At ATS STATS, transparency isn’t a marketing slogan — it’s the entire foundation of what we do. This page explains exactly how our data is sourced, how our records are calculated, and how we handle the inevitable edge cases. If you’re going to use our stats to inform real betting decisions, you deserve to know exactly what’s behind the numbers.
Every game result, line, and ATS calculation on this site comes from our proprietary database, built and maintained internally since 2001. NFL data going back to 1983, and NBA, NHL, MLB, CFL, and College Football data going back to 1996.
Game results, final scores, and box-score data come from official league sources. Betting lines are recorded from consensus market data from various sportsbooks like DraftKings, FanDuel, and offshore sportsbooks and we standardize across books to give a clean, consistent dataset.
This is the question that separates serious handicapping data from sloppy data. There are three different lines that can be used to grade an ATS result:
ATS STATS grades all results against the opening line — the line as it first hit the market before sharp money moved it. We chose the opening line because it represents the price available to bettors who acted on early information, and it gives a consistent benchmark across our entire historical database back to 1983.
We do not retroactively adjust lines. The price at our recorded grading time is the price you see, even when the market moves dramatically afterward.
When a final result lands exactly on the spread (a “push”), we record it as a push — not a win, not a loss. Pushes are reported separately in every record so you can see the full picture: wins–losses–pushes.
When we display cover percentages, we calculate them as: wins / (wins + losses). Pushes are excluded from the percentage calculation because they don’t represent a decision either way. Some sites include pushes in their denominator, which artificially lowers the cover rate. We don’t do that — but we always show you the raw push count so you can do your own math if you prefer.
Our database updates within four hours of game completion and we confirm the data every morning at 6 AM ET as a backup. Final scores, ATS results, and all derived statistics (cover percentages, situational splits, trend records) are recalculated automatically as soon as a game is graded as final.
Every report, system, and analysis on this site is grounded in Ron Raymond’s 5 Fundamentals of Sports Betting:
These aren’t abstract concepts — they’re applied in the structure of our data itself. Our stat pages show situational splits (home/away, favorite/underdog, after a bye, on a short week, etc.) precisely because Performance Cycles and Player Availability shift those numbers in measurable ways. Our line tracking and value indicators exist to support the Value and Percentage Play pillars. And our records are reported with full pushes-and-totals transparency to support Money Management — because you can’t size a bet correctly without knowing the real cover rate.
Many of our pages report ATS records inside specific situations: home underdogs, road favorites, primetime games, division matchups, teams off a bye, etc. These splits are calculated from the same underlying game data using strict definitions:
When league realignment occurs (as it does in college football regularly), splits are calculated using the alignment that was in effect at the time each game was played, not retroactively reassigned.
We are human and our data sources are human. When we get a result wrong — and over thousands of games per season, it happens — we correct the record and note the correction. If a line gets misrecorded, a final score is adjusted by the league post-game, or a push gets miscounted, we fix it as soon as we’re aware of it.
If you spot something that looks wrong on any of our pages, contact us at ronraymond@atsstats.com. We will check it and respond.
We don’t:
The reason this site exists is because too much of the handicapping industry does all of those things. We don’t. The data shows what the data shows.
A 65% ATS record over 20 games is not the same thing as a 65% ATS record over 200 games. We always display the raw record alongside any percentage so you can judge sample size yourself. As a general rule, we treat anything under 10 games as informational only — not actionable as a standalone trend.
If you have specific questions about how a particular stat is calculated, or want to suggest data we should be tracking, contact us at ronraymond@atsstats.com. Methodology improvements are how we get better.