clock menu more-arrow no yes mobile

Filed under:

How FEI and S+P see the game: Oregon vs. Washington 2014

Another week of which statistical system you like more - this time, for hate week. The stats agree that Oregon is going to win; the questions are simply by how much, whether Oregon dominates on offense or (surprise) on defense, and whether or not Washington will cover.

Jake Fisher and Marcus Mariota should dominate Washington
Jake Fisher and Marcus Mariota should dominate Washington

FEI is the Fremeau Efficiency Index, created by Brian Fremeau. Brian Fremeau is an author at Football Outsiders,ESPN and BCFToys. FEI is an advanced statistical measure for college football that tracks drive efficiency instead of per-play success.
S+P is created by Bill Connelly. Bill Connelly is an author at SBNation, RockMNationFootball Study Hall and Football Outsiders. S+P is an advanced statistical measure which combines success rate, explosiveness per play and opponent adjustments.

The first week of fully adjusted stats is on us - and things are downright weird.

How S+P sees the game:

OVERALL When Oregon has the ball When Washingten has the ball
Category Oregon


UO Off UW Def UO Def UW Off
F/+ Rk 6 (28.8%) 49 (8.0%) 4 (17.6%) 28 (9.5%) 22 (10.5%) 75 (-2.1%)
S&P+ 12 (238.0) 68 (199.8) 3 (126.3) 75 (97.6) 30 (111.6) 58 (102.2)
Play Efficiency

5 (139.5) 79 (96.4) 32 (116.0) 48 (108.5)
Rushing S&P+

7 (137.8) 72 (98.8) 42 (111.8) 51 (108.2)
Passing S&P+

10 (142.7) 82 (94.7) 33 (117.6) 61 (108.4)
Std. Downs S&P+

9 (132.2) 100 (89.2) 24 (120.7) 22 (127.2)
Pass. Downs S&P+

9 (153.5) 32 (123.1) 59 (104.4) 73 (97.5)
Drive Efficiency

10 (118.3) 90 (95.3) 15 (119.0) 64 (101.1)

Oregon moved up 10 points in value (from 228 to 238) but dropped by two points in the rankings entirely due to opponent adjustments and other teams getting that much better. Oregon's D looked better and Oregon's O looked statistically similar to most of the other games they've played so far this year. S+P is still not sold on Oregon due to how bad they looked against other opponents - but they're trending up.

And our running game isn't the worst it's been in 8 years!

Similarity scores for Washington

Washington is a lot closer to teams we will be facing (Stanford, 59th and OSU, 62nd) than almost anyone we've actually faced this season. The closest via ranking is Arizona (39th). Washington is significantly better running the ball but much, much worse passing the ball. Washington is also much worse at actually closing drives.

S+P hates Washington's defense. Hates them so very much. Because of this, Washington is one of the weakest defenses by S+P that we've seen so far. This is again likely a function of turnovers - Washington has had a ton of success on turnovers and scores via turnovers. The closest team Oregon has faced, amusingly enough, is Washington State (59th). The biggest difference is that Washington is better on passing downs. They are otherwise much worse on almost every other down in every other way. Wyoming (96th) is on the other end of the spectrum, and is not particularly comparable to Washington otherwise.

Why does S+P hate Washington while FEI really loves them (more on that in a bit)? Well, because as far as FEI is concerned the FCS game didn't happen for Washington. Them giving up 52 points to EWU? Nope, didn't happen. Combine that with their dominant performance against Cal, and you get very high values for FEI. Throw in a 52-point game against a FCS team and a defense doesn't quite look so hot.

Similarity scores for Oregon

As I usually say all year, team X (in this case, Washington) just hasn't faced anything close to Oregon's caliber. The closest is California (25th), but they are 13 points below Oregon. That being said, Washington did pretty well against that team. Cal is significantly worse at running the ball though comparable in passing and passing downs. Oregon is also much better at scoring than Cal - typically this is a function of turnovers, and Oregon is pretty low.

Stanford (1st) is the best defense Washington has faced - but next up? That's right, it's Oregon at 30th. The next closest is Hawaii - at 66th. Washington's D has faced one of the meh schedules in the nation.

Oregon's offense vs Washington's defense

By S+P standards Oregon should have a field day against the fairly porous Washington defense. The biggest weakness is on standard downs, where Oregon has a 42 point difference in value. Oregon has about the same difference passing or rushing as far as advantages - close to 35 points. If Washington can get Oregon in a long down and distance - or if Oregon shoots its own foot and puts itself there in penalties - there's a good chance that Washington can make a stop, as they have a fairly decent passing downs defense. Otherwise they're likely to give up points and yards early and often. I would fully expect option runs by Marcus Mariota and Royce Freeman to gain a significant amount of yards early in the count. This may look a lot like last week's UCLA game, especially if the offensive line remains at the current health state and Jake Fisher remains as dominant as he was.

