About

QuantusSports is a sports statistics analytics platform providing unique metrics to the NBA, NHL, and MLB. Currently, it is comprised of three main aspects, Ratings, PowerRankings, and Predictions.

Ratings

QuantusSports has two main rating metrics: QuantusRatings and QuantusIndex

QuantusRatings is a modified Elo formula. Elo is a system created by Arpad Elo for ranking chess players. Every player starts with a rating of 1000. As teams play each other, their ratings fluctuate, gaining points from a win, and losing points from a loss. The point swing is determined by the respective ratings of both competitors. If a player with a 600 rating beats a player with a 1200 rating, the point swing will be much greater than say two players with the same rating. The system is used in modern chess, tennis, and competitive video games. While the normal Elo formula treats a games results as a binary decision, win or loss, the QuantusRatings formula takes into account one more piece of information, the point differential. If a baseball team is blown out 9-0, it will have a bigger impact on their rating than if they lose a close game.

Learn more about Elo

QuantusIndex is an application of Google's PageRank algorithm. PageRank is used to determine which web pages are the most important, by taking into account the number of other important pages that link to it. QuantusIndex replaces web pages with sports teams. When a team beats another team, it not only rewards the team that won, but every other team in the network based on their past encounters. Similar to baseball's OPS+, the league average QuantusIndex rating is set to 100, with teams above 100 being above average, and teams under 100 being below average. A team with an Index of 125 is 25% better than average.

Learn more about PageRank

PowerRankings

PowerRankings are an application of QuantusRatings that attempt to determine who the best team is within a 7, 14, or 30 day period. It uses the same formula as QuantusRatings but is much more volatile. Each game is worth more because of the short timeframe. You will often see teams with a rating over 1000, or below 0 in PowerRankings.

Predictions

Predictions are another application of QuantusRatings that attempt (with varying success) to predict the winner of a matchup. It does this using the standard Elo expected score formula. The parameters are adjusted every so often to make it as accurate as possible. Despite this, in sports like baseball and hockey the accuracy is typically not much better than a coin flip. The model does better in sports with higher skill gaps and less variability like basketball. These predictions are not intended to be used to make financial decisions.

BlueSky

QuantusSports is active on BlueSky through various automated accounts. After each game, these accounts post updates on how QuantusRatings were affected. They also post the current days predictions, and the top 10 teams every morning.

@QuantusBaseball.bsky.social
@QuantusBasketball.bsky.social
@QuantusHockey.bsky.social

Made by jmi06