I was watching a video clip of a seminar, and the participants were paired up, one was practicing an attack, in this case a particular strike, the other followed with a response. This is one of the most common teaching methods, and I am assuming that the thinking behind it is to condition a reflexive response to a stimulus .... See X coming, do Y, or Y + Z + ..... until it becomes ingrained. Everyone was doing the drill in rhythm, repeating this rhythm over and over.
I kept wanting the attacker to switch it up and throw a different attack, or mess with the timing, because of course that's how I learned, and though I know that this part comes later for those who spar or do some other kinds of random drills, it made we wonder, do you really need this first step at all?
Instead of only training to deal with what your opponent DOES, would it not be better to either train to deal with what your opponent is THINKING of doing (as you will get to deal with what they actually do do as a byproduct anyway)? ....... Or alternately, train to MAKE the opponent do the strike you want them to do by giving them a reason to throw it?
The repetitive, reflexive exercise in the clip may well be training how to predict the incoming danger of a particular strike from watching the opponent's physical motion, and maybe even from the physical set up, but if the feeder KNOWS what they themselves are going to feed, then you are not going to learn how to predict the intent itself, and the smaller micro adjustments, or 'tells', that happen as they decide what to throw.
Neither are you going to understand WHY they chose to throw that strike, or HOW to get them to do so.
If you are getting your information from this drill only, you will always be behind the opponents intent, in other words always behind the timing.
Tactically any partner practice like this where one is waiting, makes better sense if there is a reason for the waiting.
One reason could be being unsure of what is about to happen, and therefore making space until you can be more sure of what is going on. This set up could train reading the opponent's intent.
Alternatively, waiting for something you know is about to happen could potentially be a ploy/bait/draw of course, but then there needs to be a reason for the opponent to enter, for instance an opening ... and for that you need to MAKE one, AND be sure that it is tempting, but does not put you in danger of actually getting hit.
Is training AT the time something is happening, then winding the clock backwards until you can SEE the moment coming, and then can CREATE the moment to come, the best way to train?
Or could you train the uncertainty, and the creativity in from the get go, so you are already closer to context?
Even if there are only 2 options - perhaps either left or right, or high or low, now or ++ now, as choices, surely it would be a more efficient way to learn?