Putting players into a game does not guarantee learning. A game can expose a problem, create repetition and give players opportunities to make decisions. It can also produce loads of activity without getting close to the thing you were trying to coach.
The game does not know your intention. It does not understand the problem you are trying to solve or whether the rules are producing the behaviour you hoped for. That is still our job.
I use plenty of games in training, but I try to start with the problem rather than the activity. What are we actually going after? What picture do we want the players to recognise? What decision or behaviour are we hoping to see?
Only then can we decide what the game should look like.
The space, numbers, scoring system and constraints all shape what happens. If the attack has too much space, the decision may be too easy. If the defence is overloaded, the players may never see the picture we want. If a rule is too complicated, they can spend more time remembering the game than playing rugby.
Once the game starts, we need to observe it properly. Are the players getting enough opportunities to solve the problem? Is one player dominating? Is the behaviour appearing for the right reason, or are they simply gaming the rule?
Sometimes the best response is a question or a quick bit of feedback. Sometimes it is changing the space, altering the numbers or removing a constraint. Sometimes the game is working and the coach needs to leave it alone.
The same game will not work in the same way with every group. The players, their experience, the available space, the time in the session and the problem you are solving all change what you need from it.
A folder full of games is useful. Knowing why you are using one, what you are looking for and when to adapt it is more useful.