xG, no-vig, DNB, BTTS, fair odds.

xG (expected goals)

A shot-quality estimate: how many goals a team “should” score from the chances it creates, on average. CC Analytics uses team-level xG attack and defense priors (plus form and finishing adjustments) to project each side’s expected goals in a matchup.

xG is an input to a score distribution — not a final score prediction by itself.

Fair odds

The American price implied by the model’s probability for an outcome (moneyline, draw, total, BTTS, etc.), after calibration. Example: a 40% model probability ≈ +150 fair.

Fair odds are model estimates. They are not betting recommendations and they are not guaranteed value.

No-vig (devig)

Sportsbooks build juice (overround) into every market so implied probabilities sum to more than 100%. No-vig removes that overround so outcomes sum to 100%, giving a cleaner market baseline to compare against the model.

On the member board: model fair vs best available book price, with consensus no-vig used in edge checks.

DNB (draw no bet)

A two-way market on the home or away team where a draw refunds the stake. Model DNB prices come from the same 1X2 probabilities with the draw removed (renormalized home vs away).

BTTS (both teams to score)

Yes / No market on whether both teams score at least once. Derived from the joint score probability mass (all scorelines where home ≥ 1 and away ≥ 1), then calibrated like other binary markets.

Edge (as we publish it)

A gap between model probability and the market, expressed as expected value at the best available price — only after qualification filters (probability gap, book count, EV floors) and hard EV ceilings. Longshots are blended toward the market so published edges don’t look cartoonish.

Edge ≠ lock. Variance is real. Confirm lineups and live prices. The full Top Edges slate is members-only.