Nylon Calculus: Thunder-Jazz win probabilities, style, matchups and more

While Russell Westbrook will receive all the attention, this is a series that will be marked by defense with guys like Paul George, Steven Adams, Rudy Gobert, Ricky Rubio, and bench defensive phenoms Ekpe Udoh and Royce O’Neale. If Andre Roberson and Thabo Sefolosha were healthy, this would arguably be the most stacked defensive series in years. Everyone, I imagine, will rate these defenders by some combination of blocks, steals, and plus-minus, but how about we rate these defenders by an arbitrary measure I just created that only looks at a slim snapshot of what a defender can do in an absurd way?

One of my favorite stats — offensive fouls drawn not including charges, which I call “takes” — was just released on pbpstats.com. As you can see, a lot of the leaders are weak offensive guards who don’t seem to offer much resistance on defense. Yet from my research that stat is still signaling some strong defensive value, at least in incremental forms. Thus, how about we look at defenders who can pick up that stat while blocking shots?

Enter the geometric mean, which deals better as an average if you want to look at players who do well in both stats rather than an average just being overwhelmed by one. Some players from this series show up here, like Gobert who picks up an impressive number of takes given his size or Roberson, who is sadly still injured. Royce O’Neale and Derrick Favors also rated well but fell outside the top 15. But Jerami Grant has the second highest mean behind unicorn Kristaps Porzingis. The Thunder, in fact, have been better when he’s been on the court, and let’s hope the playoffs don’t reduce his time on the court in favor of Carmelo Anthony, who has an aversion to defense and, unfortunately, made baskets this season too.

Table: geometric mean of blocks and “takes” per possession, min. 1000 MP

(Source: pbpstats.com)

— Justin Willard (@AcrossTheCourt)

Defensive matchups

In many ways, the Jazz and Thunder will be facing each other anew in the first round.

Their regular-season series ended in 2017: they met once within the first week, then three more in December. At the time, Joe Johnson was still in Utah; Royce O’Neale was playing just 11 minutes a game; and Rudy Gobert was in the midst of nursing a left knee injury that would sideline him for a month.

Since then, roster changes have occurred. Donovan Mitchell has solidified his standing as a Rookie of the Year candidate, completing the 2017-18 campaign with the most points and minutes on his team. He finished at 54 percent true shooting on 29 percent usage. The Jazz were about 5 points per 100 possessions better with him on the floor than off (and even more if we exclude garbage time). Part of this was attributable to the time that he shared with Rudy Gobert, Joe Ingles and other high-impact players, but even if teammates, opponents and other factors are accounted, his contributions were still estimated around +2 by ESPN’s Real Plus-Minus.

The Thunder have seen changes of their own, most notably the loss of Andre Roberson. That stands to make a big difference from a matchup perspective.

The following chart shows how Oklahoma City matched up defensively against Utah during their regular-season series. Each line indicates the defender who spent the most time guarding a specific offensive player on a given possession. The thicker the line, the more frequent the matchup.

Roberson served as the primary defender on over 40 percent of Mitchell’s offensive possessions. Although there were confounding variables, the Thunder defense appeared to perform better during these matchups, and Roberson’s on-off numbers mirrored their season averages.

How OKC adjusts to his absence will be worth monitoring. Of course, Paul George is available to assume some defensive responsibilities for Mitchell, but he also shoulders a significant offensive load. We’ll see where head coach Billy Donovan goes and whether he makes tweaks from game to game.

— Positive Residual (@presidual)

Offensive styles

These charts are not meant to evaluate whether an offense is good or bad. They are designed to help illustrate how teams go about the goal of trying to put the ball in the basket. Each team’s offense is evaluated on four stylistic spectrums.

Ball movement is measured with the average touch time for each team, from the NBA’s player tracking statistics. A lower average touch time means the ball is moving from player to player more quickly.

Player movement is measured with a combination different NBA.com tracking statistics, and works out to average distance traveled per 24 seconds of offensive possession.

Pace is measured with the average length of an offensive possession from Inpredictable, a more accurate representation for how quickly a team is working than traditional pace.

Shot selection is measured with MoreyBall percentage — in this case the percentage of a team’s true shooting opportunities that came at the rim, from the free throw line, or on a 3-pointer. It’s a generalized measure but captures something about how much each team hews to the shots that are, on average, the most efficient.

On the graph below you’ll see a line for each team’s offense. As the line moves away from the center of the graph on each axis you’re seeing more of that stylistic trait. For example, shot selection shows a (hypothetically) more efficient shot selection the further you are from center. Here’s how the Thunder and Jazz match up.

The Thunder’s offensive system is all built on Russell Westbrook’s shot selection. The push the pace and take a lot of 3s and shots at the rim, but in the halfcourt there is a lot of standing around and watching him work. Utah, on the other hand, plays a very deliberate halfcourt game with the ball flying around the perimeter as they try to create an opening.

— Ian Levy (@HickoryHigh)

Win probabilities

To project the series, I am using my in-season game projection model. The model is trained off historical game data, and accounts for rest, travel, team strength, and matchup. Since I began using the model to predict outcomes, I have been able to correctly identify the winner in about 70 percent of games.

This series could very likely be an upset for the Jazz, featuring the second-best odds to advance out of any lower seeded team in the 2018 playoffs (Pelicans have a 45 percent chance). With an over 60 percent chance of going to six or seven games, this series is absolutely one of the best in the first round. The Thunder hold the two most likely scenarios, either winning in five or seven games, but expect each game to be close and overall a really competitive series.

— Jacob Goldstein (@JacobEGoldstein)