In my last article, I laid out a rotation and minutes projection and discussed why each player will get the specific number of minutes that I projected. In this article, we are going to look at typical substitution patterns, and chart what the substitution pattern might look like for the Utah Jazz this year.
Typical Substitution Pattern
Nylon Calculus’ Seth Partnow took time to analyze the typical substitution patterns of a NBA team.
Seth first looked at how many minutes each team played their starters together. The average NBA team only played all five starters for 13.87 minutes per game. When considering the advantage that playing all the starters give you, this seems low. The team that used their starters the most were the Los Angeles Clippers at 19.78 minutes per game, and the team who used their starters the least was the San Antonio Spurs at 10.51. In 2014-2015 the Utah Jazz played their five starters 12.53 minutes together per game.
Seth then laid out a timeline of how many starters were on the floor. This is a good way to see the substitution patterns of the average NBA team. It’s evident that the most starters are on the floor at the start of the first and third quarters. The ends of the second and fourth quarters are also laden with starting players. The dip in number of starters on the court at the very end of the game is likely attributed to blow outs.
The next, and perhaps the most important, take away from his research is something you would expect to see. Starters player better with other starters. Seth went through and charted out the +/- based on the number of starters each team has in at one time. As you would guess, the more starters you have on the floor, the better.
Proposed Substitution Pattern
Below is a layout of how substitutions could be managed to meet the rotation discussed in my previous article. These are all approximate and will change on a nightly basis based off of game flow, matchup, and injuries. As I laid out this substitution patterns, I kept the information from Nylon Calculus in mind. My main objective was to get more minutes with all five starters on the court than the 13.87 average.
In the first quarter, you would see the starters play a good deal of minutes together. The first substitutions would be Joe Johnson coming in for Rudy Gobert. Johnson would play at the four while Derrick Favors slides to center. The next substitution would be Gordon Hayward checking out for Trey Lyles. At that point, Johnson would move to small forward while Lyles mans the power forward position. The last substitution of the quarter would consist of Dante Exum coming in for George Hill, Alec Burks coming in for Hood, and either Boris Diaw or Jeff Withey for Favors.
This pattern gives the Jazz seven minutes with all five starters on the court at once. The crucial moment will be the final two minutes. Can that bench unit play enough defense to hold the lead that the starters set?
The same unit that ended the first quarter would start the second quarter, with the exception of Gobert who will come back in at center. The first substitution will be Hayward coming in for Johnson. Hopefully, the early substitution would allow Hayward to play some minutes against the other team’s backups to get some easy buckets. The next three substitutions would be Hill coming in for Burks, Johnson coming in for Lyles, and Favors entering for Gobert. During this small stretch, Exum would play the two next to Hill. The last substitution would be Rodney Hood coming in for Exum. This lineup of Hill, Hood, Hayward, Johnson, and Favors would be an offensive powerhouse.
You will notice that Exum and Burks play almost exclusively together in this rotation. The defensive prowess of Exum is a good fit next to the limitations of Burks, while the offensive capabilities of Burks compliment Exum. Lyles and Gobert would have a similar setup. Gobert’s rim protection would help as Lyles develops and defense and Lyles’ shooting will help spread the floor with Gobert in. You could also likely see some four-five pick and rolls that end in Lyles lobbing it up to Gobert for the hammer.
The starters would start the third quarter. The first substitution would be Burks coming in for Favors and Exum coming in for Hill. Hood would slide to the three and Hayward to the four. The next swap would be Johnson coming in for Hayward at the four. These are crucial moves as you want Hayward and Hill to get some rest and be fresh for the fourth quarter. The next substitutions would be Favors coming back in for Gobert, and Lyles coming in for Hood. Johnson would slide from the four to the three, and Lyles will play his natural position of power forward.
While you have Hayward sitting for more than half the quarter, you still have offensive firepower with Burks, Johnson, and Lyles all in to get buckets.
To start the fourth, you would see the previous unit on the floor with Diaw/Withey in at center. This would last for a couple of minutes until Hill, Hayward, and Gobert come in for Exum, Johnson, and Diaw/Withey. Hood and Favors would then replace Burks and Lyles. The Jazz would close out the game with their best group, the starters. If any player is to get more fourth quarter minutes than what is shown, it would be Johnson due to his willingness to take a big shot.
Analysis of Proposed Substitution Pattern
This projected substitution pattern for the Utah Jazz is in line with a typical pattern. This graph lays out this substitution pattern on a linear timeline. The purple is the number of starters on the court at a time. The red line is the average number of starters on the court at that minute, per Seth Partnow’s data that was previously analyzed. Yes, there are moments where the Jazz will have fewer starters on the floor than average, but due to the depth of this team, that is bound to happen. The benefit of this pattern is the amount of time that all five starters play together.
This lineup gives the Utah Jazz 17 minutes per game with all five starters, compared to the 13.87 MPG average among the league. That gives the Jazz 3.13 more minutes with all five starters, which based off of Seth’s data gives the Jazz nearly a half-point per game advantage. That is a big advantage for something that the staff can mostly control.
The Utah Jazz will have an incredible amount of flexibility this year with their substitutions. If Snyder wants to go small, he can substitute Johnson for Gobert. If Snyder needs elite defense, we could see a lineup of Hill, Exum, Hayward, Favors, Gobert. Giving Snyder a roster with this many options has to have him drooling in excitement. Sure, this pattern will likely change on a nightly basis. Injuries, matchups, and the flow of a game will naturally have a big impact on what the game’s rotation looks like. However, if the Jazz use this rotation as a baseline, they’ll have an advantage over most of their opponents from tip off.