He can always run LI. With prediction he could not run motion every game anymore. All teams need to swap princeton/LI or motion/LI now because of prediction, making it a rock-paper game. This actually hurts outside teams more than inside teams if you think about it. Not in terms of winning, but in terms of building a resume of shooting games. Well then again, an LI vs. 2-3 with good shooters will probably give you better 3pt % than motion against 3-2.
Absolutely.
This isn't anything official because I know nothing more with the GM title than I did before, but I fully believe based on quite a few seasons of Princeton with big men I built specifically to play Princeton that there's something fundamentally wrong with the algorithm for calculating the percentage of hitting jump shots and three point shots. The main problem is that "open' shots (i.e., ones which would generate an assist, rather than having a defender contest it) certainly seem like they're opposed by the opponent's team OD. If you have high OD at every position, even shots that are open are still very low percentage. So basically, OD gets you steals, helps contest outside shots and then even if the shot is open, still depresses the shooting percentage to the point where it's depressing.
Of course, if you have high enough JS/JR, you can overcome that - which can certainly be done in the lower levels of the game, but as the level of competition increases, the cost of higher shooting (both JS and JR) grows much higher than the cost of higher OD, both in training time and in salary expense.
It's simply unfortunate for me that I would rather see my PF (only 11 JS/8 JR) take a three guarded by an opposing PF than I would see my best outside shooters take "open" shots. Same thing with jumpers, too - I had several players who shot better on guarded jump shots than uncontested ones, and that even included guards against other guards.
In terms of prediction, yeah, it killed my desire to run Princeton exclusively (though the 3-2 was already proving sufficient to cripple it most nights). There's still push the ball and the isolations (if I had built stronger offense on my bigs, maybe an inside iso would work), but it's still falling prey to the same fundamental flaw - assisted inside shots are very high percentage (usually), contested inside shots are still pretty high, and then unless you have a mismatch you're likely to shoot very low.