Engineering_Design_Process_in_Competition_Robotics _20091204.pdf
I generally recommend reading this paper around this time of year. I always re-read it the night before kickoff. It is so easy to jump to step four - unless you are paying specific attention not to, you'll find your team jumping to conclusions about the game. Those first three steps really drive the direction the rest of the season goes. (Even in this thread, I see a lot of skipping step one.)
A great brain exercise to do (in step four) is to determine the Minimum Competitive Concept. Frequently, once you identify the simplest robot that can still be competitive in the game, you can then tweak and improve it to meet your specifications (step 2). This is basically what 33 did in 2014.
When thinking about robot designs, I often think in terms of elevators, and polycord, and four bars, and shooter wheels. Another valuable thought experiment is to draw the robot size box on paper and try to optimize the travel path of the game piece through the robot without considering mechanisms. You can often zero in on how you want the robot to interact with game pieces this way. This is also a fantastic way to directly engage newer students beyond "having them come up with crazy ideas". Team 33s 2013 robot was partially the result of this thought experiment.
Cheers, Bryan