There's a common folk belief held by nearly the entirety of the college football world, and it's one I think is absolutely wrong.
The laymen's view of player progression is something like a linear graph, i.e. (the first result from a Google image search):
I have no doubt that the aggregate summation of every I-A player's career looks something like this. However, there's a really common fallacy regarding probability that's very easy to fall victim to: just because the probability of an event happening is likely (say it's the most likely outcome, or even has a greater than 50% chance of occurring), does not mean that the particular event necessarily must occur. Think about when a high school recruit says something like "there's a 60 to 70 percent chance that I sign with East Stroudsburg State". Lo and behold, come signing day, that player ends up going to Slippery Rock. Maybe things didn't work out with his first choice, or maybe the 30% chance that he did not end up there goes to fruition.
Ok, let's back up for a minute - just what the heck am I talking about in referring to "player progression"? It's, roughly, the belief that players come on campus at X level of skill, a certain percentage of their ultimate potential down the line. Through practice, maturity, learning the playbook, and physical training, the belief is that players will improve to some extent each year.
To reiterate, I think if you looked, cumulatively, at every college football player, their progression could be plotted linearly. The problem with looking at any narrow subsection, such as, say, the 2008 Rutgers Scarlet Knights, is that not only will a linear progression not really hold, in many cases it will have little to nothing to do with each player's career.
Some examples can help illustrate this point. Let's take a hypothetical player, Terrell Hairston. Terrell starts his career off on the banks like gangbusters, looking like a potential future star at both the college and pro levels. His sophomore year, he battles injuries and inconsistency, and he ends up leaving campus before all of his amateur eligibility is exceeded. Maybe Terrell just ran into a string of bad luck. Perhaps he was very polished coming out of high school, and did not have much room for growth on a college level. The opposite of this scenario can definitely happen too. How many times has a heralded player shown up on campus and played poorly for years, before suddenly everything starts to click?
Player progression is difficult to quantify. We have rough ideas of how well each skill position player does, although you obviously have to factor in opportunity, quality of competition faced, etc... There's still a great deal of subjective analysis required, even more such so when evaluating other positions. As a result, it may often be the case that two people who follow the same team in the same season may have radically differing opinions of what occurred. Even if they both have sharp minds and charitable hearts, there are bound to be some areas of disagreement. That's what makes any analysis of player progression extraordinarily difficult.
Is there a point to all of this? Aha, that's what I was getting to. Think back to Greg Schiano's recent comments about overestimating this team's level of talent entering this year. Entering the 2007 season, I thought RU's roster was, on paper, far better than it was during the 2006 season. Entering 2008, I had similar thoughts regarding this year's prospects vis-a-vis 2007. And you know what? I'm not going to back down from that sentiment at all. On paper, 2008 should have been a better year than 2007 and 2006.
Emphasis on the on paper qualifier. There's no way to project with absolute certainty whether a player will continue to improve, or not significantly regress. All anyone can really do is make educated guesses. Entering 2008, I was fairly certain that Kenny Britt would continue to be an absolute monster, and he has lived up to the hype for the most part. I was hopeful that Mike Teel's struggles last season were mostly attributable to his thumb injury, and unfortunately, that belief ended up being catastrophically wrong. In fact, he was far, far better in 2007. Perhaps one could have expected more inconsistency, but who could really expect him and Tiquan Underwood to completely fall off a cliff when both dazzled in fall practice?
Is prognostication an art or a science? To an extent, both, even though that's a waffling cop-out of an answer. What I try to do is to make decisions using a reliable, evidence-based process with predictive value, one that's under constant internal scrutiny, and subject to numerous and frequent tinkering and tweaking. Every word may not be true, but they are all earnest.