Protect Rusell Wilson by Turning Him Loose

The biggest debate in Seattle this fall hasn’t been Obama or Romney, or even Inslee or McKenna. Instead, partisan camps have formed behind Seahawks quarterbacks Matt Flynn and Russell Wilson. After his starring role in leading the Seahawks past New England, Wilson seems to have won over most skeptics, but another performance like Thursday’s second half at San Francisco, when Wilson completed just three passes, could reignite the discussion about the rookie and his more expensive backup.

In defending Wilson’s starting role after a loss at St. Louis last month, Seahawks coach Pete Carroll put the blame for Wilson’s poor numbers on his own conservative handling of the offense.

“I really think this is me holding the lid on it right now,” Carroll told reporters. “And I’m overseeing all of that, and making sure that what’s more important to us is that we take care of the football – more than anything. I don’t care about the yards.”

Given the team’s strong defense and special teams, Carroll’s approach makes sense from the Seahawks. Surely, Carroll also feels like he’s protecting Wilson by relying on powerful running back Marshawn Lynch to carry the load offensively. I don’t think a detailed analysis of the team’s play calling bears that out.

To better understand the Seahawks offense, I downloaded play-by-play from each of the team’s seven games and coded it by down and distance, run or pass (counting scrambles as passes), result in terms of yards (including penalties, and penalizing turnovers as -56 yards +/- the change in yardage per this old Football Outsiders analysis), location on the field, situation (run, pass or neutral) and success (based on Football Outsiders’ baselines, which originally come from The Hidden Game of Football).

Overall, the Seahawks have been about equally effective passing and running the ball:

Play    #    YPA    SD     Suc
Rush   203   3.8    5.0    .41
Pass   199   4.1   16.5    .47

These numbers are very different than the official team totals, since they include turnovers, sacks and penalties. When you factor all that in, pass plays have averaged slightly more yards per attempt than run plays. They’re somewhat more likely to be successful (picking up enough yardage to make a conversion likely), but far more volatile because of big plays both positive and negative. That last factor is part of why most teams, who are more effective passing than running, tend not to just pass all the time. That the Seahawks are about equal in both areas means they are a good running team and a terrible passing one.

Overall, the Seahawks run and pass almost identical amounts, but that changes substantially depending on the situation. I called run situations first down and 10 (or fewer) or plays with one yard to go on any down, called pass situations third (or fourth down) with at least 4 to go and everything else neutral. Here’s how the Seahawks stack up by situation:

Situation   Rush   Pass   Rush%
Neutral      70     82     .461
Pass         11     60     .155
Rush        122     77     .613

The percentages there make sense, especially since my definition of rushing situations is a tad more liberal than passing situations. The more important thing to note is how the Seahawks’ success in running and passing varies by situation.


Play     YPA    Suc
Rush     3.7    .41
Pass     2.2    .41

In neutral situations, the Seahawks are about equally successful. The reason for that low yards per attempt for passing plays is that four of Wilson’s seven interceptions have come on such plays, which is likely a coincidence.


Play     YPA    Suc
Rush     5.6    .18
Pass     3.5    .28

As you might expect of a team that struggles to pass, the Seahawks have had a rough time converting in passing situations. Note that the high yards per carry is a product of draws on third and long, which explains the discrepancy with the success rate.


Play     YPA    Suc
Rush     3.6    .43
Pass     6.7    .57

Here is the most interesting layer of analysis. The Seahawks are decent when they run the ball in rushing situations, but tremendous when they play against type and pass the ball. Both yards per attempt and success rate show how effective these plays have been. Narrowing to first and 10 plays only strengthens this conclusion:

FIRST & 10

Play     YPA    Suc
Rush     3.9    .36
Pass     7.2    .58

While yards per carry go up, the success rate of runs–which need five yards on first and 10 to be considered successful–goes down because we’re no longer factoring in picking up short-yardage situations. Meanwhile, passing plays become even more effective, netting better than seven yards per attempt. Of the Seahawks’ five best plays this season, four–including the game-winning touchdown to Sidney Rice against New England–have been passes on first and 10. Wilson excels at selling the play action, giving receivers an opportunity to get open deep downfield.

By running on first down, Carroll and offensive coordinator Darrell Bevell believe they are protecting Wilson by trying to avoid third and long situations where opposing defenses can bring pressure against an offensive line that struggles to block edge rushers. However, by reducing the likelihood of picking up early first downs, they’re also making third down plays more likely. One other thing the numbers show is that the Seahawks have been ineffective when it comes to converting third and short or medium–aside from third and one, a likely running down.

Dist   #    Suc
1     12   .667
2      5   .200
3      8   .375
4     13   .462
5     12   .333
6      7   .286
7      5   .400
8      2   .500
9      5   .200
10     9   .000
10+   13   .077

Now, the obvious counterpoint to arguing the Seahawks should pass more on first and 10 relates to game theory. Part of the reason these pass plays are so successful is because opposing defenses expect to run. The more the Seahawks pass, the less successful each pass will be. Here’s where I think the research Aaron Schatz published on FO based on a conversation with me comes in. Going into the San Francisco game, the Seahawks offense had been at its best by far during the first quarter, ranking eighth in the NFL in DVOA. The Seahawks were also 17th in the fourth quarter, but near the bottom of the league (27th and 26th) during the middle two quarters.

