Accuracy and Fact Checking

Maintaining Accuracy

Painstaking efforts are made to ensure every statistic and bit of information shown on this site is accurate. For example, the database has multiple constraints that cannot be broken under any circumstances. These include:

  • A team's win/loss record must equal the sum of the results of their game log
  • A team's final score for a given game must be the sum of all the goals/points/runs scored in the box score
  • Every streak and record is recalculated anytime the game database is changed in any way
There are cases in all 4 leagues where other sites, and even the leagues themselves, violate the first or second constraints because the results or scoring details of given games are not known precisely. In cases where the scoring details are not known for sure, we do not display them. We do not display any partial scores due to the above constraints.

There are certain cases where the specific outcomes of a game, mainly really old MLB games, are disagreed upon based on different sources. In those cases we've attempted to pick the source which we trust the most, which is generally Retrosheet.

Because of the third constraint, that every streak and record is recalculated for every new game, the website is always guaranteed to be in a self-consistent state. That means for every streak or record, you should be able to go to the relevant team or league pages to figure out how that number was calculated. It is simply not possible for a streak or record to include a game that hasn't been imported yet, and this also means that we do not specifically edit or alter any streaks or records individually; every streak and record is calculated from the game log, and will always be as up to date as the gamelog is.


In the course of importing hundreds of seasons and tens of thousands of games, some fun little historical quirks have come up that challenge assumptions that you might have had. Such as:

  • A division winner will always have the best record in the division
    • Nope, in the 1981 MLB season, due to a mid-season player's strike, a team could win their division without the best record in the division over the whole season.
  • The winner of a playoff series is the team that wins the most games
    • Nope, in the early NHL seasons, many playoff series were decided by aggregate goals. In fact, for many seasons the aggregate goal and best-of formats were played in the same playoffs. For example, the 1928 New York Rangers lost the final game of the Conference Semis, but still won the series on aggregate goals, then went on to win the Conference Finals in aggregate goals, and then win the Stanley Cup in a standard best-of-5.
  • The team that scored the most runs in a baseball game is the winner
    • Nope, there are countless examples in MLB of games in which the team with more runs still lost the game, due to forfeit. The most recent example being the 5-7 win by the Yankees over the Washington Senators (now Texas Rangers) on Sep 30, 1971. You can read more about the events that led to the forfeit.
  • A team with the same name is always part of the same franchise
    • Nope, not even close. Take the Washington Senators, who existed 3 separate times, from 1892-1899 in a franchise that folded, from 1901-1960 in a franchise that is now the Minnesota Twins, and from 1961-1971 in a franchise that is now the Texas Rangers.

Accuracy of pre-1996 NBA comebacks

Because NBA play-by-play data only goes back to the 1996-97 season, many historically large NBA comebacks would be underestimated just based on the box scores. Fortunately, many historically large comebacks had reputable and contemporaneous newspaper coverage, and in the cases where we could find a recap of the game, and that recap included an exact value of the largest deficit and the score and quarter when that deficit occurred, we updated our records to include that information. We've done this for tens of games, including all the top-10 all-time comebacks, and our database is probably the most accurate database for cataloging NBA comebacks. But still the database isn't perfect. If you see an approximated comeback on our site that you would like updated, please send us the link to the newspaper article for the particular game and we will verify the information and update the database.

Find a mistake?

Though we try very hard to make everything perfectly accurate, there are occasional issues that pop up. For instance, a reader once wrote in that a historical winning streak for an MLB team was 1 higher than he thought it should have been. After investigating the discrepancy, I discovered that a double-header from a rescheduled game had been imported in the wrong order, as if the night game has happened before the day game. After verifying the correct order of the games from newspaper reporting, the database was updated and when the winning streak was recalculated, it was the expected and correct value.

If you notice anything that looks like a mistake, please don't hesitate to email us at There are certain records and statistics where our site will have slightly different information than Wikipedia or other sports sites, and more often than not when that happens, our site is right, or at least our site is right based on the assumptions that we make. However if you notice something that looks suspicious, feel free to email.

Notable Fixes

  • The 1954 Knicks were erroneously listed as missing the playoffs, which incorrectly listed them as the best NBA team to miss the playoffs. They did in fact make the playoffs, but lost in the initial Round Robin playoff round and did not advance to a more traditional playoff series. This was the only year the NBA used this playoff format.