At least, I think. I admit, I don’t totally understand how adjusted plus-minus is calculated. I get the point of it, but the formula is a little tougher to get a grasp of.

• Proving causation is nearly impossible in these arguments, yes. But my intent wasn’t to provide a scientifically verifiable conclusion. I wanted to raise the question of which Pistons should have a chance at playing more with Ben Wallace (and at whose expense).
• Other teams’ worst players are playing during garbage time, too. I think that who “wins” garbage time tends to even out. In the long run, the Pistons’ bench warmers’ plus-minuses aren’t hampered by playing in garbage time.
• My intent wasn’t to analyze how playing with Ben Wallace would help the the numbers of players. I’m looking at how playing with Ben Wallace would improve a player’s ability to help the Pistons. If Wallace and Kwame Brown go up for a rebound, and Ben Wallace gets it, that hurts Brown’s stats. But it doesn’t affect the Pistons.
• I definitely see the logic of pairing Wallace with a shooting big like Villanueva or Maxiell. The Pistons have played decently when one of those two joins Wallace. But they’ve played decently with Wallace and Brown, too. Maybe they should give that combo more of a chance  – at least just to see how it works. The numbers suggest it might work better than what Detroit has been doing.
• I’m not sure whether Brown would maintain his production. But if the goal right now is to evaluate individual players, I’d give him more minutes with Wallace to see how he responds. If you don’t, you’re comparing apples to oranges.
• If you were to remove any point it would be Brown, who has played, by far, the fewest minutes with Wallace. If you did that, the correlation would be much stronger.
• I chose to use this single variable, because I think it’s very significant. Watching games, it’s clear Ben Wallace is the Pistons’ best player. After him, the order is muddled. It’s hard to compare players whose playing time with Wallace vary so much. The attempt was to find which players might deserve more of it.

Demonstrating that players who play a higher % of minutes with Ben Wallace tend to have higher per 48 +/- stats is neat, but does nothing to establish causation.  In fact, here’s a simple alternate theory: as an aging veteran who is, as you say, unquestionably his team’s best player, there is only one category of game time we expect Ben Wallace to miss consistently: garbage time.  Of course, the players who play in garbage time are not selected randomly – they are generally weaker players.

This also completely disregards any possibility of players being complementary: Ben Wallace may enable some players more than others.  Hint: the big men can be disregarded or, at minimum, must be analyzed separately from the guards.  Obviously the impact of one big on another is very different.  In the specific case of Ben Wallace, one might expect him to hurt the numbers of other bigs due to his strong rebounding and post presence.  Since his outside shot is limited, it makes perfect sense that he doesn’t share much court time with Wilcox OR Kwame.  Shooting “bigs” (Villanueva, Maxiell to a much lesser degree) likely thrive more and are clearly playing with him more…  Does this mean they are “over-rated”?

Brown is outperforming the “Ben Wallace line” significantly, but do you truly believe he’d move up the more he shares time with BW?  Seems unlikely…  It’s debatable whether he would even maintain production.

Finally, if you take out two points (Summers and Wilcox), the entire correlation looks much weaker.

Please, if you would like to leverage numbers to enhance your understanding of the game, do so in a comprehensive way.  Viewing player performance along a single variable (in this case % of time with Ben Wallace on court) is so statistically flawed that any conclusions drawn are probably less likely than random hypotheses to be correct (example being Kwame Brown’s insane overperformance relative to everyone).