Game theory (inflating the pot when behind)

That way you will have "equity" to go to showdown! But, that way any bad play will give you "equity", because, prior to your turn re-raise you had no equity whatsoever! Maybe, you still don't have it unless you invested majority of your chips on TWO OUTER!!!
My "two cents" on this play: It's a good play on some occasions and you NEVER want a showdown! But, I'm no pro!
Also, that's almost heads-up game, so, that play has more sense then on FR-game!

I think we need a specific scenario to consider this.

Sometimes raising when behind can improve your equity where the player will tend to fold a lot of equity AND the equity they fold improves our winning chances.

Usually when a player folds the majority of their equity goes to the hand which is favored to win.

In your example (briefly explained) we have two outs to the virtual nuts. In such a situation it sounds like we could play against 1, 2 or more players and hitting our two outs gives us the win. Its possible something else is at play in your scenario but if we still have two outs when we force players out then we do not have an incentive to raise against our equity and force them out.

The same as capping preflop with 67 in a 5 way pot. I'll give a good situation when i find a hand ;-)

I don't know exactly what you mean here, but I guess that you are trying to "trade mistakes". That is, you make a small mistake (you raise with a weak hand) on one street in order for your opponents to make bigger mistakes later on (pay you off in full when you hit)?

Qualitatively it does make sense. Better made hands can fold. You don't get drawn out on and you win a big pot when you win.

Does it make quantitative sense? I.e. is it a better play in the long run than just folding?

You need an actual example to analyze this. I suspect that the opportunities are rare to really make this play profitable. In 7-stud it's easier. There the play is standard (e.g. pair + AK kicker vs probable QQ) in many situations.

There is a related play. You have the best hand, but don't raise on one particular street because it's advntageous to wait until later streets with the aggression. Same thing here. You make a small mistake by not betting (your opponents would call correctly), the you bet/raise on the next street and your opponents now incorrectly calls or folds, making a bigger mistake that you did the previous street. [Not talking slowplaying here.]

As should be obvious, this is way too complicated to analyze quantitatively without a very specific example. In the latter case (excessive passiveness), you need a table full of calling stations. For the first case (excessive aggression) you need, well, a lot of things to go well.

As for the 76 example, it is a clear loser. Don't do it. If you do it with 55 against 5 opponents, it is probably a small winner.

/Johan =

Actually, the cap with 55 is useless too. The thing is that you might want to raise in a MW limped pot if you fear that your opponents will fold too easily if you hit your set. That double bet migh keep them in the pot paying you off all the way. But, most fish will call the flop and the turn anyway, making the play is pointless.

[Read Sklansky if you want rationale behind this.]

/Johan =