I know right. After all this time. Well you know what they say in crypto that if the roi is way shorter than the competition then its too good to be true

The "roi" is set by what people are willing to pay for the MHS (was KHS) being sold. That has nothing to do with scrypt.cc being a scam or not. The prices are very low per MHS because the yield is very low. It was set to 20% after the hack and never restored. Prices just adjusted to the lower return. It is a free market. That is all.
If the yield goes down the ROI point should be further away, no?
That is true for hashing power already held. Likewise when the yield goes up the break even point can be hit sooner. Generally the yield is down with mining. However the price of the MHS is controlled by the market, not by scrypt.cc.
Agreed. What I'm struggling with is - assuming a free market - why has yield risen to the point where ROI is "way shorter than the competition / too good to be true"? My assumption would be that there are external factors - that the market (scrypt.cc) isn't as free as it appears. To give an analogy, the last time I was almost tempted by something too good to be true was arbitrage between exchanges a couple of years back. At the time, it was possible to deposit fiat in a Japanese exchange, BTC in a Chinese exchange, and then buy BTC in the former and sell BTC in the latter for a (theoretical) 5-7% profit. I assessed the risk of the
Chinese exchange to be too high - I was wrong, though it did save me in the end. It turned out that the Japanese exchange - MtGox, obviously - had serious problems, and that arbitrage opportunities (like Gox withdrawals) were an illusion.
Edit: clarified that arbitrage "profits" were theoretical, obviously because withdrawals from MtGox were n't possible.