Compare up to 4 auton routines by expected value (probability × points). A reliable 10-pt auton beats an unreliable 18-pt one — this tool shows you the math.