When the NFL Draft comes around every year, all we do is talk about evaluations, projections, predictions and comparisons. The lead-up is exciting for all those reasons.

But the same can be said for the years following the draft -- that's when we look at a player's development and potential.

Have they performed as expected so far? Do they show signs of being a superstar, or do they look like a wasted pick?

It's not as often that we look back to compare players from the same draft class. How has one quarterback panned out versus the others in his group? Was it worth moving up in the draft to get them instead of waiting on someone else?

These are things we usually only consider in head-to-head matchups and other things like that. Well, we should do it more. Why? Because it's fun and we now have the means (thanks to advanced metrics) by which we can accurately compare one player to another.

The best way to do this is using Pro Football Reference's Approximate Value (AV), a single number representing each player's seasons, allowing us to compare across positions, seasons and eras. In order to evaluate a career, though, naturally we use Career Approximate Value (CarAV), which takes 100% of a player's best season, 95% of his second-best season, 90% of his third-best and so on in order to weight peak seasons over what we may consider "compiler" seasons.

Using CarAV, we will re-draft the quarterbacks from the 2004 NFL Draft based on the results and value of their careers up to this point. This will point out the poor picks, value picks and much more.

Here's how things played out for the seven most notable signal-callers.

Now -- starting from 7 and working our way to 1 -- let's break down each player's career so that we can understand who would go either later or earlier in the draft if we were to do things over 13 years later.