Ok, I did some digging, and it has to do with the plus/minus so take what you will from this.
I took the +/- for the whole year, and the minutes played, then divided the +/- by the minutes played, and then multiplied that by 100 for ease of use.
This is what I found, we outscored opponents when the following players were on the floor.
Fesenko, 8.33 pts per 100 minutes.
Miles, 2.69 pts per 100.
Price, 1.39 pts per 100.
Watson, 0.70 pts per 100.
Sap, 0.61 pts per 100
Opponents outscored the Jazz when the following players were on the floor.
Kirilenko, -1.65 per 100
DWill, -2.49 per 100
Harris, -2.64 per 100
Evans, -3.24 per 100
Al, -5.71 per 100
Elson, -9.67 per 100
Bell, -10.16 per 100
Hayward -16.01 per 100
Favors -18.24 per 100
Okur -19.64 per 100
Cousin -94.44 per 100
What this tells me is...... nothing.
You are welcome.
Why not just use 82 games on court which are already calculated per 48 minutes? Using 100 is misleading being over two games in length.
https://www.82games.com/1011/1011UTA.HTM
While Fez, Price, and Watsons numbers can be thrown out from low sample sizes and playing against backups, Miles sticks. CJ also sported an impressive 3.06 adjusted +/- and that stat usually catches fluke scores.
Anyway you slice it we were a lot better with CJ on the court last year. Statistically it is not debatable. I'm not sure the main reason for it, perhaps because he was one of the few players that could halfway spread the floor.
Can the analysis be broken down by first half of the season vs second half? I am willing to bet the issue is he played more during the first half of the season than he did in the second half, so he has less time during the losing games than during the winning games.
Correlation does not prove causation.
He actually averaged more minutes in the 2nd half of the season, averaged a radical 30 mpg in march. His +/- did fall like everyone else, but he had built up quite a cushion if I remember right. He was at +16 /48 or something ridiculous at all star break.
He had some monster games in the 2nd half like this.
https://espn.go.com/nba/boxscore?gameId=310316026