Newz Forum: FOOTBALL: Rating the efficiency of passer efficiency rating

Perhaps the NFL's original X-Man, Patriot great Steve Grogan stood alone as the outstanding running quarterback of his era. Grogan's scrambling ability was nothing short of a dominant mutation in a gene pool of drop-back passers. In fact, his single-season record of twelve rushing touchdowns by a quarterback, set nearly three decades ago in 1976, still stands today.

But Grogan's record isn't safe. The Ice Age of the NFL is upon us, and in accordance with evolution, only the strong will survive. The glaciation of T-Rex Grossman and Dinsoaur Drew Bledsoe - just to name two - is all but inevitable.

Steve McNair. Michael Vick. Daunte Culpepper. Donovan McNabb. Quite simply, the quarterbacking elite of modern professional football and the future of the game. Their ability to evade defenders doesn't just help them find open receivers - it enables them to run the ball effectively.

The widely accepted passer rating formula manipulates percentages of success to define "effective" quarterbacking in terms of efficiency. For example, the feat of throwing for 300 yards and three touchdowns in a single game increasingly loses merit with every attempt beyond 30. A laborious analysis of the implications of statistics yielded can go on forever, but the basic gist of it is "the higher, the better."

Contrary to popular belief, passer rating is remarkably easy to compute. (1) Yards per attempt, (2) completions per attempt, (3) touchdowns per attempt and (4) interceptions per attempt are manipulated to yield the following formula:

Rating = [25 + 10(%Completions) + 40*(%TD) - 50(%INT) + 50 (YD/ATT)]/12

As the quarterback prototype evolves, so too must the statistic that quantifies his effectiveness. Although the current passer rating forumula accurately measures a signal caller's passing efficiency, it fails to comprehensively account for his overall effectiveness. After all, today's elite quarterbacks do more than just pass the ball.

Expansion of the passer rating formula would give birth to a comprehensive quarterback rating, justifying the quantification of the performance of the modern quarterback by acknowledging his ever-emerging role as a more complete football player. To accurately gauge efficiency as a measure of "effectiveness," key rushing statistics, including yards per attempt, touchdowns per attempt, and fumbles per attempt must be accounted for in the numerical evaluation of a quarterback.

A quarterback's ability to "make plays with his legs" is much less an intangible quality than physical production that can be measured in raw statistics. Did first-year starter Michael Vick have an above-average '02 season in leading the Falcons to the playoffs and establishing himself as arguably the Most Valuable Player in the NFL? Rhetorical questions ask for rhetorical answers, but Vick's 81.3 passer rating somehow begs for the contrary.

A rating system might be in order for all the offensive skilled positions, wide receiver included. After catching 106 balls in '01, Tampa Bay's Keyshawn Johnson earned a spot in the Pro Bowl. But Johnson's 11.9 yards per catch and lone touchdown ranked at the bottom of the league among starting wide-outs. A wide receiver efficiency rating based upon (1) touchdowns per pass thrown to, (2) catches per pass thrown to and (3) yards per catch would help dispel the myth that, despite his opinion, Keyshawn is an NFL superstar.

Our love for the NFL is surpassed only by our passion for scrutinizing its players. Few would argue with the current rating's ranking of Steve Young, Joe Montana, Dan Marino and Brett Favre, respectively, as the modern era's four greatest quarterbacks. But in order to properly analyze today's quarterback as more than a passer, we must scrutinize the very criteria by which we scrutinize him. Evolution demands it.



[Edited on 5/10/04 by TRD]

so are you saying that we should make a rating for efficent to running qbs?

the statistical rating of a football player isn't as easy as for a baseball player, in baseball while a team sport there are more individual accomplishments that can be isolated and recorded, hits, runs, errors etc. in football it becomes much more difficult, for the qb for example, his passing stats depend as much on his offensive line, coaches game calling, field position, receivers abilities, respect for his teams running game, all kinds of variables, you can be the best passer in the world but your stats may not reflect it if the circumstances don't support the passing game

its true that football is more of a team game, and that is gong to remain the same for years and years to come, its very hard for a team to have one standout player, even though there are top players on the team, they cant get the job done without the supporting caast.

Dean- are you trying to say that there should be a passer rating for more indinviduals than a team effort?

yes it seems as though the qauterbacking position is taking a new effect into the running qb's i personally dont like this as much, i think its going to hurt the game with more running and stuff than a manning type of qb.

Originally posted by freak29
yes it seems as though the qauterbacking position is taking a new effect into the running qb's i personally dont like this as much, i think its going to hurt the game with more running and stuff than a manning type of qb.

i dont like this course alos, i will alwyas have the impression that the qb should stay in the pocket and mover out when he has too!

how would we make a rating system for all offensive players? id ont get how that would work out.

Hey guys.

A few things:

-Yeah, I'm saying that by emcompassing rushing statistics and other factors such as net yards gained (which are a direct reflection of elusiveness and overall pocket presence) into the widely accepted passer efficiency rating formula, the metric would be more efficient.
After all, this measurement of efficiency is only as effective as it is efficient and vice versa.

-As far as the stat being for individuals, yeah, that's the point. Football is a nature a team game, there's no questioning that. But it's fun to see if we can nail down an individual players performace statistically. We don't factor in offensive line prowess when looking at a running back's yards per carry, so why do it for quarterbacks. I'm not quite sure what the issue is.

-Creating an efficiency stat for other players is something that the guys at MIT shoudl look into, not me. But it would follow the same principles as the quarterback rating shown in the article: the higher, the better in the macroeconomic sense, and in the microeconomic sense, the higher, the better for yards per reception, yards per ball thrown to, catches per ball thrown to, touchdowns per reception and the lower the better for fumbles. In other words it's a better, more efficient performance if a player catches all four balls thrown to him for 60 yards (15.0 per catch) and a touchdown than a guy when catches eight of ten for 80 yards (10,0 per catch) and a touchdown.

-DC

Hey, I like your ideas, DC. Do you think Tagliabue would be up for this change? Or, even Madden?

Not sure what the Commish has to do with the stats. But if you put him on the spot, I'm he'd give it a thumbs up, or at the very least some expression of Taglia-bue-yah.

Madden is a bit old school not to mention a former coach, so I'm not so sure he'd be into it. He's more concerned about what goes on at the line of scrimmage, and he should be: that's where the games are won.

But for the media and fans, it would be fun to follow. After all, without their media-friendly philosophy and intensely loyal fan base rooted in superior marketing and - more importantly - a superior product on the field, the NFL wouldn't be America's Real National Pastime. Go stats.

To everybody its all about the stats, i think that the pats with their 18 game winning streak arnt so much into the stats as other teams are. Some teams or players on those teams are all about the money and the stats which leads to the fame.

well, can we collectively come up with a better system? Me thinks that we could. What stats would be important to include for a running QB?