Oregon's defense vs Washington's offense

While Oregon is better, it's not so insanely better that Oregon can expect a major shutout. In particular, Washington is very good (22nd) running the ball, which is about a 10 point advantage over Oregon. I would expect Shaq Thompson, Lavon Coleman and Dwayne Washington to all get some yards. That is basically where their advantages end. Oregon has dominant advantages in the passing game, on standard downs and on passing downs. As long as Oregon can get a stop or two in the running game Washington should be shut down.

How FEI sees the game:

This week we have FEI full data as

OVERALL When Oregon has the ball When Washingten has the ball
Category Oregon


UO Off UW Def UO Def UW Off
F/+ Rk 6 (28.8%) 49 (8.0%) 4 (17.6%) 28 (9.5%) 22 (10.5%) 75 (-2.1%)
FEI Rk 1(.279) 22(.163) 5 (.628) 13 (-.554) 21 (-.436) 79 (-.116)
Field Position 24 (.537) 5 (.581)

Raw Efficiency 15 (.173) 18 (.160) 1 (.857) 8 (-.538) 86 (.239) 78 (-.117)
First Down rate

12 (.769) 35 (.597) 123 (.815) 98 (.625)
Available Yards rate

6 (.603) 23 (.357) 99 (.541) 99 (.404)
Explosive Drives

4 (.250) 11 (.048) 78 (.148) 84 (.107)
Methodical Drives

44 (.154) 87 (.161) 120 (.222) 54 (.143)
Value Drives

6 (.553) 12 (.250) 85 (.442) 104 (.298)
Special Team rank 53 (.478) 56 (.445)

Field Goal efficiency 84 (-.220) 52 (.080)

Punt Return efficiency 65 (-.093) 58 (-.070)

Kickoff return efficiency 98 (-.237) 37 (-.101)

punt efficiency 40 (-.167) 36 (-.195)

kickoff efficiency 10 (-.336) 98 (-.049)

Go home, FEI. You're drunk.

Okay, here's why FEI has Oregon as the best team in the nation: Oregon has one of the hardest schedule in the nation. Oregon has played well against Arizona (the 5th best team by FEI) and dominated UCLA (14th). Oregon doing only so-so against Washington State isn't so bad, since they're ranked 34th (only two spots above Michigan State). Basically, the entire Pac-12 is amazing and Oregon is the most amazing amongst the amazing teams, and they've played a hard schedule.

This is almost entirely unlike how Oregon typically gets to the top of FEI - FEI typically looks at the Pac-12 as a fairly weak conference but Oregon's high game efficiencies put them to the top anyway in spite of playing a weak schedule. This year, FEI thinks the PAC-12 is the SEC. And Oregon is the best of that lot.

I don't think anyone but the most absurd Oregon fan would argue that Oregon is the best team in the nation right now - even after beating down UCLA. It's kind of hard to give FEI a lot of credit at this point. This has been true in previous years as well - FEI early on in the season is often quite odd and very volatile. I don't believe it's that good for absolute values or determining, say, if Oregon or FSU would win - but it's okay for relative values.

And both FEI and S+P agree - Oregon should dominate against Washington. They just differ as to how Oregon will do it.

Game Factors: the best and worst of the teams

Again, I'll be talking game splits, as game factors won't be available for a bit.

Oregon's offense put in its second-best performance of the year while the defense did...well, it was there. And not as bad as other times. The real key in the UCLA game was two factors - turnovers and field position. Oregon massively tilted the field against UCLA with an 8.6 difference. Oregon also got 9.4 points via turnovers. When you get 18 points of value from turnovers and field position, chances are good you're going to win as long as you didn't look too bad otherwise. And they didn't.

Washington's offense has been consistent - always providing a positive value (again, no EWU) and having fairly high scores against some teams. Their offense sans the Stanford game has been consistent as well - not good, but consistently decent. The big value that Washington has is that they've gotten 63.2 points of value from turnovers. That's massive. And barring some very different game, a very good sign for an Oregon team that hasn't turned it over all that much.

Similarity scores for Washington

UCLA is the best team in the nation on offense, according to FEI. This is almost entirely because they have played decently against the hardest set of defenses in the nation. This is one of the reasons why Oregon is #1 - because Oregon held UCLA to 24 points in non-garbage time. Oregon holding WSU to 31 points also helped, as they're #4. Arizona is 13th.

I say this because FEI thinks Washington's offense sucks. Even for the PAC-12, they suck.

