One interesting approach is to check the average ETF flow. Since many institutions are banking on BTC and are the ones with the money to actually move the market, I think this is an important case to add.
Of course it's always a good idea to check "real" market movements like ETF flows. However, my main problem with that approach is that the you generally see what happened "after the fact".
If you see a day with negative ETF flows, and the price tanked that same day, then you can conclude that ETFs contributed to the price weakness. But you don't see what ETF holders will do tomorrow.
Another interesting case: A day with negative ETF flows, but the price goes up. This could mean that retailers are buying and institutionals are selling. So you could interpret that more weak hands enter the market.
However it's not really clear to me if ETF buyers are really stronger hands than CEX buyers. I mean to remember that in the 80k dip last year, institutionals were one of the main drivers. And CEX buyers not necessarily means "retailers" - whales which are not restricted to ETFs can also be institutional whales.
It would be interesting if there is a study about that. Did ETFs "overreact" in the case of strong panic crashes and FOMO movements? Or is it (as it is commoningly expected) vice versa, i.e. CEX users are more likely to panic selling and FOMO-buying?