I think many teams need to take the time to stop and look for weaknesses in their designs. Kinda related to
Amdahl's law, you want everything to work fast, without there being a bottleneck somewhere.
You could have a really fast and accurate shooter, but that's no good if your ball-collecting mechanism is slow and clunky and can't actually catch and send balls to the shooter fast enough. The shooter would be hindered by the collector, so trying to make the shooter even better wouldn't do much good unless you improved the collector first.
The opposite also applies. You could have a great collector mechanism that quickly grabbed balls off the field, but that's no good if the shooter takes a while to fire, or if it's not very accurate. The collector would be stuck doing nothing waiting for the shooter to fire a ball, as you can only hold 3 balls at a time. If it missed, the robot would need to spend even more time recollecting more balls to shoot.
Basically, focus on making the weakest part of the robot better. Any team can do this. If you don't, all the other parts of the robot won't be able to perform at their best.