Wyoming's offense is close (98th). But that's really it. Arizona, WSU, UCLA are all way, way better, and South Dakota isn't on here. Washington is by far the easiest offense in the PAC-12 that we have faced, and chances are fairly good by FEI standards that Washington will be the worst offense we face this season.

Washington on defense is an entirely different beast. Washington will be the hardest defense Oregon has faced so far this season, and that is also not really close. The next closest is Arizona (37th), with a -.270 differential. From what FEi can see this is going to be tough. Washington has faced the 4th hardest schedule on defense so far.

Similarity scores for Oregon

Washington has actually faced something like Oregon...hah, had you going. They haven't. Much like S+P California (28th) is the best offense they've so far faced, but they're not really comparable; Oregon has a .320 advantage over Cal.

With Oregon's dominant win against UCLA their defense skyrocketed from the 70s to 21st. This should tell you two things - FEI is super volatile this early in the season and Oregon has played one heck of a hard defensive schedule (13th in the nation). Washington has played harder in Stanford (2nd) and less hard in Hawaii (43rd).

Oregon's offense vs Washington's defense

Oregon is great at everything save methodical drives. Washington is great at stopping explosive drives and getting three and outs, but is really horrible at stopping methodical drives and gives up a lot of available yards. Based on this, I would expect Oregon to be able to get a lot of success in drives until they get to the red zone, and then bog down a bit. But only a bit, mind you. Oregon can likely crack this game very open if they can get a big play or two. Otherwise it might be (gasp) a bit of a field goal and field position kind of game.

Oregon's defense vs. Washington's offense

S+P figures that Oregon will dominate on offense and hold serve on defense; FEI thinks Oregon will do okay on offense but completely murder on defense. Let's split the difference and say that Oregon will murder the Huskies on offense and defense, okay? Cool cool.

Seriously, as far as FEI is concerned Washington is bad at everything. Not even 'meh' - actually bad. The one thing that they are decent at is methodical drives. They're not likely to get explosive plays, first downs, get much in available yards or really do much of anything. Oregon does have a major weakness, however - they do give up methodical drives quite often (120th in the nation, at about 1  in every 5 drives). Oregon's defense is also massively bolstered by their degree of difficulty - Oregon has faced the 13th ranked hardest schedule so far.

Special Teams

Washington and Oregon are both...okay in special teams. Oregon is bad at field goals per usual (gods I hate typing that) and Washington is better. Washington is bad at kickoffs, but Oregon is even worse at kickoff returns.

A sidenote: Charles Nelson, please stay the hell in the end zone on kickoffs. This crap about taking it out and getting it to the 15 has to stop. Just take your 25 yards and be happy. Please. Just stop.

Anyway, neither team has particularly exploitable advantages here. The best we can say is that Oregon is very likely to continue their good kickoffs, negating one of the advantages Washington has.

So what does this all mean?

This is what I said last week:

Based on S+P, that is again true this week. UCLA doesn't have a killer defense and Oregon has major advantages in the passing game that should be available to be exploited all night. However, that goes fairly against what we've seen of Oregon's offensive identity through 5 games this season. We aren't WSU and aren't going to turn into WSU after a week. Oregon's defense is also a good matchup for UCLA in theory and should stop them at least some of the time.

That's the S+P view. The FEI view sees UCLA as a team that can get a lot of turnovers, get drive-stopping plays and puts a ton of pressure on the other team via big plays. While I think FEI is a bit confused right now, I do think that the relative values of PAC-12 teams are likely right - and UCLA has been better than Oregon so far. If Oregon can limit the turnovers against UCLA I think they've got a good shot.

S+P was much more right. Boy, was it much more right. Oregon did have advantages in passing and running the ball and used them well. Oregon's defense was good enough for most of the day. UCLA's turnover advantages and drive efficiency did not manifest. Oregon limited the turnovers, got a balanced offense for the first time this season, did not hurt themselves with significant mental miscues and played the first truly complete game for the entire game all season.

In this game I also think S+P is more right, especially with Jake Fisher coming back. While Washington's defense has been spectacular they've also had some fairly scary holes - and more importantly have not played great, balanced offenses. Washington's offense, meanwhile, has been at best adequate. Cyler Miles has had flashes of brilliance, but he's not as mobile as Brett Hundley and not as consistent as Connor Cook. I don't think Oregon is going to suffocate Washington's offense but I don't think that Washington can score much without getting very lucky on turnovers - something Oregon has been able to limit handily.

If Washington wins it will be because of big return yards from turnovers and a bad turnover ratio in their favor. Otherwise Oregon should win handily and likely blow out Washington. Based on the numbers I predict a win and a cover.

Goes to 11

This one goes to 11.