2016-17 Gonzaga Bulldogs: Which big man was better?
Gonzaga had an embarassment of riches at the center spot with Przemek Karnowski and Zach Collins, but which big man was actually better that season for them?
This is the final article in a series about the 2016-17 Gonzaga Bulldogs. Click here for my overview on the team, and here for my analysis on their schedule.
For my subscribers, I’ll be pausing the Hoops Hindsight series for a few weeks to focus on the current season analysis through the NCAA Tournament. I’ll return once the tournament is completed with a special overview of this year’s champion, plus a series on the 2013-14 UConn Huskies.
The 2016-17 Gonzaga Bulldogs upheld their program’s tradition of excellent basketball from a traditionally mid-major conference, along with a legacy of standout frontcourt players. Continuing that particular legacy of big men were 2 players in particular:
Senior center Przemek Karnowski, who was returning from a medical redshirt year
Freshman center Zach Collins, the highest rated recruit in program history at the time
Both players played less than they might have at another program, with Karnowski playing 23 mpg and Collins 17. Both players had very successful seasons, with Karnowski winning the Kareem Abdul-Jabbar award as the nation’s top collegiate center and Collins being drafted 10th in the 2017 NBA draft. The natural question for curious minds when two players at the same position for an elite team have success is: which player was actually better? Could Gonzaga have had more success by prioritizing one over the other a bit more? In this edition of Hoops Hindsight, I look at the evidence and try to determine which player was better, and how Gonzaga might have adjusted their rotation to get better results.
A wealth of data to help with research
In order to answer the main quesiton of who was better, I’ll rely on a few sources:
KenPom.com, and their player data
Sports Reference, and their advanced player stats
Synergy, and their offensive and defensive player ratings
EvanMiya.com, a great source of data by Evan Miyakawa with player on/off, lineup, and teammate chemistry data
I’ll focus on judging our two protagonists in three areas: individual offense, individual defense, and teammate impact.
Individual Offense
Both Karnowski and Collins put up tremendous offensive numbers during the 2016-17 season, with Gonzaga putting up an elite offensive season thanks in large part to their big men.
Karnowski had an effective FG% of 59%, with all but 1 field goal attempt coming inside the arc. Karnowski had the third highest assist rate on the team, and a slightly positive assist-to-turnover ration of 1.9-1.8, which is notable for a big man. He did a decent job drawing free throws but only hit 59% from the line, making his free throws not much better than his field goal attempts. Karnowski was a solid but not spectacular offensive rebounder, ranking 474th in offensive rebound rate per KenPom. He rated in the 86th percentile per Synergy in points per possession, and was most heavily featured on post-ups, as expected. Karnowski struggled a bit against zone defense per Synergy, dropping to the 71st percentile, as you’d expect of a post-up specialist.
Collins’s offensive numbers were maybe even better. He had an eFG% of 68% thanks to 67% shooting on 2’s and 48% shooting on a small volume (21 FGAs) of threes. Collins was a 74% FT shooter, which was a further boost. Cwas more turnover -prone than Karnowski and not as good of a passer, with 16 assists vs 60 turnovers on the season. He was an excellent offensive rebounder, collecting 12% of Gonzaga’s misses and placing 85th nationally. Synergy rated Collins in the 97th percentile as a scorer, and he was more effective than Karnowski on post-ups, as a roll man, in transition, and as a spot up shooter.
The two players both put up very strong offensive stats per 40 minutes. Collins averaged 23 points, and 14 rebounds per 40 minutes, while Karnowski put up 21 points, 10 rebounds, and 3 assists. Collins appears to have the edge, however, in most individual offensive categories. EvanMiya.com agrees, as their Offensive Bayesian Performance Rating favors Collins pretty clearly. This metric adjusts for teammate and opponent strength, so it seems like Collins was the superior offensive player.
Individual Offense edge to Collins
Individual Defense
It’s tougher to measure individual defense, as stats like blocks and steals occur so infrequently that they don’t accurately measure good defense. There are some useful metrics and advanced stats, which I’ll incorporate, but let’s take a holistic look at each player.
Karnowski was a solid defensive rebounder, getting 18% of opponent misses (362nd nationally). He also blocked about 4% of opponent shots, 261st nationally. Karnowski was fairly foul-averse for a big man, committing only 3.7 fouls per 40 minutes. Synergy rated Karnowski in the 94th percentile as a defender, as he rated as Very Good defending post ups and excellent on spot ups thanks to his large frame.
Collins was an even better defensive rebounder, getting 23% of opponent misses (83rd). He blocked almost 10% of opponent shots, 22nd nationally. He was fairly foul-prone, committing 6.2 fouls per 40 minutes; he fouled out of 6 games that season despite playing less than 20 minutes in all 6. Synergy rated Collins in the 87th percentile defensively, as his higher propensity for fouling negated some of the benefits of his shotblocking.
