In the hand you posted there's t675 in the pot going to the flop and villain comes out with a bet of 200. That's just above 4:1 odds, which is what you're looking for as far as odds to draw to a flush. Don't forget to factor in the fact that you have overcards and might be able to count an additional 6 outs if a 9 or Q falls.

What drives my decision in these situations is the table action. If you are in the pot with a few people, like you are here, you need to decide how to get value out of your hand if you hit. That's when you decide between calling and hoping for overcalls, raising hoping to get more money into a pot that you might win if your flush hits or raising to eliminate people from the hand. It's more about why you make these moves in each situation and not a set percentage that I would assign to each move. Consider the possibility that villain bet out 200 on the flop and, instead of folding, you had 2 callers around and the action is now on you. In this spot you might opt to put in a raise and hope that everyone calls around again. If you hit your flush you're taking down a monster pot.

Your position also has a lot to do with how you play it. Checking the action on this flop into villain and letting him bet out might be best for you since you might get the callers between you to commit more chips not buying into his PFR/CB line holding a small-medium PP. This would give you great odds to make a call and you can even consider popping the raise to inflate the pot even more. I especially like this play on a draw heavy type board like this as you could see a dream card like the 2,3,7 or 8 of hearts fall on the turn completing someone elses straight while you are holding the flush that beats them.

In the situation you posted I would probably just call with proper odds to draw to my flush with overcards and re-evaluate on the turn based on the card that falls and villains action. The raise for value isn't as effective here as you are now HU with the villain after everyone else folded. If you miss your draw on the turn and decide to see a river card you could have to call off a large portion of your stack. You want the extra players in the pot to make a raise for value the right play here. As this hand played out I think a call and re-evaluate on the turn is the best move. That way you can keep the pot small and decide on the upcoming streets if you want to commit more of your chips.

Another option that you have when you are in position, which you aren't in this hand, is to raise the flop hoping that villain will check to you on the turn and you can deicde there to either take the free card or bet out. Because you are first to act this line won't necessarily work in this particular situation. If you raise on the flop and then check to villain it's almost a dead giveaway that you are weak since very few players try to go for 2 check raises in a single hand. Yet this is still another option to consider when you are on a draw and want to see 2 cards cheaply. Your flop raise will often be less than a turn bet would amount to and you can sometimes see both the turn and river for a lower total price by using this tactic.

I love HOH and everything that Harrington teaches in his books but formulas like these that he lays out make people too dependent on these fomulas and less dependant on reading players. Psychology is an important aspect of poker and goes hand in hand with the mathematics that drives our decisions. Different situations and opponents should be a factor in deciding which moves to make just as much as the probabilities. I'm appreciative of his examples using his watch to determine frequencies, but my watch only tells time and isn't affected by a players tightness/aggressiveness (or lack thereof). If I were to rely solely on my watch to make my decisions for me I think I would be in serious trouble.