"True point guard." What do you think when you read that phrase? The immediate connotations to the college basketball fan (or, for that matter, the NBA partisan) are practically hard-wired: A true point is a player that "sets up his teammates," "takes care of the ball," is the "coach on the floor" or the "quarterback of his team."

The college game is so often dictated by guard play -- it is an accepted theorem (at least) that good guard play leads to deep tournament runs -- that college coaches, fans and analysts tend to get a little too caught up in defining a point guard on purely restrictive terms. Does he do this? He's a pure point guard. If not, he's not.

At least, that's the takeaway from Basketball Prospectus writer Drew Cannon's interesting Tuesday ditty about point guards, in which Cannon explains why evaluating point guards is -- or at least should be -- the least zero-sum of any position on the court. Wing players and centers can, with minimal exceptions, be dropped into particular lineups without much of a perceived drop-off in their abilities. Point guards, on the other hand, are incredibly context-dependent, as Cannon shows in his breakdown of UNC point guard Kendall Marshall:

Here's the short version of what I'm talking about. Trade Kendall Marshall for Maryland's Terrell Stoglin or Pittsburgh's Ashton Gibbs. In either trade, both teams would be worse. Maryland and Pitt need their point guards to score because their other players can't create their own shots. Without Stoglin, Maryland would have tons of trouble scoring, even with Marshall zipping passes around -- their top scoring option would probably be freshman Nick Faust, who's been a fine player thus far but not capable of carrying an ACC offense. At Pitt, Nasir Robinson and Tray Woodall have been extraordinarily efficient thus far, but it's unlikely that would still be true if defenses could focus on them rather than Gibbs. And on a team like North Carolina, Marshall is the perfect point guard, just putting Harrison Barnes, John Henson, and Tyler Zeller in the best position to succeed. A guard whose value is more predicated on getting his own shot would destroy that balance.

Cannon runs through a few more examples -- Ohio State's Aaron Craft, Duke's Seth Curry, Wisconsin's Jordan Taylor. In each case (with the possible exception of Taylor, whose longtime scoring efficiency and low turnover rate make him a good candidate to succeed just about anywhere) each guard's performance is in many ways dictated by the personnel that surrounds him. Craft couldn't focus on his amazing perimeter defense and canny, opportunistic offensive play if the team around him wasn't so loaded with talent. Curry couldn't set up teammates the way Marshall does; Duke's offense would then have to run through its wing players and forwards, rather than Curry's perimeter scoring skill.

In adding to Cannon's point, John Gasaway writes that it's time to banish the term "true point guard" from the lexicon in much the same way intelligent basketball observers have banished everyone's favorite, and oft-cited but horribly unrevealing, "rebounding margin." John's key point, I think, is this: "Point guards come in all shapes and sizes."

That seems simplistic, but it's true. A point guard can be, say, Evan Turner, who was really a 6-foot-7 wingman doing his best impression (one good enough to win him the Naismith, no less) of a point guard. A point guard can be whatever its team needs it to be. If you merely need a very good setup man, then you can skew more toward the traditional description of a "true" point guard. If you need scoring from your point of attack, you can go in a different direction. Or, if the glove fits, you can play point guard by committee.

Any and all of these styles can work, depending on the team and the coach and the particular player. What matters is not what the player looks like or how tall he is or the raves he draws for "quarterbacking" his team, (though that can be just as important a function as any other, if more difficult to pinpoint). What matters is how this player -- or players' -- perform. Does he keep turnovers low? Does he make the lineup around them better? Does he accentuate his team's strengths? Does he minimize its weaknesses? Does he fit?

The search for a "true point guard" is inherently frustrating. It asks the searchers to compare practical reality with a vague ideal. Most players don't fit inside such peremptory classifications. Each is different, and each has the potential to be effective in variable ways on variable teams with variable traits each their own.

Is Marshall a "true point guard?" Is he the truest of them all? It doesn't matter. He's a fantastic fit at UNC, and that's all that does.