implied odds

i am pretty much a newb at the game with just a 50k hands history at microstakes. so please bear with my ignorance...

i understand the concepts of pot odds and implied odds and use them in my play. i am just often finding it difficult to find a good estimate for implied odds. like with the 15x rule for raises on pocket pairs i am not altogether convinced this is always correct in practice. since a lot of the time i wont be able to get my opponents money in the pot to justify calling a 4x or 5x bb raise preflop. i believe i really have to factor in variables like loose-/tightness of opponent, relative position, no. of opponents and pair value. so maybe it should actually be called the n-times rule with n = f(LT, RP, NOP, PV). maybe even additional variables? in fact with very deep stacks the rule might not be any guidance at all since it would suggest to even call very big raises preflop which just doesnt feel right at all.
another situation is the frequent flush-draw or open-ended str8-draw at flop or turn. again i am finding it difficult to estimate implied odds. these situations vary immensely i feel. when i hit, sometimes i find it very easy to take down oppenents stack, at others i will hardly make a few bb.
so how to judge implied odds? when unsure i go with about 1 1/2x pot size for calculating implied odds. but not at all sure this is in fact a helpful guideline.
maybe a good indicator might be the strength shown by opponent up to that point, since the more strength he has shown the more likely he will be to move money in the pot when i hit. but then again assuming oppenent not to be stupid he might already have read me being on a draw and seeing a card hit the board that would have me hit would have him immediately tighten up.

any recommendations by skilled players as how to estimate implied odds in situations like these or others would be great. what criteria do you go by? again a function delivering a multiplier for the pot as a basis for doing implied odds math would be great. what would the variables in such a function be? how would these variables be weighted?

cheers

The 15 times size of bet for stacks is a general rule. If you have a read that a player will not pay you off when you hit, then you can play accordingly. However, the 15x rule does take into account times when you dont get their stack and for a beginning player it is good advice. With deep stacks, PP become even better because if there is a large raise pre, and you set up, it becomes even harder for your opponent to fold their good hand.

Implied odds for flush draws aren't usually very good. It is often very obvious when you hit and most players will not call large bets with marginal hands on 3-flush boards. Straight draws where you are using both of your hole cards generally have pretty decent implied odds especially when your straight makes the villian 2 pair. However, flush and straight draws are much more read dependent, and position because much more critical. Also, an important thing to consider is how expensive it will be to see the river if you call with a draw on the flop. I sorta rambled there so i hope it helps.

Calculating implied odds is more of an art than science. Just because you have 200BB and villain has 200BB does not mean you have implied odds to call any pf raise. In general, these are the principals I use when calculating implied odds:

1. Villains aggressiveness. The more aggressive, the better your odds are. If he pushes every flop and turn after raising pre-flop, then he is ripe for a stacking if you hit and you should make note of this tendency. This is probably the most important factor. Also watch for patterns when they have QQ-AA...do they play them uber-aggressive or not? Many players bet these differently pf...i.e. I see very few players who bet it the same. For example, many will raise pf less with QQ-AA because "they don't want to scare away their action". When I see this I call with any 2 because I know they have a monster and they will stack off if I hit. Plus they make it SO cheap, even 2 7 has implied odds.

2. Position. If you have position on the pf raiser, then you can hide behind his betting or raise him at the right times to get his stack in. At some point when you are OOP, you have to bet out and that usually kills your action. With position, he might check the turn or river to you but he still can't be sure you are betting because you are trying to steal or if you really have a hand.

3. Calling a button raiser from the blinds rarely leads to a stacking. You don't call here for implied odds, you call or raise to defend your blinds. Why? In many cases the button is raising with a marginal hand (esp when it is folded to them) and thus will not be inclined to stack off.

The bottom line is that you need to study your opponents and make notes on their tendencies and aggression. When you understand this, you can profitably call them with the proper implied odds.

thx guys for your replies
seems some of my gut feelings are being confirmed here. sklansky in "theory of poker" states stacksize of opponent as no. 1 important variable but then goes on to loose/tightness as no. 2. beyond that he remains rather vague actually. from experience i would definitely agree with djzcko that relative position is also immensely important.
if we stick to the case of pocket pairs for the moment the 15x rule would thus only take into account variable no.1 .
i know of course there is no exact way of calculating implied odds. these will always remain something of an estimate. but i do believe in quantifying estimates.
so i still have this idea of maybe modifying rules like the 15x rule towards taking into account additional variables like loose/tightness and position aswell.
a rather simple adaption for example could be:
n-times rule with
n = 15 + LT + P (LT = loose/tightness modifier, P = position modifier)
- LT positive for tight opponents, negative for loose opponents e.g.: LT = + 3 when tight opp. LT = -3 when loose opp
- P positive when oop, negative with position. again, just for the sake of the argument let P = +3 when oop and P = -3 with position.
then for different situation n would come out with values:
n=21: oop & tight opp.
n=18: oop & neutral opp.
n=15: oop & loose opp. or position & tight opp.
n=12: position & neutral opp.
n=9: position & loose opp.
the weighting of 3 for P and LT is at this point pretty arbitrary just a guess to illustrate what i am looking for. probably weighting for P and LT should not be the same.
which of the 2 is actually the more significant variable?
anybody have suggestions what good values for such weights could be?
also ideas to formulating similar rules for flush draw and openended str8draw implied odds would be great. again i think such rules should take variables like position and loose/tightness into account.

cheers