Esports Math Glossary

A zero-sum skill rating algorithm calculating comparative win probabilities based on point differentials. Used as the historical foundation for CS2 and Dota 2 team power rankings.

An advanced Bayesian rating framework created by Mark Glickman that expands classical Elo by modeling uncertainty through Rating Deviation (RD) and skill volatility (sigma).

A measure of confidence in a player or team's estimated rating, equivalent to one standard deviation in a Gaussian distribution. RD inflates over inactivity intervals and collapses after matches.

The sensitivity multiplier determining how many rating points are transferred per match outcome. High K-factors (32+) provide rapid reactivity but excessive noise; low K-factors (16) provide long-term stability.

Quantitative index measuring the breadth and statistical viability of a team across the 7 Active Duty maps, weighted by sample volume and winrate thresholds.

The map a team systematically vetos in the first ban phase across all official fixtures, eliminating exposure to low practice hours and forcing opponents into contested ground.

The algorithmic game-theoretic sequence where each team makes ban and pick decisions that maximize their worst-case series payoff against a rational opponent.

The binomial path summation calculating the overall probability of winning a Bo3 or Bo5 series from individual map win probabilities.

The structural multiplier where winning a pistol round triggers high-probability conversion of the subsequent anti-eco round, producing an immediate 2-0 score momentum.

A situation where a team loses a round immediately after winning one, capping their loss bonus at $1,400 and forcing an eco round, creating sustained multi-round EV swings.

A high-risk tactical investment where a team spends all remaining funds on sub-optimal rifles and pistols instead of saving, yielding ~28-32% upset conversion.

A designated concession round where a team spends minimal funds (<$500) to build a maximum loss bonus bankroll for the next round.

The structural map design skew that causes round win probabilities to deviate from 50% between Counter-Terrorists and Terrorists (e.g. 54.8% CT on Ancient vs 52.6% T on Anubis).

The mandatory role reversal at round 13 in CS2 MR12 regulation where economies reset to $800 pistol baseline and map asymmetry inverts.

The conditional probability of overcoming a substantial half-time deficit (e.g. 4-8 or 3-9) based on map side asymmetry, pistol conversion, and rating skill delta.

The mean health points subtracted from opponents per round played. Unlike KPR, ADR captures assist contribution and chip damage, correlating with true win share at r=0.74.

The percentage of rounds in which a player achieved at least one Kill, Assist, Survived, or was Traded within 4 seconds of dying. Measures baseline consistency.

The industry-standard composite index evaluating competitive CS2 performance by integrating KAST%, KPR, DPR, ADR, and Impact Rating into a centered distribution around 1.00.

A Gaussian z-score standardizing individual performance against specific tactical roles (Anchor, Entry, AWP, IGL, Lurker), removing systematic role bias.

A metric quantifying the degree of statistical reliance a team has on a single star player. High CPI (>1.35) highlights teams vulnerable to solo target-banning and bad individual form.

The probability-weighted average payout of a wager over an infinite horizon. In esports betting, positive EV (+EV) represents a mathematical edge where model probability exceeds bookmaker implied probability.

The cumulative margin extracted by a sportsbook by pricing reciprocal outcomes below true fair odds. Low overround (sub-4%) preserves player capital, while casual books extract 8-12%.

The percentage of total wagered turnover mathematically returned to bettors across balanced market distributions. 1win Esports markets benchmark at ~96.8% RTP versus ~92.5% across industry averages.

Statistical dispersion of betting returns around mathematical expectation and the maximum observed peak-to-trough bankroll contraction. Esports live markets exhibit elevated variance due to sudden economy eco resets.

The mathematically optimal fraction of bankroll to wager on a positive-expectation proposition to maximize logarithmic capital growth. Half-Kelly (f*/2) and Quarter-Kelly (f*/4) are recommended in esports to hedge against hidden map pool variance.