How We Review and Score Gambling Sites
Every score on our reviews comes from the same published rubric: six weighted categories, graded on a 0 to 5 scale against written anchors, using facts we can source and price data we collect ourselves. This page explains exactly how a GamblingSite.com rating is built, where the numbers come from, how often we re-check them, and what we refuse to do.
The Six Categories and Their Weights
A sportsbook’s overall score is the weighted average of six category scores. Odds value carries the largest weight because price is the one thing every bettor pays on every wager, and it is the category where we hold measured evidence rather than impressions. The weights below apply to sportsbooks; casino, poker, racing, and exchange reviews swap the first two categories for equivalents that fit the product (game library and live dealer for casinos, games and traffic for poker, pricing and liquidity for exchanges) while the last four stay identical.
Measured hold versus the regulated books we track, how often the brand posts the best available price, and the pricing posture on props and same-game parlays.
Sports breadth, live betting depth, props and parlay tooling, streaming, cash-out, and the features that actually change how you bet.
Deposit and withdrawal method breadth, the operator’s stated processing windows, fees, and minimums, all quoted with a verification date.
Attributed app-store ratings, navigation and bet-slip quality, stability history, and desktop parity, assessed as product research.
Licensing footprint, regulator history, the depth of responsible gambling tools, and account security posture.
What an offer is realistically worth after the playthrough math, plus ongoing promotions and loyalty value. Headline dollar amounts never move this score by themselves.
What the Numbers Mean
Scores are graded to one decimal against fixed anchors, so a 4.2 means the same thing in every review. We publish the anchors so you can hold us to them:
- 5.0: verifiable best-in-market evidence
- 4.0: clearly above industry standard
- 3.0: industry standard
- 2.0: below standard with concrete gaps
- 1.0: disqualifying problems
Where Our Numbers Come From
Three data sources sit under every review, and none of them is a vibe. Each review names its sources inline, with sample sizes and dates, so you never have to take a rating on faith.
Our Own Odds Archive
Since June 2026 we have captured sportsbook prices roughly every six hours into a permanent, append-only archive, using odds data from The Odds API. For each review we compute the brand’s hold (the margin built into its prices) from pregame snapshots taken in the final 24 hours before each game, and we count how often the brand posted the best available price among the US-regulated books in our panel. The panel members are named in every data chart we publish, and the math is deliberately conservative: hold is calculated in probability space, and best-price comparisons only happen between books quoting the same line. Lower hold means better prices for you. When a brand is not yet in our panel, its review says so plainly instead of borrowing someone else’s numbers.
A Public, Graded Picks Ledger
We grade our own betting picks in public, win or lose, on our track record page, including closing line value on the lines we called. That discipline matters here for one reason: the same standard applies to reviews. Claims get sources, results get logged, and misses stay visible.
Verified Facts With Dates Attached
Every volatile fact in a review (the welcome offer and its terms, live states, banking methods and stated payout windows, app ratings, support channels, licensing) lives in a per-brand fact ledger. Each fact carries its source, the date we last verified it, and a confidence tier based on whether the source was the operator itself, an app store or regulator, or a third party. The “Facts verified” date at the top of each review is that ledger’s date, and it only moves when a full verification sweep actually ran.
What We Do Not Do
Trust is easier to keep than to win back, so we draw hard lines and put them in writing:
- No invented testing claims. We do not claim deposit-and-withdrawal timing tests, secret shopper panels, or “500 hands played” sessions we did not run. Our hands-on evidence is our own price data and documented product research.
- No fake reviewer armies. One named analyst scores each review and our Editor-in-Chief approves it. We do not publish aggregate star ratings from raters who do not exist.
- No pay-for-placement. Scores and rankings cannot be bought, and advertising relationships never touch the rubric. Some links on this site may earn us a commission, which is disclosed on our affiliate disclosure page and has no input into any score.
- No “guaranteed winners” language. Anywhere. Betting involves risk, and any probability we quote is the market’s implied number, attributed, never our secret model.
- No silent edits. Material changes to a review’s facts or scores are logged in that review’s change history.
How Fresh Is a Review?
Reviews decay because products change, so freshness is a process here, not a promise. The two most volatile facts (the welcome offer headline and the live-state count) are checked for every brand weekly. Deeper sweeps rotate through the full fact set so that every brand completes a full verification cycle at least monthly, and detected changes are re-checked through a second source before we act on them. Our odds data refreshes monthly with a new archive month. When something changes, the review’s change table shows what moved and when, and the “Facts verified” badge only carries the date of the last completed full sweep.
Re-Scores, Corrections, and Rubric Versions
Category scores are re-assessed quarterly, or immediately when something material happens (a market exit, a licensing action, a product overhaul). Every re-score is logged with the old score, the new score, and the reason. When the rubric itself changes, the version number changes with it and the change is noted on this page, so a score from one era is never silently compared against another. Factual errors are corrected under the standards in our editorial guidelines, with the correction noted in the review’s change history.
Who Writes and Approves These Reviews
Each review is scored by a named analyst and carries their byline, with Matthew Buchanan leading sportsbook and platform reviews. Every scorecard is then reviewed and approved by Paul Wilson, our Editor-in-Chief, before it publishes, and his name appears on each review as the fact-checker of record. If you spot something we got wrong, tell us and we will check it against the sources, fix it if you are right, and log the correction. Whatever you decide after reading a review, please gamble responsibly: our responsible gambling resources are always one click away.
Frequently Asked Questions
A few questions readers ask about how our scoring works.
Why should I trust your sportsbook scores over other review sites?
You do not have to trust us, and that is the point. The rubric, the weights, and the scoring anchors are published on this page, every review shows the price data and dated sources behind its numbers, and our own betting picks are graded in public on our track record page. You can check the work.
Do sportsbooks pay you for better reviews or higher rankings?
No. Scores cannot be bought and advertising never touches the rubric. Some links on the site may earn us a commission, which is disclosed on our affiliate disclosure page, but commissions have no input into any category score or ranking.
What does the facts verified date on a review actually mean?
It is the date of the last completed verification sweep across that review’s full fact set: bonus terms, live states, banking methods, stated payout windows, app ratings, support channels, and licensing. It only moves when a sweep actually ran, never as a cosmetic refresh.
How do you measure which sportsbook has the best odds?
We capture prices from US-regulated sportsbooks roughly every six hours into our own archive, then compute each book’s hold in probability space from pregame snapshots and count how often it posted the best available price on the same line. Every chart states its sample size and date range, and lower hold means better prices for the bettor.
How often do review scores actually change?
Volatile facts are checked weekly and every brand completes a full fact sweep at least monthly, while category scores are re-assessed quarterly or immediately after something material like a market exit or licensing action. Every re-score is logged in the review’s change history with a reason.