[Tom Line]

You will get a very wide set of answers depending on how old the teams are that are answering.

For instance, most well established teams have a playbook that they draw from that already has a wide chassis, a narrow chassis, plus the trannies and drivetrains to run each.

Those teams will immediately focus on overall game strategy, which should take no more than a couple of hours to talk through, and then how best to start designing their manipulator. Likewise, those same teams will already have a well developed autonomous program that is reasonably quick for them to implement.

In many instances the "ideal" manipulator will drive your chassis design.

If I could conjecture a bit: some people have suggested that the game may radically change this year. I think perhaps they're right (crosses fingers). I think that when teams can start selling pre-built chassis, it's a pretty fair bet that the game design is in a rut.

To wrap it up:

Determine a strategy using rough engineering guesses ( torques from the motors, engery calculations, etc). Estimate the power required for the tasks, since the one true limitation we have is motors. That strategy should determine the direction you'll take for the game. Then decide if your manipulator will need to drive the chassis direction, or if you can develop them in parallel. Then break into teams and get to work!