refreshingly nice discussion

We don't need half the hands to get AI with someone post flop. We just need to average an 8x call return. The extra IOs required for set mining are to make up for the times that doesn't happen and the occasional RIOs.

With 4x already in the pot pre, that just means that we need to average 4x more post flop when we hit. So in OP's scenario the cost is $25 and the pot is $125, so we need to win $100 on average post flop to make set-mining profitable. That means that with 4 Vs, we need just one of them to pay of one bet of 4/5 pot, or 2 of them to pay off a <1/2 pot bet. If you can't average at least that much when you hit (you can), you should be calling planning to bluff post (you shouldn't).

I don't like the bolded hands as much unless you are positive you can get away from top pair or know when you can continue with top pair. Also, a lot of your cards are usually in other people's hands.

Curious, how often do we get and hit a combo draw vs. how often do we hit a set? I honestly have no idea.

Short answer. With SCs, there's about a 8% chance of flopping an OESD, and about an 11% chance of flopping a FD. That means that there's about a 19% chance of flopping one or the other (additive), but only a 0.88% of flopping both at the same time (multiplicative). That excludes when you actually flop the made hand, of course, but yeah, combo-draws are way less common than sets. You have to be in a situation where you'd be happy with one draw or the other for SCs to be worth playing. This isn't such a situation, unless FE is very high post allowing super-profitable semi-bluffing, or if the game plays super passive post, allowing very cheap drawing.