Here's a simple thought experiment. Take a look at my PG,
Kevin Bassler. He's at ~70 BH and ~70 Pass. Through 8 games, he's averaging 1.8 TOs per game (24.1 min) against the #1 rated SOS. Last year as the starting PG against the 8th-rated SOS, with the same 69 BH and an average of ~65 Pass, he averaged 1.7 TOs in 22.1 minutes. So that number of ~1.8 TO/game has been pretty reliable and well-established at this point. He takes a little over 4 shots per game (actually almost 5 this year).
So here's the easy part for most coaches. Let's say you could add or subtract 10 points of BH from Bassler. How much do you think it would impact his turnovers? My guess is that if you added 10 points of BH, he'd improve his turnover rate by no more than .2 to .3 TO/game. If you removed 10 points of BH, it might be a change of .3-.4, which could be mitigated back to the .2-.3 range by having him shoot a little less frequently. Those are very liberal estimates. On a reasonably efficient offensive team, you might stand to gain or lose approximately .2 points per game of offense from that 10-point BH change.
Now the hard part. If you added or subtracted 10 points of passing from Bassler, how much would that impact the team? Passing almost certainly has less impact on turnovers than BH, so let's just totally ignore that. The truly important factor is the impact of passing on teammates' shooting percentage. My best guess is that 10 points of passing in either direction would change the offensive expectation by something in the neighborhood of one made basket per game (I don't want to be more specific than that, but it's close enough, within 100% error). That's more than 2 points per game. An order of magnitude more important than 10 points of BH. If anybody has done some investigation into the impact of passing on FG% and thinks I have more than 100% error (IE impact less than 0.5 baskets/game or greater than 2), feel free to refute this. But otherwise, the point is valid. It's basically an order of magnitude difference in importance.