I have played a lot of games both on the console and PC. My conclusion is that AI defense needs to be overhalled, NOT mine. Unlike the computer I don't go for a steal every three seconds, and unlike the computer I actually watch my man on transition D so he doesn't get easy baskets. Another thing is I know that at least on the console it was a snap for me to drain threes with Shaq and most likely with other players like him.
I fail to see how tweaking is cheating, considering that it affects all players and slows the game down to where calling pick and rolls are actually a strategic move and not just for looks.
Out of the box, blocking shots is as easy as breathing. With DK's roster it takes proper timing and positioning. So now when I challenge a shot, just like in real life, I don't get a block every single time. In fact, I have used D Armstrong on defense for 10 games on my season so far and he has like 2 blocks. Out of the box I could probably get 4 a game with him playing REAL DEFENSE. That is jumping not sticking my hands up, in real life the latter only works on 8 year-olds.
With DK's latest roster some people notice that point guards don't shoot loads of random threes in a game.
This what I have experienced on a tweaked roster: opponents FG% usually ranges from 35% to 45%, although I played the Kings and they managed 50% shooting. Both teams never get over 10 blocks, usually much less but the computer can still block jumpshots with abnormal frequency. In terms of strategy the game takes a lot more thought, I can still shoot 50% but it takes lots of pick and rolls and passing. With that I easily average 30+ assists per game. 95% of my points are not scored by taking advantage of the "steal, steal, steal" mentality computer AI defense.
It is important to note that even on a tweaked roster the computer can shoot in the 60-70% range from the field when proper defense is not played. If you don't pay attention guarding J Kidd he will still launch a three and make it.
It has been awhile since I played with the default rosters but it also would be interesting to test how fast the players travel compared to the speed of a ingame pass(which are usually bullet passes). I know on the slower tweaked rosters passing is much faster relavtively, as it should be. Thus, passing to the open man is never in vain.
I honestly have not used freestyle much, except on the console. I do not have a dual analog stick controller so taking advantage of those new features are, to me, more burdensome then not.