The difference between the way the Seahawks play in the first quarter, when their plays are scripted, and the second and third periods can be traced in part to their run/pass balance. Here’s how that looks by quarter in run situations:

Qtr  Run  Pass   Run%
1     26   20    .565
2     28   17    .622
3     33   16    .673
4     35   24    .593

The Seahawks come out aggressively, attacking down the field with play action, then settle into a more conservative style of calling plays, especially after halftime. During the fourth quarter, they’ve frequently had no choice but to pass when trailing, with generally positive results.

It makes no sense for the Seahawks to morph into the Saints and start winging the ball around the field 50 times a game. That would be a waste of Lynch’s talent, and too much to put on Wilson, the line and a group of receivers that have been plagued at times by dropped passes. By maintaining their more balanced play calling from the first quarter deeper into games, however, I think they can avoid some of the three and outs that have been all too common during the middle quarters. Letting Wilson pass may ironically be the best way to take the pressure off of him.


Rethinking Expectations for the Huskies

There’s a certain mindset you hear from time to time from Washington football fans, whether it’s muttered in the stands, declared at tailgates or used as trash talk against rival supporters. The thinking goes that any day now, the Huskies are going to return to the place atop the conference they occupied during the early 1990s, when Don James led Washington to three consecutive Rose Bowl appearances, including a shared national championship in 1991.

In the context of that perspective, Bob Rondeau shared a fascinating statistic before Saturday’s disappointing loss at Arizona. Reminiscing about the famous 1992 game where the Wildcats’ Desert Swarm defense snapped a 22-game Husky winning streak, Rondeau pointed out that since that day, both Arizona and Washington had exactly .500 records. Of course, since that includes some solid years under Jim Lambright and Rick Neuheisel, that means the Huskies have been substantially worse than that over the decade since Neuheisel was fired in 2003. Washington is 40-76 over the last 10 seasons, including this one in progress, and even less competitive in conference play. Check out the cumulative standings among the holdover Pac-10 teams during that span:

Team                W     L      %
USC                66    17    .795
Oregon             61    21    .744
Oregon State       48    34    .585
California         46    37    .554
Stanford           41    41    .500
UCLA               39    43    .476
Arizona State      39    43    .476
Arizona            31    51    .378
Washington         25    57    .305
Washington State   21    61    .256

It’s getting more and more difficult all the time to attribute those issues to specific coaches. Using the same method I did last season to evaluate game-by-game performance using Sagarin ratings to create a baseline, I went through every game over the last three seasons. The lines indicate different seasons, and the double lines reflect Washington’s two coaching changes:

If there’s a trend here, I’m not seeing it. While Washington has never struggled as badly under Steve Sarkisian as during the final seasons for Keith Gilbertson (2004) and Tyrone Willingham (2008), the best years of the Sark era are relatively indistinguishable from other good campaigns with the exception of resulting in bowl appearances.

This year’s results have highlighted an interesting pattern. The Huskies have tended to get blown out more under Sarkisian than Willingham. Aside from 2008, when the team fell apart and later quit after Jake Locker‘s injury, Washington suffered relatively few lopsided losses during the Willingham era. From 2005-07, the Huskies were beaten by at least four touchdowns just once (a 56-17 loss to Cal in 2005). By contrast, Washington has lost three blowouts this season (at Arizona, at LSU and at Oregon) and eight during the last three seasons.

The Huskies are having more difficulty playing elite opponents during Sark’s tenure–and there have been plenty of them on the schedule this year, with one more to come in Oregon State next Saturday, the fifth team Washington will face ranked No. 11 or higher at the time. At the same time, Sarkisian’s Huskies have done a better job of taking care of business than Willingham’s teams. During the last four years, Washington has lost only one game that a team playing identically at the level of this year’s Huskies would be expected to win: last season’s loss at Oregon State with Nick Montana at quarterback in place of the injured Keith Price. Letdowns against less talented teams were commonplace under Willingham, including multiple Apple Cup losses.

If that trend holds, it’s good news for the rest of the season. After Washington gets through the Beavers, the schedule will start to even out. The Huskies will face a winnable game at Cal, then be favored in their last three games. If Washington plays consistently, this can still be a bowl team.

In the larger picture, though, the numbers suggest patience and realistic expectations. The Huskies are no longer going through a slump; 2004 and 2008 aside, this is a consistent level of play. Today’s recruits weren’t alive when Washington was the top team in the Pac-10, and the Huskies dominated a conference whose landscape was entirely different. Looming possibility of sanctions or not, Oregon isn’t going anywhere, and Oregon State has become a consistently solid team over the past decade. Both schools, particularly the Ducks, offer an enticing alternative for California kids who are interested in coming to the Northwest.

In their own way, the Beavers offer hope. Anyone claiming back in 1992 that within two decades Oregon State would be one of the Pac-X’s top three teams over a 10-year period would have been laughed out of the room. No college’s fate is set in stone. Washington can get better, and eventually will get better, but it’s time to stop thinking that a return to the early ’90s is only a matter of time.