This was written by Greg Cossell on Yahoo! but is one of many articles stating the evolution of the safety position in the last 5-10 years. I hope this helps clear the confusion some people have in a safety. For further reading you may want to read his "the safety switch part II".
I have always thought watching the progression of NFL offense over the last decade, both with multiple wide receiver spread concepts and two tight end personnel, that safety was an increasingly important position. It’s hard to say whether NFL teams agree or not. There have been seven safeties selected in the top 15 since 2000: Roy Williams, Sean Taylor, Donte Whitner, Laron Landry, Eric Berry, Earl Thomas, and Mark Barron. (Rolle was the eighth pick in the 2005 draft but he came out of Miami as a corner). Speculation is both fun and futile, but it’s certainly interesting to ponder whether any of those players would be chosen that high in the 2013 draft.
The larger question is, what traits are now needed to play safety at a high level in the evolving NFL. There was a time when there was a clear delineation between strong safety and free safety, and the requisite skill sets that each position demanded. Strong safeties, in many ways, were glorified linebackers; bigger, more physical, able to play the run and match up man-to-man on the slower, less athletic tight ends that were then prevalent. Free safeties were better athletes, faster with more range, capable of roaming deep from sideline to sideline. All this derived from an earlier offensive era in which two backs, one tight end and two wide receivers was the league-wide personnel template.
It seems to me, and others may disagree, the safety position has been exceedingly slow to change and adjust as offensive concepts have expanded over time. In fact, I think you can make the easy argument that safety generally has not been seen as a premium position in the NFL. For years, the feeling among personnel people in the league was that you could get quality safeties beyond the first round. As is always the case with any position, there are numerous examples that make that point. John Lynch was a third-round pick, while Darren Woodson, Darren Sharper and Brian Dawkins were all second-round selections. Again, it brings us back to the same argument that infuses all discussion at this time of year: draft value. It often points to the different world view of personnel executives and scouts on the one hand, and coaches on the other. Value at the end of April has absolutely no meaning to coaches game-planning on a weekly basis in the fall and winter, especially when they don’t have the players to match up effectively, and therefore must limit their schemes to camouflage and compensate for weaknesses and limitations.
My sense from extensive tape study and many conversations is that defensive coaches would like to have interchangeable safeties, players with strong and free safety attributes, and therefore multi-dimensional in their scheme utilization. Again, what does it come back to, and what is the objective of defensive coordinators? Having the resources to solve problems presented by the multiplicity of offenses, both from a tactical and personnel perspective. If you cannot match up to an athletic receiving tight end in man coverage, you don’t feel real comfortable playing man. Your choices are reduced. The offense knows that, and can more easily exploit your defense. That’s the way it works on NFL Sundays. That’s how games are won and lost. That is as clear as the sun in the summer sky when you meticulously study tape.