Advanced Stats to Know for NBA Betting

Jimmy Butler #22 of the Miami Heat. Tim Nwachukwu/Getty Images/AFP

The NBA has reached the conference finals, and that means that the top four teams in the league will be battling it out to see who is going to play for the trophy in the Finals.  

As the number of games remaining in the season dwindles and there are fewer games available to bet in a given week, it’s important to make sure you have all the information you need to make the right call in the contests that remain. Here’s a look at some of the advanced statistics that may help you.   

Going Back to School 

The NBA is a different game from college basketball, but gambling on the sport can be aided by many of the same tools that NCAA basketball bettors utilize. Just like in college, it’s important to take any counting statistics with a grain of salt, due to the vastly different paces at which teams play. There are a number of stats that address the pace issue by stating things on a per-possession (or per 100 possessions) basis.  

NBA teams also generate efficiency ratings on offense and defense that are pace independent—essentially measuring the points per possession a team scores and allows.  

The four factors (shooting, turnovers, rebounding, and free throw rate) are also just as important in the pros as they are on campus, and many of the same measures are used in the NBA to evaluate teams and players. These include rebounding rate, free throw rate (how often a player or team gets to the line as a function of shot attempts, as opposed to just what percentage of free throws they make), and effective shooting percentage (which looks at the different values for two and three-pointers).  

Get a Slice of the PIE 

There are a number of different “all in one” measures that attempt to determine who is the best all-around player. One of the newest is the Player Impact Estimate or PIE. It was developed by the NBA and combines all of a player’s box score stats to determine a player’s contribution to the events in a given game. It is similar to PER (Player Efficiency Rating) and attempts to improve on that older measure by putting more weight on defensive statistics.  

It looks at a player’s points, shots made and attempted, free throws made and attempted, offensive and defensive rebounds, assists, steals, blocks, turnovers, and fouls, as well as the totals in those statistics for the entire game.  

The benefit of PIE is that it is easily calculable by hand and uses readily available information. The downside is that it doesn’t take advantage of all the tracking technology that has helped to revolutionize basketball analytics. As a result, there are more powerful tools you can use to evaluate players for prop bets.   

Plus Minus 

There are a number of statistics that either calculate or estimate a player’s plus-minus rating. In other words, how much better does a player’s team do with him on the floor than without? It’s similar to hockey’s statistic that measures goal differential with a player on and off the ice, except there’s much more scoring in basketball. There are also a number of adjustments that can be made to try to get to a “true” plus-minus that eliminates any sources of noise in the data.  

While player plus-minus is interesting at times, NBA teams are more likely to use lineup plus-minus to see the impact of various groupings of players on the court together—something to be aware of when participating in live in-game wagering.   

Acronym Soup 

As we move up the level of sophistication in NBA data models, we find the acronyms get more involved. It seems that stat folks are almost as good at coming up with cute abbreviations as they are at crunching through piles of data.  

The folks at FiveThirtyEight have come up with RAPTOR (Robust Algorithm Player Tracking On/Off Ratings) to evaluate players and teams. It uses box score data, play-by-play data, and player tracking information. That means it can make use of data as specific as “distance traveled on the perimeter by defender.” 

Basketball Index has a competitor to RAPTOR in LEBRON (Luck-adjusted Estimate using Box Regularized ON-OFF)

Caveat 

When building a team (real life or fantasy) or settling a bar argument about who was better, you may need an all-in-one stat to help make a case. When wagering on an NBA game, you need to evaluate the entire team, not just its star, and when making a prop bet, you may not need to know who is the all-around best, just whether a given player will hit a given benchmark.