Glossary
What every stat on Bracket Hero means. The same definitions appear when you hover or tap the ⓘ beside a stat.
Efficiency and strength
Per-possession numbers take pace out of the picture, so a fast team and a slow team can be compared fairly.
- Offensive efficiencyAdjO
- Points scored per 100 possessions, adjusted for the strength of the defenses faced. Higher is better. Higher is better.
- Defensive efficiencyAdjD
- Points allowed per 100 possessions, adjusted for the strength of the offenses faced. Lower is better. Lower is better.
- Efficiency marginAdjEM
- Offensive efficiency minus defensive efficiency: how many points per 100 possessions a team is better than an average Division I team. The best single measure of team strength. Higher is better.
- PaceTempo
- Possessions per 40 minutes. A style, not a strength: fast and slow teams both win. A style: neither higher nor lower is better.
- Points per gamePPG
- Points scored per game. Fast-paced teams score more, so efficiency is a fairer comparison. Higher is better.
- Strength of scheduleSOS
- How good a team's opponents have been on average, in points per 100 possessions above an average team. Higher means a tougher schedule. Higher is better.
Shooting
How often a team's shots go in, and which shots it takes.
- Field goal %FG%
- Share of all shots, twos and threes, that go in. Higher is better.
- Three-point %3P%
- Share of three-point attempts that go in. Higher is better.
- Three-point rate3PA rate
- Share of a team's shots that are threes: how much it relies on the three. A style, not a strength. A style: neither higher nor lower is better.
- Free throw %FT%
- Share of free throws made. Matters most late in close games. Higher is better.
- Free throw attemptsFTA/g
- Free throws attempted per game: how often a team gets to the line. Higher is better.
Rebounding and ball control
Extra possessions won and possessions wasted.
- Rebounds per gameREB
- Total rebounds per game, offensive and defensive. Higher is better.
- Offensive rebound %ORB%
- Share of its own missed shots a team rebounds. More second chances. Higher is better.
- Defensive rebound %DRB%
- Share of opponents' missed shots a team rebounds, ending their possession. Higher is better.
- Assists per gameAST
- Baskets set up by a pass, per game. Higher is better.
- Turnovers per gameTOV
- Possessions lost to turnovers per game. Lower is better. Lower is better.
- Turnover rateTOV%
- Share of possessions that end in a turnover. Lower is better. Lower is better.
Defense
What opponents manage against the team.
- Points allowedOpp PPG
- Points opponents score per game. Lower is better; slow-paced teams allow fewer. Lower is better.
- Steals per gameSTL
- Steals per game: turnovers the defense takes by force. Higher is better.
- Blocks per gameBLK
- Shots blocked per game. Higher is better.
- Opponent FG %Opp FG%
- Share of opponents' shots that go in. Lower is better. Lower is better.
- Opponent 3P %Opp 3P%
- Share of opponents' three-point attempts that go in. Lower is better. Lower is better.
Roster
Who's on the floor, built from player minutes.
- Average heightHeight
- Average height of the players, weighted by minutes played. Higher is better.
- Program experienceExp
- Average years players have spent in the program, weighted by minutes played. Higher is better.
- Bench pointsBench
- Points per game from players who came off the bench. Higher is better.
Bracket and model terms
- Percentile
- Where a team ranks against the comparison group, 0 to 100. The 80th percentile is better than 80% of teams. On defense and turnovers, where lower raw numbers are better, 100th is still best.
- Power ranking
- Bracket Hero's order of all Division I teams by efficiency margin from our own model, refit every day from game results. Arrows mark teams that moved 3 or more places in a week.
- Projected field
- Our guess at the 68-team bracket if the season ended today, from the model's ratings and each conference's likely automatic bid. The real field is set by the selection committee on Selection Sunday.
- Seed
- A team's line in its region, 1 (strongest) to 16. Four teams share each seed line, one per region.
- First Four
- Four play-in games before the Round of 64: two between the last at-large teams and two between the lowest-rated automatic qualifiers.
- Model win probability
- How often the model expects a team to win a given game, from both teams' ratings and where the game is played. A 70% favorite still loses about three games in ten.
- Upset
- A win by the team with the worse seed. In the upset history, an underdog is a team at least three seed lines worse.
- Simulation
- Playing the whole tournament out thousands of times with the model's game odds, to see how often each team reaches each round.
- Credits
- What full simulator runs and premium tools cost. Every account starts with free credits; packs and the Bracket Pass add more. See Pricing.