I am fully convinced that you are not arguing in good faith.
Finding a few counterexamples on a gradient doesn't mean the gradient doesn't exist. Finding a few counterexamples means that strikes don't always work, but they certainly work more often than solo negotiations.
At best you are doing this thing that smart but narrowly focused people do where you try to break it down to a binary. That might be possible, but if this most charitable interpretation is correct your binary is "Are unions perfect" which they clearly aren't. Instead your binary should be "Are union usually better than solitary negotiations?" and that the answer to that is clearly yes.
I also doubt this is your actual issue, because you have introduced data and brought in hypothetical mom and pop stores that are clearly unrelated. Because of abject non-sense like this:
> You can direct your labour resources towards the union, but that comes at the cost of not being able to direct it to your paid job.
We are talking about strikes. You know and have demonstrated a knowledge that withholding labor is the power a laborer has, not directing it to a union. And in some situations (like the Kellogg's Strike over the pandemic) the union pays people on strike, this is a problem with many solutions and people get more options as they pool effort and resources.
You are ignoring repeated explanations and willfully choosing to answer the wrong questions. You are not engaging with this topically honestly.
> but they certainly work more often than solo negotiations.
Generally, least as long as the members are moderately rich and have the richest, and therefore most powerful, entity – the government – willing to hold their hand. But how successful do you think labour unions would be if the government eliminated all laws that protect labour unions? Let's be real: Most of them wouldn't stand a chance. A lot of them aren't rich enough to exert the needed power on their own.
> Instead your binary should be "Are union usually better than solitary negotiations?"
It seems the problem here is that you're doing that weird programmer thing where you think everything is binary. The first comment described four possible states on the blatant surface. It cannot be distilled down to a binary state. To try and force it into to a binary state loses everything that is being talked about. The quantization you are trying to apply is invalid.
With respect, no wonder you're so confused about the subject. Hopefully recognition of your error can get you back on track.
> the union pays people on strike
Assuming we're still talking about a union of poor people, and not going off on wild tangents, how? Where has this money magically appeared from? Sure, rich people can fund a union and pay out dividends to workers on strike, but then you're talking about rich people. We recognized from the very beginning that unions are beneficial to the rich.
Again, where do you think these poor people are magically finding all this money out of thin air? You keep going there, but never seem to be able to answer to it. Let me be clear: Poor people, not rich people.