Its like talking to a wall.The release happens far above the atmosphere. Hence, the interceptor is not yet fired. Even after release, there will be 1-2 minute to strike. The missile is fired only during that last 20-30 seconds of strike. So, recalculation is possible. 10 seconds of tracking is enough to determine the trajectory after release.
1) The release happens the bloody bus moves and releases agaon. The problem is not that the RADAR cannot track the change, its the interceptor won't know where the bus will be whehn **it** gets there. Its like a fielder catching a ball. He can see its trajectory and place himself inline to catch it. If the ball swerves in the air (like a bus moves) he is now in the wrong place and he is going to struggle to get to the new place in time. Now imagine if if keeps swerving.
2) You also forget that an interceptor has to move through the air to get into space. That takes times. At the same time the RV is nearing its target. Its a dynamic system.
The interplay between the two is known as an "engagement envelope". Meaning the period of time that the interceptor has to hit the incoming RV. That is calculated as E=Time to reach Target(Tt)-Time to achieve intercept (Ti).
E=Tt-Ti.
Ti is the time an interceptor takes from acuqistion+ to firing solution+ to launch+ to arrival.
With an MIRV bus, when it changes its trajectory, the whole process has to start all over again. It =has to reaqcuire the target, plot a new solution and fire. All the time E is reducing. By the time it is ready to engage, BOOM.