This is actually exactly backward. What the paper is trying to quantify is how many people are trying to use bitcoin as an investment, versus how many are tryi

This is actually exactly backward. What the paper is trying to quantify is how many people are trying to use bitcoin as an investment, versus how many are trying to use it as a currency or other financial tool (the whole thing about how many of the investors are underwater is a small point of the paper that Gizmodo is fixating on because of Gizmodo's preexisting hatred of crypto). What the paper finds (and we can certainly quibble with any paper) is evidence that the majority of people who bought Bitcoin in the 2015-2022 period did so in patterns that suggest that they're interested in an investment asset, not in a currency.