I think that most people would agree that the vast majority of blockbusters don't have much artistic value, and that you are more likely to find it in lesser-known films. Despite this, blockbusters are immensely successful. Why?

More generally, there are a lot of domains for which you have well-known, popular, and not really outstanding solutions, and lesser-known, potentially outstanding solutions. (Think going to McDonalds or Starbucks vs going to that original little restaurant next street.)

There are other contexts in which the most popular solution is worse than alternatives and where this fact can be accounted for by inertia and migration costs. Windows vs Linux, Qwerty vs dvorak, and so on. But this doesn't work here, because there are no migration costs.

Another explanation is people's unwillingness to take risks. Original movies can be outstanding, but they can also be horrible, whereas blockbusters are usually neither good nor bad. This is probably a factor which accounts for part of the phenomenon.

However, I would like to propose an additional, different explanation, which works for any choices which are made by groups of people, and which is totally independant from the relative merits of the different solutions. My point is that some solutions are sufficiently popular to be canonical in the sense that choosing them isn't even perceived as a choice anymore, whereas choosing something else is perceived as a deliberate choice. From this, it follows that if you decide to go see a canonical solution, it can be either good or bad, but if it is bad, you won't come out as having made the wrong choice: the film was bad, period. Whereas if you choose something original, it can be either good or bad, and if it is bad, you are sure to carry the responsibility of having made the wrong choice.

To summarize: popular choices stay popular because people are unwilling to take the risk of choosing something else and being wrong. (This is not the same thing as being unwilling to take the risk of being disappointed.)

This also explains the fact that TV is still popular even though being unable to choose precisely what you will watch and when you will watch it really sucks. Choosing something else to watch requires you to make a choice, expose your tastes, and take risks, whereas no one will blame you if you chose the canonical solution of "watching TV" no matter how bad it turns out to be.

A more cynical way of seing things is the following: canonical choices, no matter their quality, give people something to comment on and to talk about. It is socially consensual to go to the canonical solution and criticize it, whereas it is risky to take the initiative to pick something original. In fact, you can comment on the canonical choices more freely precisely because they are canonical: saying that you found some given blockbuster boring won't offend anyone because there was no real choice involved in deciding to go and see it, whereas saying that you didn't enjoy something which someone really chose to see won't really be nice to that person (because he either finds it interesting and wanted to share, or thought that it could be interesting and turned out to be wrong).