Material implication -- the "→" operator of propositional calculus -- is interesting because of how it differs from our "natural" notions of how implication should work. For example, it is easy to see that a false hypothesis P implies anything. Mathematical logicians don't care about these matters, but ("real") logicians do, since they want a theory explaining how people think.

This is unfortunate, as it leads to unintuitive consequences. First, consider a false conclusion: "if the moon is made of blue cheese, then I am a porcupine". Here I'm generating a fairly logical sequence, one which is even found in everyday speech. But consider the converse "if the moon is made of blue cheese, then Richard Nixon was president of the U.S.A.". Here I am deducing a correct (and true) conclusion from a false hypothesis. Since I'm not likely to be able to establish the hypothesis, it still doesn't look too bad.

But it means that material implication doesn't capture what we think of as implication, i.e. an ordered "logical" process. Because no amount of argument will lead you from "the moon is made of blue cheese" to "X was the president of the U.S.A."! So this implication means very little -- just that you are unable to establish "the moon is made of blue cheese".

Unfortunately, if you're looking for an implication that is a logical operator, there's no alternative. A logical operator depends only on the truth values of its arguments, and it's easy to find examples of things we'd like to call "implication" satisfying each of the lines in the truth table for P→Q:

   P   Q P→Q
   T   T   T
   T   F   F
   F   T   T
   F   F   T

To generate a "real" implication, we'd want to add stuff requiring some (semantic? syntactic??) connection between P and Q. Nobody has any ideas how to do this, so we're stuck with MI. And even if you could do it, you'd lose the ability of formal logic to deal with deduction without considering the "character" of statements: you'd no longer be able to replace statements with letters, and letters with statements, knowing this does not affect truth values.