Personally, I'm using the TGM's recommended MDF of ~45%, wc means I'm folding ~55% (villain's actual RFE for his 3b is ~68%, but author reduces it to ~55% for the fact that villain has postflop equity). Obv, my MDF consists of flatting, 4b value, and 4b bluffs.

What portion (% wise) of your overall opening range do you defend vs 3b or you think is an optimal% defend?

Another concern of mine is when we get 3b by a known fish or tight players w/ non-optimal 3b freq (on a nitty 3b freq side) but using a standard 3b sizing (so their RFE is equal to decent players). Bc if we exploit them by overfolding, aren't they also automatically exploiting us? How should I think about this spot?

Thanks

There's been hundreds of threads on this issue, but my feeling is well known: MDF has very little utility pre-flop or even on the flop.
You should continue with certain hands vs a 3-bet simply because calling (or 4-betting) is more profitable than folding.
Your aim isn't to prevent villain making money with air (because most villains aren't 3-betting total air) or making him "indifferent" with bluffs (a good calling range will automatically reduce the profitability of his weakest hands to <1bb). Your aim is to maximize your own EV, given the range you think villain is playing. He'll win a bigger share of the pot (or win the pot more often), because he has a stronger range, but many of your calling hands will win often enough.
If he only 3-bets KK+, then obviously you're not gonna do well flatting AK or QQ (never mind the weaker stuff), but if he's 3-betting 10-15% of hands, then loads of pairs, suited Broadways and suited connectors will have the right price and equity to call. After all, if villain makes a pot-sized 3-bet, you only need to win a third of the time for calling to be better than folding. There are a LOT of hands that will win at least a third of time vs a typical 3-bet range.
In short, I wouldn't do too much thinking about indifference or frequencies. Just consider your equity vs villain's range, and whether you'll be able to realize it.

Preflop MDF would only apply if all remaining players are playing 4-bet or fold, and it applies to all players collectively. Most strategies recommend having a flatting range, even OOP, so I’d just not think about MDF much preflop.

ok that makes a lot of sense, thanks guys ; )

edit: what's the common foldto3b% for a solid reg?

It varies a lot, especially from one stake level to another. At 2NLz, there are decent winners that fold to 3-bets more than 60% of the time. Higher up, the number is probably lower than 50% for most winners.
FWIW, I had a quick look at Snowie's 100NL BTN v SB 3b strat last night, and it has the BTN continuing (with a call or a 4-bet) around 57% of the time. (i.e. it only folds 43% when faced by a pot-sized 3-bet in that particular positional battle. It continues a fair bit more often than the basic MDF number when in position. When it's UTG v MP3b, the number is completely different.)

What is that UTG number?

UTG vs an MP 3-bet?
Snowie opens around 15.2% UTG at 100NL (about 202 combos), but only continues (inc 4-bets) with about 76 combos (77+,ATs+,AKo) vs a 3-bet by MP. So it folds to 3-bets in that instance about 126/202 = 62%. Recall that MP's 3-betting range vs UTG is exceptionally strong, because he's 'saying' that not only can he beat UTG, he's also not too worried about 4 other players waking up with a better hand. Ergo, UTG has to give MP more credit for a hand, and can continue only with a very strong range. I think the folding frequency is a fair bit lower in UTG v BTN 3-bet, because the latter's 3b range is weaker/wider than MP's would be.
Being OOP means it continues less often than it would when IP vs a 3-bet. (Because it's easier to realize equity in position).