I don't think anyone claimed that there was 1:1 relationship between the volume of this (or any other) fraud and a change in bitcoin's market cap. As far as bitcoin is concerned, the particular details of the fraud are entirely irrelevant.
I don't think anyone was as exactly specific as that, but in reading discussions of MMM's possible contribution, the description was almost universally (on Reddit and HN) comparing, explicitly or implicitly, the previous MMM's total volume to the Bitcoin market cap, showing that pretty much everyone seems to have been intuitively confusing stocks with flows - the unspoken argument seems to be 'the old MMM's total was about as large Bitcoin's market cap is now; the new MMM is also quite large and may be a large fraction of Bitcoin's market cap; therefore, the recent near-doubling of Bitcoin's market cap may be mostly or all due to new MMM's large total'.
But the float is much smaller than the total volume of the fraud. If MMM is doing $10b of fraud a year, then it's not boosting Bitcoin's marketcap by $10b as one might naively guess ('$10b fraud drives Bitcoin price growth!' scream the headlines), it's boosting the price by much much less, by just the float (which could be more like $27m).