Below is an explanation of the Teammate Correlation stats, what they show, and some examples from the 2019-20 NBA season.
What is Teammate Correlation?
The Teammmate Correlation tool looks at the relationship between two teammates’ performances (minutes per game, DraftKings points per game and FanDuel points per game), measured by correlation.
Correlation measures how two values move relative to each other, and is expressed as the correlation coefficient, a number on a scale from -1 (perfectly negative correlation) to 1 (perfectly positive correlation).
The closer the correlation coefficient is to 0, the weaker the relationship, and the further it is from 0, the stronger the relationship.
For a positive correlation, the values move in the same direction. If Value A increases, Value B increases. If Value A decreases, Value B decreases.
For a negative correlation, the values move in opposite directions. If Value A increases, Value B decreases. If Value A decreases, Value B increases.
Finally, there are also cases of no correlation. In these cases, a change in Value A doesn’t impact Value B.
A final point on correlation: we are not dealing with any perfectly correlated relationships. For example, in the positive correlations we will look at below, when Value A moves one direction, Value B will typically move in the same direction, but not always.
The closer the correlation coefficient is to 1 (or -1), the more often the relationship will be true.
The general idea is: how does a change in Player A’s performance impact our expectation for Player B’s performance? If we expect Player A to have a big game tonight (which we do if we’re playing them in a large field tournament), how does that impact our prediction for Player B?
Correlation in the NBA vs NFL
Any NFL DFS player should be familiar with the correlation between a QB and his pass catchers. Nearly every scoring play for a pass catcher will also be a scoring play for the QB, meaning you’re scoring points twice for a single event.
The QB-WR level of correlation doesn’t quite exist in basketball. Two players can both score on a single play when one is credited with an assist, but the total value of those assists is relatively low.
For example, Lebron James averaged 2.7 assists per game to Anthony Davis last season, by far his highest average with a single teammate – and that’s still just 4.05 Draftkings points per game.
So we don’t get the same type of obvious and strong positive correlations in the NBA, but we can still pull out a lot of interesting observations from correlations.
There are still some interesting positive correlations, but perhaps more importantly, there are some strong negative correlations that can be helpful to consider when building a lineup.
I’ve highlighted the most important types of correlations below, provided examples and touched on how to use that information in your DFS lineups.
Negative Minutes, Negative DK Correlation
I am starting with a negative minutes, negative DK points correlation example. This is possibly the most useful relationship we can find using this correlation tool.
Remember, the negative correlation means as Player A’s minutes/DK points/FD points increase, Player B’s minutes/DK points/FD point (typically) decrease – the values move in opposite directions.
This type of relationship is often between two players filling the same role on a team, and the example shouldn’t be surprising.
DeAndre Jordan and Jarrett Allen both are centers and rarely share the floor. Like most center duos, the only way one gets extended run is if the other is limited (foul trouble or injury).
Their DraftKings Points are also negatively correlated, although not as strongly as their minutes. Still, an above average performance by one player often happens at the expense of the other player.
This type of correlation is the worst for DFS, and pairings like this should only be used when dictated by the slate – if both players are significantly under-priced or there aren’t other viable options.
Typically though, this correlation has limited upside and an increased risk that one player has a below-average game.
Positive Minutes, Positive DK Correlation
These are the relationships closest to the QB-WR example from football. This positive correlation means these teammates’ stats tend to move in the same direction.
The positive minutes and DK points correlation makes these type of duos strong tournament plays, but can make them riskier for cash games. Their performances tend to move the same direction – which can mean both having big scores or both having below-average nights.
Eric Bledsoe and Khris Middleton are a great example of this positive/positive correlation. Their minutes are positively correlated, which isn’t surprising for two starters on the same team.
I’d rather focus on their DraftKings points correlation.
Now this is an interesting graph. Not only did they both have their best games of the season in the same game (that dot in the top right), but Middleton topped 40 DK points in 9 out of Bledsoe’s 10 40+ DK point games.
These are two players I would have no concerns about playing together in a tournament, as they tend to have great games together, and do not appear to limit the other’s upside.
Positive Minutes, No DK Correlation
Donovan Mitchell and Rudy Gobert have a strong positive minutes correlation. This isn’t surprising, as star teammates will often have strongly correlated minutes.
Does that make pairing Mitchell and Gobert a good play? It’s not necessarily good or bad. As shown below, their DraftKings Points show no real correlation. The performance of one player doesn’t help us guess the other’s performance.
If Mitchell and Gobert both project as great plays, you should feel free to play them together. However, a big performance by one player neither increases nor decreases our expectations that the other player has a big game.
No Minutes, No DK Correlation
Bam Adebayo and Tyler Herro last season provided a good example of players without a strong correlation. Given that they play different positions and roles on the Heat, this makes sense.
Their DraftKings points scored looks similar, with no real correlation.
Ultimately, whether or not two teammates are a good fit together in a DFS lineup will be dependent on the slate and game type you are playing.
Even teammates with a strong negative correlation can occasionally be optimal together in a lineup, depending on their salary and the other options at that position.
These correlations should be another tool for your DFS research and lineup building, and hopefully they will help you get the edge over time.
The Teammate Correlation tool is available here.
Data from Basketball-Reference.com