The success of every single NFL play is judged based on yardage gained toward both a touchdown and a first down. Then each play gets compared to the NFL average on similar plays, based on down, distance, and other variables. Players are judged not based on how many yards they get, but on how important those yards are in the context of the game. Once we have enough data for the season, beginning in Week 4, we also adjust for the quality of the opposing defense. Each player's performance is then translated into an approximate number of actual yards that such success (or failure) is worth when compared to a generic bench scrub (also called a "replacement player," or, by his parents, "Billy Joe"). This statistic is called Defense-Adjusted Yards Above Replacement (DYAR).
Need an even quicker and easier way to see how good a player was? Then use another stat we call Effective Yards (EYds). Effective Yards take the player's performance, adjusted for situation and opponent, and puts it on the exact same scale as standard yardage. If a player has more Effective Yards than Yards, he was better than standard stats make him look. If he has fewer Effective Yards, he was worse than standard stats make him look.
We rank players based on DYAR rather than Effective Yards because Effective Yards give too much weight to playing time and not enough to quality. Still, Effective Yards is a good way to figure out if that 300-yard game really was as effective as we usually think a 300-yard game is.
Among the advantages of this system:
• Gives value for first downs, which are never discussed by conventional statistics even though they are the most important part of sustaining drives.
• Does not punish players on offenses that are always in bad field position because of poor defense, nor does it punish quarterbacks who are always stuck in third-and-long because of a poor running game.
• Players receive bonuses when they play well against good defenses, and they don't get rated as world-beaters when they shred the Falcons.
• Quarterbacks get credit if they can run, and get penalized for sacks if they can't recognize or avoid a pass rush. Running backs get credit if they gain yards out of the backfield.
• Judges players not only on receptions but on all passes, because the blame for incomplete passes should fall on both the quarterback and the receiver.
• Punishes players not only for turnovers, but for all fumbles, because research shows that jumping on a loose ball is entirely random. (We all know that the officials just award the ball to whichever team is pointing more strenuously.) Different fumbles have different penalties depending on how often that type of fumble is recovered by the defense.
• Five-yard gain on third-and-10? Worthless!
The system cannot yet split the responsibility for a play between 11 players on the field, so be aware that a player's rating will be influenced by the talents of his teammates and the offensive system in which he plays. (This is especially true for a quarterback and his receivers, and for a running back and his offensive line.)
An even longer explanation of these numbers can be found here.