Game Analysis

Stanford (+3 ½)  23   WASHINGTON  21

This game could decide the winner of the Pac 12 North division and we’ll find out just how good Washington really is. While Stanford has been beating good teams Kansas State, USC, and UCLA by an average score of 25-12 the Huskies started the season with three home games against bad teams (Rutgers, Idaho, and Portland State) before being tested last week at Arizona. Washington needed overtime to tame the Wildcats, which calls into question just how good the Huskies are (since my model doesn’t think much of Arizona).

Washington’s offense has averaged 7.2 yards per play and 46 points per game but the Huskies have faced 4 bad defensive teams that would combine to allow 6.8 yards per play to an average offensive team. The best defense that Washington faced was Arizona last week (Arizona is 0.5 yppl worse than average) and the Huskies were limited to 28 points in regulation. Washington was only 0.3 yppl better than average last season and their starters have been +0.5 yppl this season so far, which is pretty close to what I projected before the season began.

That unit should be held in check by a good Stanford defense that has limited 3 better than average offensive teams to just 12 points and 5.2 yards per play. Stanford will be without their two starting cornerbacks Meeks and Holder, but the two filling in, Murphy and Alexander, have combined for 5 passes defended and the Cardinal defense played great in the second half last week after Meeks and Holder were injured. I did adjust the Stanford pass defense a bit even though there is no evidence that they’ll be any worse. My model does project Washington quarterback Jake Browning to have pretty good success (7.2 yppp) but the Huskies’ run game should be held in check by a strong Cardinal defensive front.

Washington’s defense was very good last season and has been very good so far this season, as the Huskies’ starting unit has been 0.4 yards per rushing play better than average and 2.0 yards per pass play better than average (they gave up a lot of passing yards in the first 3 games after their starters were pulled and I don’t use those stats in my model). Washington’s run defense is actually better than they’ve shown (4.5 yards per rushing play allowed to teams that would average 4.9 yprp against an average defense) but Arizona’s elusive quarterback Dawkins ran for 180 yards on 11 runs last week. Those numbers aren’t relevant in this game since Stanford’s quarterback Ryan Burns has run only 9 times in 3 games (for 49 yards) and I used on Washington’s run defense numbers against running backs in my model for this game. the Huskies’ run defense has been great against running backs – although tonight they’ll be tested by Christian McCaffrey.

McCaffrey has averaged a good, but not great, 5.5 ypr so far this season but the Cardinal have faced teams that are really good defending the run (those teams would allow only 4.2 yprp to an average team), so he’s actually been really good when compensating for the defenses he’s faced so far. Washington should do a good job containing McCaffrey and the pass defense has an advantage over Burns but Stanford is historically good at converting 3^rd downs and keeping drives alive and that has been the case against this season (50% 1^st downs on 3^rd and 4^th down despite facing 3 good defensive teams).

While Washington should average more yards per play my model projects Stanford to run significantly more plays and for the totals yards to be pretty even. The Cardinal have the edge with their special teams, which helps with field position and converting stalled drives into points with a kicker that is among the best in the nation (24 for 26 on field goals since last season). I’ll call for a mini upset and side with the Cardinal plus the points.