While Collins tended to make more of the notable defensive plays, I think Karnowski’s quieter contributions were also quite beneficial. The advanced stats around defense aren’t great, but they are pretty split. Per Sports Reference, Karnowski had more defensive Win Shares 2.6-2.4, although that stat is influenced by playing time. Box plus-minus favored Collins 6.0-4.4, although that is largely due to his higher block and defensive rebound totals.
EvanMiya.com tracks the number of defensive possessions for which a player is on the floor, and Synergy tracks the number of possessions where a player is the defender on a play. Using this we can estimate how often a player was targeted, by dividing the Synergy possessions by the total defensive possessions. Karnowski was targeted on 15% of defensive possessions, while Collins was targeted on 17%. Collins graded out as a little worse defender, and was targeted a bit more on his worst play type (isolation) than Karnowski was. Given this, it looks like teams felt like they could pick on Collins a little more than Karnowski, so I’ll give PK the nod for individual defense.
Individual Defense: Karnowski
Team Impact
Measuring the impact each player had on the team is really difficult. It’s nearly impossible to control for teammates, opponents, situations, and still have a decent sample size. Evan Miyakawa’s site does a nice job, however, so I’ll lean heavily on his analysis.
Per Evan’s data, Gonzaga was a little better on defense with Collins on the floor and almost identical on offense. However, Gonzaga’s opponents were a bit better when Karnowski played. The Bayesian Performance Rating metric favors Collins, likely because he didn’t play as many minutes with the better Gonzaga players.
One of the nice features on Evan’s website is the ability to look at how Gonzaga played with various combos of players on the court. For example, Karnowski played a lot with Gonzaga’s starting PG Nigel Williams-Goss (about 1,300 possessions). Gonzaga had an efficiency margin of +32 points per 100 possessions with the two of them together; that’s about 3 points better than the average of all Williams-Goss’ minutes. When Collins played alongside Williams-Goss, Gonzaga had an effiency margin of about +28 per 100, slightly below Williams-Goss’ average. So, Karnowski was a better frontcourt partner for Williams-Goss according to this data.
We can do the same thing for each of Gonzaga’s other key rotation players:
Josh Perkins had slightly more success with Karnowski, 26.0-25.3
Jordan Mathews had more success with Karnowski, 35.2-32.2
Jonathan Williams had more success with Karnowski, 30.2-25.5
Killian Tillie had identical numbers with both, 42.7-42.7
All of these players had more success alongside Karnowski than Collins. This seems like Karnowski was clearly more impactful, but I am not quite convinced. After all, Karnowski played a huge portion of his minutes with these players, and with several of them together; his minutes with Williams-Goss almost always had Williams as well, for example, whereas Collins more often played with only 1 or 2 of these starters at a time.
There are 4 main 4-man combos that played 50 possessions alongside each of Collins and Karnowski. When we control for the other 4 players and swap out Collins for Karnowski, we see:
When Collins played alongside Melson/Perkins/Williams/Williams-Goss, Gonzaga had an efficiency margin of +30/100; with Karnowski it was +10
When Collins played alongside Melson/Perkins/Tillie/Williams-Goss, Gonzaga had an efficiency margin of +40/100; with Karnowski it was +36
When Collins played alongside Mathews/Perkins/Williams/Williams-Goss, Gonzaga had an effiency margin of +21/100; with Karnowski in place of Collins it was +37
When Collins played alongside Mathews/Melson/Tillie/Williams-Goss, Gonzaga had an effiency margin of +34/100; with Karnowski it was +42
Karnowski was WAY more effective with the starters (lineup 3 above), while Collins largely held his own with the other groups.
It’s a really close call to pick out which player had a stronger team impact. Gonzaga played a bit better with Collins in, but against weaker opponent lineups. When it comes down to it, I’d probably favor Karnowski due to his better chemistry with the starters. That starting lineup played over 600 possessions together, almost half of Karnowski’s total and over 3x as many as the second most used lineup. Karnowski seems to have had stronger chemistry with the other Gonzaga regulars and was able to bring that to bear in the most important lineup.
Team Impact: Karnowski
Conclusion
Zach Collins had a monster freshman season for the Zags, and his contributions were huge in getting Gonzaga to the title game. His individual skill level was probably higher than Karnowski, as evidenced by his individual stats and hit NBA future. The 4 additional years Karnowski had in the Gonzaga program probably gave him the edge, however, and it seems to have helped Gonzaga reach their potential a bit more.
I’d have to say that Karnowski was probably the better player by a narrow margin that season, and Gonzaga was probably right to give him more time than Collins. However, having a player of Collins’ ability available to come in and wreck opposing bench units was a massive luxury for the Zags. Collins came up huge in the national semifinal, but Karnowski was a bit more consistent and blended well with the rest of the starters. It was nearly a dream season, and the two big men played a huge part in the team’s success.
As I noted at the beginning of the article, I’ll be taking a few weeks off from Hoops Hindsight to focus on the current season. Look for an article covering the 2021 national champions the Thursday after the title game!