2012 Pro Bowl by the Numbers: AFC
So the NFL Pro Bowl Rosters were released yesterday. That's nice. I'm sure that the trip to Honolulu will totally make Peyton Manning or Tom Brady forget about the Divisional playoff game that they just lost to the other. They seem like the type of people that like to do nothing more than relax.
These days, arguing about the Pro Bowl is a bigger waste of time than trying to replicate the recipes you see on Chopped. You're just never going to get it right no matter how hard you try; you're always going to leave someone out.
Unless you have the numbers. Oh, those beautiful, intelligent numbers. Specifically, our Net Expected Points (NEP) figure, that I once explained in a past MVP Watch article, as such:
Every single situation on the football field has an expected point value; that is, how many points an average team would be expected to score in that situation. For example, the Chiefs may be facing the Pittsburgh Steelers, with a third and two on the 50 yard line. That's a ton of variables, but luckily, numberFire has data from the past dozen years of every single play, so most situations have come up at least once. According to our data, an average team may be "expected" to score 1.23 (estimated number) points on that drive. However, Jamaal Charles reels off a 32-yard run to bring the Chiefs into the red zone, increasing the "expected" point value of the next play to 4.23 (still an estimated number) points. Jamaal Charles then gets credit for the difference, in this case 2.96 points, as his NEP total. That's Net Expected Points.
Since passing is often more efficient than running the ball, you'll often see running backs with negative NEP per play scores, meaning that they are losing their team expected points every time they touch the ball. Receivers and tight ends, meanwhile, will usually have high, positive NEP per play scores, since receivers don't touch the ball unless it's a high-yardage completion. Quarterbacks can be in the middle, either positive or negative: completions typically help their score, while incompletions lower it. So when you're looking at NEP, it's important to look at the numbers based on position.
So, we decided to look at our ideal Pro Bowl against the actual Pro Bowl based on the numbers, using NEP as our guide. Since we're using NEP, we're only going to be looking at the offensive skill positions - quarterback, running back, wide receiver, and tight end - as those are the positions the formula explains the best. First up is the AFC, where the league nailed twp of the skill positions and missed another horribly.
Quarterback
The Actual Roster: Peyton Manning, Tom Brady, Matt Schaub
The Ideal Roster: Brady, Manning, Schaub
Well what do you know, the NFL actually got it right! I'm not going to automatically assume that they were using numberFire's analytics, but if it happens to fit...
Brady's an obvious selection; he's even our likely MVP this season. Manning isn't far behind, though, sitting at 165.35 NEP to Brady's 206.44. Both players are in the top five of the league in total NEP.
Matt Schaub actually sits all the way down in 10th on the NEP list among NFL QBs, with 124.05 points above expectation gained for the Texans this season. But he's the victim of a little conference bias - seven of the top nine QBs on our list this year all reside in the NFC. So even though Cam Newton may be a better QB by our standards, not even his leftover funds from his Auburn recruitment could buy him a Pro Bowl slot. Schaub, meanwhile, gets in by almost a 30 NEP advantage over fourth-place Ben Roethlisberger.
Running Back
The Actual Roster: Arian Foster, Jamaal Charles, Ray Rice
The Ideal Roster: C.J. Spiller, Rice, Stevan Ridley
One of the keys to the Pro Bowl is attracting big name talent; that's why Arian Foster and Jamaal Charles will be packing their bags for Honolulu at the end of January. It's certainly not for their play.
In terms of Net Expected Points per rush, Foster actually wasn't that bad this season. At -0.08 NEP per rush, Foster sits just below Rice for fourth-best among all AFC starting backs. But with so many rushes, Foster had the opportunity to lose a substantial amount of value for the Texans, entering Week 17 at -19.20 NEP amassed. Charles was even less effective, finishing at -0.09 NEP per rush this season and -15.15 NEP amassed total for the Chiefs. To be fair, they probably would have been selections No. 4 and No. 5 for the numberFire AFC squad.
But keeping them in the game over C.J. Spiller is an absolute joke. And not the funny Dave Chapelle type, the "What did he just say?" Carlos Mencia type. Easily the most efficient back this season, Spiller's 0.11 NEP per rush is poised to become the second-highest average since 2000 (the first year with NEP data) for a back with at least 150 carries. The only player better: Charles himself in 2010 with 0.18 NEP per rush. In a league where 95 percent of the starting backs have lost their team net points due to the inefficiency of running, Spiller gaining the Bills 37.75 NEP of value is a minor miracle. (Think turning water into Propel rather than wine.)
