Good games, weird data

Paul Millsap is the barometer for the Nuggets' playoff success

Paul Millsap is the barometer for the Nuggets' playoff success

My fantasy basketball addiction manifests itself in mostly harmful ways—attention span and cellular data taking the biggest hit—but it does provide one hidden benefit. Instead of watching the local teams struggle, I tune in to see how my squad performs across the league. I’ve particularly enjoyed watching the Denver Nuggets, an exciting young squad built around the Point Center trickery of Nikola Jokic. Jokic pulls together statlines that transcend big man expectations. Per Basketball Reference, there’s been only nine seasons where a player had 20+ points, 10+ rebounds, and 7+ assists. Jokic had the second highest true shooting percentage among these occurrences, after Wilt Chamberlain. It doesn’t hurt that Jokic has a hilariously herky-jerky, below the rim game that looks a lot like a pickup game superstar.

Jokic’s greatness produced some chatter around MVP votes from smart writers like Zach Lowe, and he ended the year firmly in the top five of my own MVP scoring model. Yet, when I watch the Nuggets, I fight this feeling that Jokic’s performance, while vital to the team’s success, isn’t the biggest indicator for victory or defeat.

Jokic may offer a consistently high floor, but does a good game from the Joker mean the Nuggets are likely to win? Is the more volatile Jamal Murray a better barometer for success? As the Nuggets fight to avoid a first round upset against the Spurs, I thought this question deserved further examination.

Ideas around methodology

Basketball Reference helpfully includes a Game Score metric for every game played. Think of it as shorthand for, did this guy have a good game?. Game score powered my Trae Young resurgence piece, and I think it will be useful here to see how players do in wins and losses.

I filtered out games for players who had less than ten minutes to ensure enough time in-game to measure. To benchmark Denver’s output, I added three other teams with different production levels.

As you’d expect, players typically had better performances in wins than in losses. A Nuggets player typically produced a game score 26.1% higher when they win, while a great team like the Bucks had a lower change (11.5%). A bottom-dweller like the Knicks had a much higher delta (39.8%). Intuitively, these ranges make sense: a team that often underperformed like New York would need a bigger bump to find success, while consistently great squads like Milwaukee have a high floor in wins and losses. The team-wide calculation allows us to set a benchmark for performance increases in a victory. 

I’m interested mainly in players with a higher difference in game score between wins and losses. Do these fulcrum players impact winning more than expected? When on, does this indicate a higher likelihood that the team will do well?

How does this look for the Nuggets?

For guys like Paul Millsap and Jamal Murray, they show a higher than average change in their performances in wins vs losses. Jokic fits roughly the overall team benchmark, fitting the gut feeling that he’s performing well regardless of outcome. At the other end of the graph, we have poor Trey Lyles and Monte Morris.

Do these performances impact team record?

Lift alone can’t prove or disprove this theory, but it gives us a framework to analyze certain players. In games where Jokic outperformed his average game score, the team went 27-11: in games below his typical output, the production only slightly dips, to 26-16.

Murray has a similar split in his gamelogs. The Nuggets went 25-15 in his poor games and 24-11 in his above-average ones. Millsap, however, appears to be the truest bellwether. In instances where he had a below-par performance, the Nuggets had a record of 17-17. When he’s on, Denver went an insane 32-4putting them on a pace for a 73 win season. 

So far, the trend holds for the early playoff results. In their game one loss, Jokic led the team in game score while Millsap struggled mightily, putting up half of his season average. Millsap rebounded in game two with an above-average output, helping to spark a Nuggets win. He again struggled in the Nuggets’ loss in game three at San Antonio.

Playoff samples might be small, but this trend shows a ton of promise. Millsap is not the most valuable player on Denver, or potentially even the second-most. Yet, for the Nuggets to fully weaponize their young promise, they need their grizzled vet Millsap to win the day.

Subscribe to Warm the Bench

Sign up now to get access to the library of members-only issues.
Jamie Larson