One of the things I find, is that when students start countering too soon, everyone focuses on technique X instead of technique A. This means, A is done poorly.... which means that X seems to be learned much better.... However, then they go up against someone who focused on technique A, and does it correctly.... this being the first time the student has tried X on a proper A.... means that A works and X fails.
The trick is to focus on A, enough so that the student can use A and rely on A to work.... and then you can learn X, being careful not to sacrifice A. The main trick here is for each student to focus on their own technique, doing it correctly.... without focusing on "winning." "Winning" makes technique worse and sloppy. "Training" is for improving technique. You need a balance. For each student, that balance will be different.
One way I like to approach this is with adjustable difficulty ecological games.
Here's an example.
Suppose I'm teaching a simple single leg takedown. There are a ton of details and variations and what-ifs that I could throw at my students when I demonstrate the technique.
Problem: there's a definite limit to how much of that detail students will be able to absorb at once.
Problem: takedowns and throws are probably the most technically demanding aspect of martial arts, so they will need a lot of those details to pull off the move against anyone who's any good in sparring.
Problem: If I have them just drill the technique for weeks before using it in sparring, they won't have the feel for how everything works in context, they'll flub various details they've learned, and they'll lose confidence in trying the move.
Problem: If I throw them into trying the takedown in sparring when they've only had a barebones introduction, then they won't have learned essential details, they'll have low success, and they'll lose confidence in trying the move.
Problem: In either of the two scenarios above, if they do have success executing the technique against a sparring partner who doesn't know how to defend properly, they'll likely start burning in bad habits that will hurt them later against more skilled opponents.
One possible solution:
Step one is to demonstrate a basic entry and finish, giving them just enough detail so they can drill the technique with a cooperative partner without tripping over their own feet.
Step two is the ecological games.
I like to start one student out having already picked up their partner's leg. This allows them to ignore (for now) the difficult tasks of setting up and entering to get the leg. Now they start live. The attacking partner's job is to finish the takedown using the leg that they are controlling. The defending partner's job is to avoid the takedown and get their leg free. This setup should give a clear advantage to the attacker, but beginners may not be able to fully realize that advantage.
I'll watch for a few rounds and see who is winning more - the attackers or the defenders. If the attackers are winning more, then I'll offer tips to the defenders on how to counter. If the defenders are escaping more, I'll offer tips and corrections to the attackers. This way each job gets progressively harder as both the attacker and defender get better at their role.
This also has the advantage that I don't have to throw every detail and nuance at the students from the get go. I can watch and see which are the most common problems which come up when trying to execute the move under pressure and fix those first.
Once students have done this game enough times, it should settle into an equilibrium where the attacker will consistently win more often, since they start with an advantage. At this point they will have the confidence that they can have success if they can just secure the leg, so I can switch to other games aimed at progressively improving their ability to do the necessary setup and entry to get hold of the leg. (While also progressively improving the ability of their partners to avoid having their legs grabbed.)
