The obvious answer is building a winning robot. Let's look at what that really means.
First and foremost, your analysis of the game (just after kickoff and before the design phase) must be spot-on. You have to deeply understand the game and all its nuances, and how it will play out - not just week one or week six, but divisions and Einstein too. Without this, the rest is not going to be effective.
Then you need to understand the capabilities your robot needs to have to score a lot and be hard to defend (i.e., play the game well). This means WHAT things you must do, not HOW you will do them.
Only when those two very important parts are done, can you start on actual design and prototyping. Design a robot that will do what it needs to, then fabricate it the very best you possibly can. Many teams feel that "cut & try" is good enough. Maybe it is, but that's not the path to Einstein.
Last but possibly more important than all the rest is your drive team. They should have a few hundred hours experience with the robot (or its duplicate <hint>), so that driving from Here to There, lining up to Score, driving around defenders, and EVERY other thing you CAN imagine they will need to do during a match, they can do with their eyes closed. Then, when something you didn't imagine comes up, they can concentrate on managing that, since the rest is so well-practiced it doesn't need much mental effort to be perfect.
All this is easy to say, but not easy to do. Miss any of it, you're watching Einstein instead of playing it.
1676 also struggles with these. We've been finalists at CMP before (beaten by 1114 & 469) and in the Elims many times, but we have yet to crack onto Einstein.
Maybe next year
