My experience with that demographic is that they're pretty ready to tell you about it.

I guess but wouldn't that leave some group of centrists who "somewhat" support martial law while also thinking the police station arson was at least "partially" justified? That doesn't seem reasonable at all unless the polls are just inaccurate. In any case the results are somewhat surprising.

NYC cops arresting an essential delivery driver who is exempt from curfew laws.

The reasons for their not fully supporting martial law are probably the reasons why they think that burning down a police station is partially justified. What's most likely is that these are different polls surveying different people with small enough sample sizes to have seemingly conflicting results. The dissonance that you're pointing out would be more compelling if it was within a single poll using the same subjects.

It's called dumb moderates who don't know what they believe but they know they don't like bad things.

Obviously I agree it would be easier to judge of it was the same group of people, but if the polls can't be assumed to roughly reflect the general population then isn't that the same as just saying they're inaccurate/poorly done?

I honestly can't speak on the methodology or what the average sample size is for a good poll on this sort of thing. The two in question had 807 and 1,624 participants.

dumb moderates who want to burn down police stations AND bring out the troops? We sure got some interesting moderates.

Nah.

Isn't the entire purpose supposed to be gauging the general sentiment? Maybe if they were targeted towards specific different groups I could see your point but I think both are meant to be representative of Americans in general.

What about the police station probably deserved to get torched because it was where that cop worked but the troops were needed because unrelated stores were being looted?