DFS Glossary

44 essential terms every DFS player should know. From bankroll management to game theory optimal play.

Contest Types & Structure

GPP (Guaranteed Prize Pool)

A tournament-style contest where the prize pool is guaranteed regardless of how many entries fill. GPPs pay out to the top ~20% of entrants with a top-heavy payout structure. Also called tournaments or tourneys.

Cash Game

A contest format where roughly the top 50% of entrants win a fixed payout (typically ~1.8x-2x the entry fee). Includes Head-to-Head, 50/50, and Double Up formats. Lower variance than GPPs.

Head-to-Head (H2H)

A cash game where you compete against exactly one other player. The winner takes the pot minus the site's rake.

50/50 (Fifty-Fifty)

A cash game where the top half of the field wins a fixed payout (roughly double the entry fee minus rake). Also called Double Ups.

Single Entry

A contest that limits each user to one lineup. Reduces the advantage of high-volume players and mass multi-entry (MME) strategies.

MME (Mass Multi-Entry)

A strategy where a player enters many unique lineups into the same contest to increase their chances of having a top-scoring lineup. Common in large-field GPPs with high max-entry limits.

Satellite

A contest where the prize is an entry ticket to a larger, more expensive tournament rather than cash.

Ownership & Field Strategy

Ownership Percentage

The percentage of lineups in a contest that roster a specific player. High ownership means a player is popular (chalk); low ownership means they are contrarian.

Chalk

A player who is highly owned in a contest. Chalk plays are considered "safe" but offer less differentiation from the field. In GPPs, too much chalk limits your ceiling.

Leverage

A strategy of rostering players at a lower rate than the field to gain a scoring advantage when those players perform well. Leverage plays differentiate your lineup from opponents.

Pivot

Switching from a popular (chalk) player to a similar but less-owned alternative. Pivots provide leverage while maintaining a similar projection ceiling.

Fade

Intentionally not rostering a popular player. Fading chalk is a core GPP strategy when you believe a player is over-owned relative to their projection.

Game Theory Optimal (GTO)

A lineup construction approach that balances projected points with ownership to maximize expected value. GTO strategies account for what the field is doing, not just raw projections.

Field

All the other players entered in your contest. Understanding the field's tendencies (chalk vs. contrarian) is crucial for GPP strategy.

Projections & Analytics

Projection

An estimated fantasy point total for a player on a given slate. Projections are derived from statistical models that factor in matchups, pace, usage, and other variables.

Floor

The low end of a player's expected fantasy point range. A high floor means the player is unlikely to score below a certain threshold. Cash game lineups prioritize high floors.

Ceiling

The high end of a player's expected fantasy point range. A high ceiling means the player has blowup potential. GPP lineups prioritize high ceilings.

Boom/Bust

A player profile characterized by high variance: they either significantly exceed their projection (boom) or fall well short (bust). Common among deep threats and low-usage players.

Value Play

A player priced below their expected output, offering more fantasy points per dollar of salary. Value plays free up salary cap to pay for studs elsewhere in the lineup.

Stud

A premium-priced player ($8K+ in DraftKings NBA) who is expected to have a high fantasy point output. Studs anchor lineups and are often high-ownership chalk.

$/Point (Dollars Per Point)

A value metric calculated by dividing a player's salary by their projected fantasy points. Lower $/point indicates better value. Also called price-per-point.

Implied Total

The expected number of points a team will score in a game, derived from Vegas betting lines. Higher implied totals signal better fantasy scoring environments.

Pace

The speed at which a team plays, measured in possessions per game. Higher pace creates more fantasy opportunities for all players on both teams.

Lineup Construction

Stacking

Rostering multiple players from the same team or game to create correlated outcomes. When a game goes high-scoring, stacked players benefit together. Essential in GPPs.

Correlation

The statistical relationship between player performances. Positively correlated players (same team, same game) tend to score well together. Stacking exploits positive correlation.

Bring-Back

Adding a player from the opposing team in a stack to capture scoring from both sides of a game. For example, stacking a QB with his WR and adding the opposing QB.

Game Stack

Rostering multiple players from both teams in a single game. Game stacks exploit high-total games where both offenses are expected to score.

Lineup Lock

The deadline after which you can no longer edit your lineup. Typically at the start of the first game on the slate. Late-breaking news (injuries, scratches) before lock can create edges.

Late Swap

The ability to swap players in your lineup after the initial lock, as long as those players' individual games haven't started yet. Available on DraftKings but not all sites.

Bankroll & EV

Bankroll

The total amount of money you have allocated for DFS play. Proper bankroll management (typically risking 5-15% per slate) prevents going broke during inevitable downswings.

EV (Expected Value)

The average return you can expect from a contest entry over the long run. Positive EV (+EV) means you expect to profit; negative EV (-EV) means you expect to lose. Calculated as (probability of winning x payout) - entry fee.

ROI (Return on Investment)

Your net profit divided by total entry fees, expressed as a percentage. An ROI of +10% means you earn $1.10 for every $1.00 wagered. Sustainable winning ROI in DFS is typically 5-15%.

Rake

The fee that the DFS site takes from each contest. Typically 10-15% of the entry fee. For example, a $10 entry might put $8.50 in the prize pool and $1.50 to the site. Lower-rake contests are inherently more +EV.

Overlay

When a guaranteed prize pool contest doesn't fill completely, the site still pays out the full guaranteed amount. The difference between the guaranteed pool and actual entries collected is the overlay, which makes the contest more +EV.

Variance

The natural randomness in DFS outcomes. Even with a perfect strategy, short-term results will fluctuate. Understanding variance prevents tilt and ensures you stick to +EV decisions.

Tilt

Emotional decision-making after a loss or bad beat, leading to poor strategic choices (chasing losses, over-entering, abandoning process). Bankroll management and process discipline combat tilt.

Player Archetypes

Shark

A highly skilled, profitable DFS player who consistently beats the field. Sharks tend to use advanced tools, build optimal lineups, and exploit recreational players.

Whale

A player who enters many contests at high stakes. Whales may or may not be profitable; the term refers to volume and stakes, not skill. Whale tracking is a key part of competitor intelligence.

Fish

A recreational or less-skilled DFS player. Fish tend to play popular players, ignore ownership, and make suboptimal lineup decisions. Identifying fish-heavy contests improves your EV.

Grinder

A dedicated DFS player who plays consistently, manages bankroll carefully, and treats DFS as a serious pursuit. Grinders focus on process over results and seek long-term profitability.

DFS Backtesting & Accuracy

Backtesting

The process of comparing projected fantasy points against actual results after games are played. Backtesting reveals projection accuracy, identifies systematic biases, and builds confidence in your data source.

MAE (Mean Absolute Error)

The average difference between projected and actual fantasy points across all players. Lower MAE means more accurate projections. A key metric for evaluating any projection system.

Calibration

How well a projection system's confidence intervals match reality. A well-calibrated system with 80% confidence intervals should contain the actual result 80% of the time.

Star Bias

A systematic tendency for projection models to over- or under-project high-salary players. Tracking star bias helps correct for predictable errors in premium player projections.