Quote:
Originally Posted by Kevin Leonard
Also discussing the theoretical highest level of play. And I was asked a question about what death cycles were, so I explained them.
Furthermore, I believe the highest possible level of play for Aerial Assist was never met, as perfect death cycles were never executed, and repeatable catching was rarely executed.
The idea of this thread was to ask "How do you determine what the highest level of play is during build season?" because I seem to foresee it being higher than it ends up being, or at least different than it ends up being every year.
If you want to debate the merits and pitfalls (of which there are many) of death cycles with me, we can do so privately.
|
You've just answered your own. You look for highest possible levels of play. That's worthless. You need to look for highest achievable levels of play.
Now, how do you determine what is going to be important at those levels of play? I typically put together a simple simulation (excel) that lets me do various what if scenarios. I then tweak the number of things and see what happens. This often shows things that aren't worthwhile or things that are limiting factors. 2013 - it fairly easily showed that climbing could be out scored by a decent shooter[1]. 2014 - showed the importance of auton and that solo cycles were almost worthless. 2015 - That uncapped stacks were worthless.
Aside about Death Cycles - I think you VASTLY overestimate these. Actually, not pulling punches, I think you're full of crap when you say it was optimal AA play. Why? Because it requires 3 robots to be perfectly suited for each other. Which then has to happen in an actual alliance... which given how the draft works is all but impossible. Not only that, but given the ease of adding some PVC up to 5' to a defender bot (bonus points, backboard for inbounding) it makes the inbound to a stationary scoring bot VERY difficult.