Oh yeah, and Rice and Ridley aren't bad either. Rice isn't the most efficient runner - although at -0.07 NEP per rush he's still better than Foster or Charles - but his main value comes through the passing game. He's gained the Ravens a running back-high 31.55 NEP of value through the passing game alone. And Ridley doesn't have the high stats because of his offense, but he was easily the second-most efficient starting back in the AFC this season at -0.04 NEP per carry.
Wide Receiver
The Actual Roster: A.J. Green, Andre Johnson, Reggie Wayne, Wes Welker
The Ideal Roster: Johnson, Demaryius Thomas, Eric Decker, Welker
To be fair, the NFL didn't select a particularly poor roster here. Their four AFC receivers are ranked No. 1 (Johnson), No. 4 (Welker), No. 6 (Green), and No. 10 (Wayne) in the AFC according to our numbers. It's just that in choosing their vacation selections, they followed the same pattern vacationers have been using for years: fly right over the Rocky Mountains and never look back.
Until he was overtaken by Andre Johnson this past week, Demaryius Thomas had been our number two receiver in the league behind Megatron for a while. With a catch rate of 65.4 percent, Thomas ranks in the upper-third of all receivers of converting targets that come his way. And he's had plenty of them - his 133 looks is tied for 12th among all receivers. It's resulted in 76.65 NEP gained for the Broncos after incomplete targets are removed. Normally, this would be the spot where I wax poetic about "People pay too much attention to touchdowns, they're random, and I'm pretentious, yada yada yada." But I can't even do that here - Thomas's nine TDs ranks third among all AFC receivers, behind Green and his teammate Decker.
Oh yeah, what about that Eric Decker? Well, he has nine fewer receptions, but he's also been thrown to 18 fewer times - his 68.4 percent catch rate is third-best among the NFL's top 15 receivers, only behind Randall Cobb and Andre Johnson. And he's made those looks work for him in a big way - his 11 TD receptions is tied with Green for the most in the AFC. Only his 988 yards holds him back, but that's an empty figure - the 73.34 NEP he's gained for the Broncos is fifth-best among all NFL receivers.
But what about A.J. Green and all of his TD receptions, or to a lesser extent, Reggie Wayne? Well, they need to start catching the passes thrown their way like a true receiver should. Green may have 95 receptions, but that's come on 164 pass attempts thrown his way for a below-average 58.6 percent catch rate. He's given the Bengals 108.36 expected points when just his catches are taken into account, but when you remove the passes thrown his way that he didn't catch, that number freefalls to 40.78 expected points gained. Wayne's even worse; his 56.0 percent catch rate has resulted in gaining only 23.03 points above expectation for the Colts.
Nice try, NFL. You almost got it right. But the big yardage numbers don't necessarily mean that a player's helping his team, especially if he can't catch the ball.
Tight End
The Actual Roster: Rob Gronkowski, Heath Miller
The Ideal Roster: Gronkowski, Miller
Cue the Hallelujah chorus, the NFL got another completely right! Not that it was too difficult; only Jermaine Gresham, Owen Daniels, and maybe if you're really stretching Brandon Myers had a case. But these two are easily the most deserving of the bunch in the AFC this year.
Think about Gronkowski this way: he's only played in ten games so far this year. And despite getting two-thirds of the playing time of other tight ends, he's still second in overall value added to his team. At 57.71 NEP gained, Gronkowski's added about a touchdown per game to the Patriots; the only other tight end even over five NEP per game is Tony Gonzalez. And while his 70.7 percent catch rate isn't the ungodly 75 percent he put up last season, it's not a bad figure in its own right.
Miller's 70.3 percent catch rate is right there, though, and remains a big reason why he's still Pro Bowl-worthy. At 48.57 NEP this season, Miller has added a bit over a field goal per game to the Steelers this season; that figure was up even higher before a three-for-nine catching day this past week dropped him six NEP. Jermaine Gresham ended up making the race to Honolulu closer than many thought it would be, but Miller outpaces him by 4.95 NEP entering Week 17.
