We are The Pulling Linemen: dedicated to bringing you stories and opinions from the guys at the centre of football - Linemen!

Contact us at info@pullinglinemen.com, or on twitter @PullingLinemen

Wednesday, 12 September 2012

Offensive Line Rating - A New Grading System

Sanchez was safe all day long (P: Paul Schultz)
The NFL is becoming a more and more stat driven league. Fans are hungry for figures they can rate they favourite players on, use in arguments as to why their QB is better than yours, and for some small factor that might give them the edge in their fantasy draft this year. For a running back, you can analyse his YPC and his every-down-capability. For a receiver, you can look at his catches, targets, drops and YAC. For a quarterback, there are a plethora of measurements you can make, from simple completion % and TD:INT ratio, to the Passer Ratings and the new ESPN Total QB Rating (QBR) that take into account a myriad of different numbers boiled down into one simple figure. But what of the offensive line? The 5 guys at the heart of every offensive snap get very little love at the best of times, from all quarters, and now they're being left behind in the statistical revolution too. Of course, each player can be graded on their assignments for each play: Did they pick up the blitz? Did they move their DT out the hole? These can be accumulated to give each player, and each unit, a % score on their performance, but is it really enough? An O-line is a unit. You can't separate each players performance and rate it, as if 4 guys block perfectly but the 5th allows a free blitzer through at the QB, you can't call them 80% successful really can you? No, what we need to really grade an offensive line is a score that covers the performance of the unit as a whole, that gives the linemen credit when a team scores 3 rushing TDs in a game, that puts value in their QB only being hit once all game, and that considers a rush for negative yards a failure on their part.

The Pulling Linemen would like to exclusively introduce the OLR - Offensive Line Rating.

Put simply, the OLR is an equation devised by us that scores offensive line units based on 6 key stats from a performance:
  • Rush yards per carry
  • Rushing TDs
  • Rushing 1st downs
  • Runs for negative yards
  • Sacks allowed
  • QB hits allowed
These stats, plus a few little mathematical intricacies, give us an OLR score. A score of 0 (zero) is basically equivalent to a completely average game, a score of around -1 is a terrible performance, and as you'd expect, scoring 1 or above is statistically excellent. To best explain the OLR, and to show you why we think it's a pretty good measure of how the lines got on, it'd be easiest to look at the NFL's week 1 ratings and highlight the most interesting scores.

Carolina Panthers: OLR= -0.975

The Panthers offense was just terrible on Sunday, managing only 10 yards on the ground in 13 attempts, an average of just 0.8 YPC. They scored only 1 rushing 1st down in the entire game, whilst the OL allowed 3 sacks and 4 QBs hits on Cam Newton. It's fair to say this unit was shocking, and it's OLR of -0.975 was unsurprisingly the lowest of the week.

New York Jets: OLR= 1.053

Entirely the opposite of the Panthers, the Jets scored the highest OLR of week 1. Their score was boosted to 1.053, the only one above 0.500 let alone 1.000, by the fact the OL didn't allow a single QB hit or sack of Mark Sanchez, which undoubtedly aided the team in their demolition of the Bills. The running game wasn't a show-stopper itself, with a YPC of 3.3 and just 1 TD for Shonn Greene, but 5 rushing 1st downs compared to only 1 run for negative yards (out of 36 attempts) shows the OL remained consistent keeping the Bills out of the backfield on both running and passing downs. The Jets score is one that might be hard to beat for the entire season, as keeping a QB untouched for a whole game is a relatively rare statistic.

San Francisco 49ers: OLR= -0.060

The 9ers have an OL often considered as one of the best in the NFL. Joe Staley at LT and Mike Iupati at LG are a pretty formidable duo, but the OLR for this 9ers unit, even as they impressively beat Green Bay on Sunday, does not manage to break 0. It might look surprising, as their run game was potent, with 186 yards on 32 carries and 0 runs for negative yards. However, it's their pass-pro that let them down, conceding 4 sacks and 5 QB hits on Alex Smith. The OLR score doesn't let you just excel in one category, you must play well in both phases to get top marks.

Philadelphia Eagles: OLR= -0.125

The Eagles are a great example of how OLR might be affected by different offensive styles, either for better or worse. With Michael Vick as their QB, there is always an increased chance that this OL will give up QB hits, and potentially sacks, as Vick moves around in the pocket and tries to extend the play with his feet rather than throwing the ball away. Against the Browns in week 1, the Eagles conceded 11 QB hits (most in the NFL) along with 2 sacks. It was this total that gave the Eagles the 4th worst OLR of the 16 winning teams from week 1, despite their excellent 5 YPC rushing average and 10 rushing 1st downs. Of course, there will be better games for Vick personally this season, and the team's OLR will benefit from the extra rushing yards and TDs he'll undoubtedly contribute that a standard "pocket passer" wouldn't. In this particular game, however, the mobile QB factor was a negative influence on OLR.

Tampa Bay Buccaneers: OLR= 0.064

The Buccaneers are an interesting team to look at, as they basically come out as average across the board, giving them an OLR of pretty close to 0 (of the winning teams from week 1, only the Cowboys and 49ers had more "average" OLRs). With a YPC of 3.6 and 8 rushing 1st downs, the run blocking was sound from TB, despite not being able to punch the ball in for a TD from 1st-and-Goal on the Panthers 4. In the pass game, Josh Freeman was only sacked twice and hit twice, one of the better records in the league from week 1, but not as standout as the Jets. Unlike the 9ers unit who performed well in one category and were let down by the other, the Buccaneers performed reasonably but not excellently in both, and the OLR averaged these out to give them both ratings of very close to 0.

Josh Freeman sets to throw in a perfect Tampa Bay pocket (P: Cliff McBride)

Of course, there are bound to be a few little kinks that need working out in the OLR system, as well as a few caveats to the scoring (such as the role of RBs in pass pro and runs of 20+ yards skewing the YPC), but we feel that this system could be incredibly useful in the future for looking at the performance of OL units. We've also heard it said many times that the offensive line is responsible for who wins matches, the "skill" positions are just responsible for the margin of victory. We'll be testing that theory out with OLR, trying to see if there is a correlation between a team's record and their average OLR score over the season. Of course, it's impossible to make any conclusions based on just one weekend of football, but the link between winning and OLR looks promising so far, and as the season goes on we can start to really look inside the numbers.

Get used to hearing about OLR folks, we're hoping it might just catch on!

- Phil Gaskin (@sosayitisaid)
- The Pulling Linemen


  1. Cant wait to see this at the end of the year and see what the OLR says

  2. Just wondering if you've thought of including penalties into the mix?

  3. Currently penalties are not included, but it's something we're considering as an upgrade soon.