Ok, these responses are longer than I intended, but these are good questions and I want to give them their due.
I've heard several candidates say that there are a handful of teams we just can't beat as an explanation why simply making the semifinals at Worlds is a good result.... 1) Imagine that you have a one-off, "unwinnable" game. ... Any record of those working for you at the club level?
First of all, I don't think there are any teams we
can't beat. But I also don't think that we assume we win, say, 60%+ of the time vs teams like Italy and Spain, and that's what I base expectations (as opposed to goals) on.
As for quirky tactics, it really comes down to lineup management. Sure, FCP/Princeton give some unexpected looks, but the main variability comes in who you play and where. This is, frankly, much harder at the club level given the smaller player pool, but I've had good luck with some different lineups (playing 3 centers against, well, you, going 4 guards to max out OS/OD, etc). I'm hesitant to say I'd do too much of this at the NT level, because I think there should be very few unwinnable games for us, and using these sorts of tactics generally works by adding a lot of variability to the outcome.
2) The US team has tended to leave several spots open on its roster for much of the competition, since it costs enthusiasm to drop a player (but not to add one). Would you keep the same policy? If not, why?
3) Given the above, is there room for a "specialist" player?
I'm going to answer these together. First, I'd keep the roster as small as needed. We have a good sense of who manages GS well, but you never know when injuries, transfers, big cup games, etc will throw a wrench into this and require you to add a specific position.
That said, we certainly have room for a specialist. In fact, there's a fascinating guy out there now with only 13/13 IS/ID, but 20 rebounding. You'd certainly have to game plan him in, but I think that could be used very effectively. There's another PF that was a backup much of last season that has pretty poor (by our standards) rebounding, but some incredible JS/IS ratings. Again, depending on who you put around him, that can be a very valuable weapon.
4a) Do you understand the game engine better than your opponents? Why (or why not)?
I've done a lot of work using the BBAPI to research tactics. I've been around since mid-way through season 2, and have watched countless games at the club, NT and u21 level, but I think we've all done that. The stuff I've done recently with the bbapi has been illuminating, and has turned up some surprises.
For example, if you're opponent is playing LI, both 2-3zone and 3-2zone are equally effective (in fact, in upper divisions, 3-2 is slightly better, 53% vs 50% chance of beating the LI team on a neutral court). This makes some sense (shutting down the passers prevents the inside guys getting the ball in good shooting positions), but is not intuitive.
Another example, if you think your opponent will play a 2-3 zone, the worst thing you can do is be patient (only a 39% chance of winning). Better to take the open shot as soon as you see it (both motion and RnG predictably kill it, so much so that to ever play a 2-3 zone, you have to know their outside game is incredibly weak, which is rare in competitive NT games).
A final example, everyone gets enamored with the inside scoring ratings Italy put up in the world's, but unless you can get those type of insane ratings, a lower outside scoring rating can actually be more useful. What I mean is, each level of outside scoring is about
twice as valuable as inside scoring in predicting outcome and point differential.
There's more, but I'm running out of characters in this post, so I'll continue in a sec...