellent. Below 40% = struggles in the paint. Scale 0-100%.
Percentage of shots taken from the paint (non-rim). Shows shot diet in the mid-paint area.
High = floater game or post moves. Low = rim or perimeter focused. Scale 0-100%.
Field goal percentage on mid-range shots (outside paint, inside 3-point line).
45%+ is strong midrange game (rare skill). Below 38% = avoid midrange. Scale 0-100%.
Percentage of shots from mid-range. Traditional but less efficient shot zone.
High = old-school game. Low = modern rim-or-3 shot diet. Scale 0-100%.
How evenly distributed shots are across zones. Higher = versatile shooter from anywhere.
High = multi-level scorer. Low = one-dimensional (rim only or perimeter only). Scale 0-100.
Corner 3P% minus Above-Break 3P%. How much better from corners vs above the break.
Positive = corner specialist. Negative = prefers above-break or equal from both. Scale -20 to +20%.
Field goal percentage when guarded by guards (PG/SG). Shows post-up or size advantage.
High = exploits smaller defenders. Low = struggles vs quick guards. Scale 0-100%.
Field goal percentage when guarded by wings (SF). Most common matchup type.
This is baseline efficiency. Compare to vs Guards and vs Bigs. Scale 0-100%.
Field goal percentage when guarded by bigs (PF/C). Shows ability to handle length.
High = can score over size. Low = bothered by length/rim protection. Scale 0-100%.
Field goal percentage when guarded by top-25 defenders. Tests against the best.
High = performs vs elite D. Low = gets clamped by top defenders. Scale 0-100%.
FG% drop when guarded by elite defenders vs average defenders. Lower = more consistent.
Below -5% = significantly worse vs elite D. Near 0% = immune to defensive quality. Scale -30 to +10%.
How consistent FG% is regardless of defender type. Higher = adaptable scorer.
High = scores on anyone. Low = matchup-dependent (exploits certain types). Scale 0-100.
Opponent FG% when defending guards. Lower = better perimeter defense.
Below 42% is strong. Above 48% = struggles defending guards. Scale 0-100%.
Opponent FG% when defending wings. Most common defensive assignment.
Below 44% is solid wing defense. Compare to Guards and Bigs. Scale 0-100%.
Opponent FG% when defending bigs. Shows interior/post defense ability.
Below 50% is strong. Above 55% = gets backed down by bigs. Scale 0-100%.
How well a player defends all position types. Higher = can switch 1-5.
High = modern switchable defender. Low = positional specialist. Scale 0-100.
Opponent FG% when defending top-50 offensive players. Tests against elite scorers.
Below 45% = can guard stars. Above 50% = struggles vs elite offense. Scale 0-100%.
How much elite scorers struggle vs this defender compared to their average.
Negative = holds elite below avg (true stopper). Positive = elite exploits them. Scale -15 to +10%.
The defender type this player scores most efficiently against.
Identifies exploitable matchups and posting-up opportunities. Scale 0-100.
The defender type this player struggles most against.
Shows defensive types to avoid or scheme around. Scale 0-100.
Defensive attention this player draws, on a scale where zero is league average.
Positive = draws extra defenders (creates space for teammates). Negative = ignored by defense. Scale -100 to +100.
THE CURRY SPECIAL - Gravity drawn without the ball on the perimeter. Elite movement shooters dominate this.
High = warps defenses just by existing on court. Curry/Klay territory. Scale -100 to +100.
Ball-handler gravity - defensive attention when you have the ball on the perimeter.
High = creates driving lanes and open teammates. Luka/Trae/SGA territory. Scale -100 to +100.
Lob threat gravity - attention drawn without the ball in the paint. Vertical spacers and cutters.
High = lob threat that commands help defense. DeAndre Jordan/Clint Capela type. Scale -100 to +100.
Post presence gravity - attention when you have the ball in the paint.
High = post-up threat that demands doubles. Jokic/Embiid territory. Scale -100 to +100.
Total video frames tracked for gravity calculations. Sample size indicator.
More frames = more reliable gravity data. Low frames = limited sample. Scale 0-500K.
Contested rebounding ability score combining contested OREBs and DREBs weighted by difficulty.
High = wins battles on the glass. Low = relies on uncontested boards. Scale 0-100.
Rebound positioning intelligence - how well a player converts rebound chances vs expected.
High = great positioning and anticipation. Low = poor box-out or positioning. Scale 0-100.
Offensive rebounding aggression score. Contested OREBs weighted by effort and difficulty.
High = motor and hustle for second chances. Steven Adams/Drummond territory. Scale 0-100.
Total rebounding effort combining contested rate with rebound distance covered.
High = crashes glass hard from distance. Low = waits for easy boards. Scale 0-100.
One sentence, then the catch
Every entry opens with a definition that stands on its own, then says what the metric cannot tell you. The second line is usually the one that changes a decision.
A rate is not a total
Most metrics here are rates — per 100 possessions, per 36 minutes, per game. A rate says what a player does when he plays; pair it with minutes and games before you act on it.
Thin numbers are labelled
A metric computed on too little play prints with its sample in plain words rather than quietly. See thin sample and withheld.
Price and projection differ
Master Value is what a player costs; projection is what he is expected to do. Trades live in the distance between them.
Open
Dynasty Rankings
Master Value, market rank, tiers and the hype gap, one row per player.
Open
Trade Calculator
The fairness band and the value waterfall, with every step of the verdict shown.
Open
Analytics Suite
Four tools every day — lineups and on/off, teammate splits, the stat bench and the dossier — plus two more that change each morning.