This team is taking a step backwards next year if Rozier is the starting PG.
YOU ARE LOSING KYRIE IRVING. Duh you are taking a step in TALENT backwards, but as far as ball movement? Idk man.
The ball movement story again. Last year, Terry held the ball as long as Kyrie did during his front court touches. NBA tracking data shows Kyrie at 4.27 sec/touch and Terry at 4.22. Does that .05 of a second mean the ball is popping when Terry is in? https://stats.nba.com/players/touches/?Season=2018-19&SeasonType=Regular%20Season&TeamID=1610612738&sort=AVG_SEC_PER_TOUCH&dir=1
It is true that Terry passes a bit more per minute of playing time, on average; 1.8 passes/minute compared to 1.55 for Kyrie. But his passes have no purpose; he's just getting rid of the ball after dribbling to nowhere. Per basketball reference, Kyrie had an assist %age of 35; Terry had an assist %age of 17.7. Per NBA tracking data, Kyrie's assist-to-pass %age (adj) was 15.6; Terry's was 8.8. That put Terry below Jaylen Brown and Jayson Tatum as an assist creator - i.e., bad, especially for a point guard.
Be careful how you interpret this data. The sample for Kyrie is dominated by time spent with Kyrie as the primary ball-handler. The sample for Terry has a huge portion where he shared the floor with either Kyrie or Marcus and was really in a 'wing' role. His touches in that role would be more often those of a finisher.
Try isolating on just the games that Terry started (where he would spend a much larger percentage of his time as primary ball handler) and his assist-to-pass percentage goes up significantly.
Another way to look at this is to isolate on lineups with Terry but without Kyrie or Marcus sharing the floor. Over the last two seasons, with either of those guys on the floor with him, Terry's AST% (percentage of teammate's shots that he assisted) is around just ~17%. But in lineups where Terry is the only small guard, his AST% is closer to 25%, which is a typical PG assist rate. Note that even in that role, he still has a microscopic TOV% rate of just 8.4%. So increased playmaking responsibilities didn't result in increased turnovers.
What do you make of the NBA tracking stat - %age of player passes that end in an assist? I can't see why playing with Smart or Kyrie would tank that stat for Rozier; I think it's pretty strong evidence, confirming the eye test observation that he doesn't know how to create good shots or find open teammates in position to shoot. Note that Rozier's number is lower than Tatum and Brown's.
To me, this actually fits with his low turnover stat. He dribbles around for about as long as Kyrie, then just dumps the ball back to the nearest teammate (or launches a low percentage shot).
On the Bball reference stat, can you send a link or a screenshot to show Terry's assist % when he's in a lineup without Kyrie or Marcus? I don't see how to derive that from the site.
With regard to the nba.com tracking stat: assist-to-pass percentage, role on the team has a lot to do with it. Let's look at 2017-18 for a second. During the regular season, Kyrie was the dominant ball-handler and during the regular season he had an AST-TO-PASS% of 9.7%. His last game was Mar 12.
From the start of that season through Mar 12 (ignoring the small handful of starts he got before then) Rozier was primarily coming off the bench and sharing a lot of time with either Kyrie or Smart. He had an AST-TO-PASS% of just 6.4%.
After Mar 12, Rozier got to start most games. Through the end of the regular season he averaged 68.2 passes and an AST-TO-PASS% of 7.3%. During the playoffs that went up further to 8.6%.
Note that when you look at these samples you do see that Horford's share of the play-making went up after Kyrie went down. That probably explains why Rozier's number went up to close to, but not all the way to where Kyrie's number was.
To isolate on different 5-man lineups, the nbawowy tool is pretty useful:
https://www.addmorefunds.com/nba-wowy/You will need to create an account but it is free.
Here is a specific link to data for Rozier on the floor over the last two seasons without Kyrie or Smart (note - some playmaking would still be made by Horford or Hayward in these configurations) with a 24.7% AST%:
http://www.addmorefunds.com/nba-wowy?query=1apx62ohaoaIf you further trim Hayward from the sample, then Rozier's AST% rises to 26.3%:
http://www.addmorefunds.com/nba-wowy?query=az7p60spbfdNow, if we simply add Kyrie back in to the latter data set, Rozier's AST% immediately plunges to 15.8%:
http://www.addmorefunds.com/nba-wowy?query=inddxop98tWhat's interesting if you look at these three is that Al Horford's AST% also drops when Kyrie gets added to the mix, from almost the same as Rozier (22-24% in the first two samples) down to just 17.4% in the w/Kyrie sample.
So while when Rozier is the primary ball handler he does dish out more assists, he shares that playmaking more with Horford (and Hayward this last year) when they are on the floor with him. But when Kyrie gets added to the floor, he tends to dominate the playmaking.
The latter is not necessarily a bad thing. It is just an indicator that they played a different style of basketball with Kyrie on the floor.