The simple idea
If Sports Rounders publishes a team at +120 and the same market closes +105, the earlier ticket captured a better return. That is positive closing-line value. If the market closes +135, the published +120 missed a better later price.
How we calculate it
For moneylines and price comparisons, we convert both sides of the same sportsbook market into implied probabilities, remove the bookmaker margin, and compare the selected side’s locked fair probability with its closing fair probability.
The report also shows the relative percentage change because it is easy to compare across prices.
Spreads and totals need two measurements
A move from +3.5 to +3 is not just an odds change—the half-point itself has value. We report line movement separately from price movement and do not collapse them into a misleading single number without a validated conversion.
What CLV cannot prove
One positive close does not prove a model is profitable. CLV becomes useful across a forward, consistently captured sample. It also does not turn a losing ticket into a win; it evaluates the price decision independently from that game’s result.
How the free tracker scores a bet
The tracker’s primary measure is EV against the no-vig closing pair at the same sportsbook, market, period, and line. It uses the site’s power de-vig method (power-v1). For fair close probability p and the user’s decimal odds d, EV per dollar is p × (d − 1) − (1 − p). It also reports whether the user’s price beat the same-book close, as a separate secondary measure.
If the exact line is no longer posted, the tracker marks “line moved,” shows the available closing line, and assigns no EV rather than guessing a conversion. The close is the latest valid, independently fetched two-sided snapshot recorded before scheduled start; it is not a per-user odds request. Integer spread and total lines can push, so two-way no-vig EV does not fully model push probability and may be slightly optimistic.
Player props, parlays, live bets, futures, unsupported books, and late entries are not included in headline statistics. The tracker measures price quality, not game outcomes. Every average is shown with its sample size and a bootstrap confidence interval; small samples should not be treated as evidence of a reliable edge.