The problem with fair use is that it is very often in practice decided by "judge gets a gut feeling of whether or not it's fair, and uses the factor to buttress

The problem with fair use is that it is very often in practice decided by "judge gets a gut feeling of whether or not it's fair, and uses the factor to buttress their opinion." It's not an objective reading of the factors in practice--read the dueling analyses in SCOTUS's opinion Google v Oracle. One analysis finds all four factors favor fair use, whereas the other finds that the three factors that really matter (as who really cares about factor 2) all disfavor fair use. The latter was the dissenting opinion, but quite frankly, it's the kind of analysis you're likely to see a trial court judge actually use. Let's run through that kind of analysis here. Factor 2 disfavors fair use, because the original work is clearly "creative"; what more needs to be said? Factor 4 is the marketplace factor, and clearly, people who are watching the game aren't going to be playing the game themselves, so it's having a depressive market effect--fair use disfavored. Factor 1 is again disfavoring of fair use, because you're reusing market value copyrighted content in a commercial context. It's not transformative because it's not commenting on the original work [1]. Factor 3 again disfavors fair use, because you're reproducing essentially the entire video game in the playthrough. Conclusion: all 4 factors disfavor fair use, so how can it be fair? I personally disagree with this analysis, and I even think that it misapplies SCOTUS-level precedent (precedent at the circuit court level is even more incoherent), but this kind of analysis is more likely to be what I expect to see, especially because (as in Thomas's minority opinion) the monetary and commercial implications of the use tend to dominate the thinking. [1] My understanding of practical US fair use analysis is that "transformative" is viewed far, far narrower by the courts than by lay people.