Bambi

October 14th, 2018 at 1:31 PM ^

S&P looks at per play numbers. Army averaged 4.4 YPA passing and 4.3 YPC rushing while OU had 11 and 7.5 respectively. Army managed 21 points but had 4 drives of 15+ plays. S&P views them as lucky to have continually gotten first downs on those TD drives when only averaging 4 yards per play, since one bad play on 1st or 2nd down would normally kill the drive.

Same with Texas. OU had like a 92% post game win expectancy after that game. They killed Texas on a per play basis and Texas only won because of 2 turnovers by OU, including 1 where Kyler Murray literally placed the ball on the ground. 

They dominated every other game.

AnthonyThomas

October 14th, 2018 at 1:33 PM ^

OU lost to Texas but it didn't really harm their S&P+ ranking because they were the better team on a yards-per-play basis. From what I remember their probability for winning that game was well over 50% according to S&P. Couple that with preseason predictions and all of the top-10 clunkers yesterday, and you get the high ranking.

CLord

October 14th, 2018 at 1:00 PM ^

This game  reminded me of the Penn State game two years ago when we crushed a depleted team that  was missing half of it’s starting defense.  Not getting too excited just yet. 

MileHighWolverine

October 14th, 2018 at 1:15 PM ^

Hard to know what to make of these rankings when the #33 team just beat the #7 team....obviously fluk-ish but that seems like too wide a variance IMO. One of those teams is seriously misranked.

FreddieMercuryHayes

October 14th, 2018 at 2:17 PM ^

Not according to the usual stats.  Connelly just actiallu wrote about this.  MSU finished with 4.1 more TOs than expected which usually leads to a 20 point swing.  So they got out played and got a ton of TO luck and pulled out a bunch of trick plays.  Remember, MSU fumbles 4 time...and recovered them all.  PSU fumbled once and MSU recovered it.  Fumble recovery is 50-50 by all stats over enough time.  Plus PSU dropped a couple TOs themselves.  Those huge TO swings are what leads you to win despite getting outgained by a yard and a half on a per play basis.  UM outgained UW by 0.66 per play.  That’s how much MSU got dominated and lucked into the win.

Problem is that this will happen to UM against MSU next week too since that’s what MSU does 

Sambojangles

October 14th, 2018 at 2:18 PM ^

OSU, ND and LSU are the only teams in the top that have Second-Order Wins which are at least 1 game below their actual win total. It suggests that they are not as good as their records, and have been somewhat fortunate to win all their games. 

On the other side, Oklahoma, Penn State and Washington are all +0.8 on 2ndO, unsurprisingly, since they have all played well, statistically, in games they lost. Michigan is at +0.3.