Another mediocre week for myself going 1-4 to bring my overall record to 27-19-4. Dr. Bob went 2-2 to bring his record to 26-23.
Dr. Bob is crushing it so far in basketball and is faring better on his paid picks than he is here.
On to this week's picks. First for Dr. Bob...
Pittsburgh (-7) 26 Oakland 12
The Raiders have won and covered in 3 consecutive game but teams that have done that are just 60-109-6 ATS in road games when facing a team with a winning record and not getting more than 7 points, including 15-46-3 ATS if the streaking team has a win percentage of less than .700. In other words, mediocre and bad teams tend to come back down to earth after a 3 week run of positive variance and I don't mind backing a superior Steelers team that is probably in a nasty mood after getting whipped on national TV by the Patriots. Oakland also applies to a negative 40-97-6 ATS situation while Pittsburgh applies to a very good 143-69-3 ATS contrary indicator. One of the keys to Oakland's recent success has been the strong running of Darren McFadden, who has averaged 5.5 ypr, but running on the Steelers is extremely tough to do (Pittsburgh has allowed just 64 rushing yards at 2.9 ypr) and Raiders' quarterback Jason Campbell isn't nearly as effective throwing the ball without the support of a good rushing attack. My math model favors Pittsburgh by 9 ½ points and my other model favors the Steelers by 12 points, so there is some line value to go along with the technical analysis.
Detroit (+6 ½) 26 Dallas 24
Remember 2 weeks ago when everyone thought how overrated and horrible the Cowboys were? Apparently one good win is all it takes to get back in the good graces of the media, fans and odds makers. I'm not buying it though. Dallas had a great offensive game in their 33-20 upset win at New York (as a 13 ½ point dog), but Jon Kitna is still Jon Kitna and one great game doesn't mean that he's suddenly a great quarterback. That win last week disguised the fact that the Cowboys' defense is still having problems, as the Giants racked up 480 yards at 6.3 yards per play. It was really just one play that turned that game around, as the Giants were picked off in the endzone and Dallas returned it the length of the field for a TD of their own - a swing of at least 10 points. Dallas was due for a good effort last week after getting destroyed by Green Bay the week before, but this week the Cowboys are due for a letdown after their big upset win over a division rival and now hosting a team with a 2-7 record that is coming off back-to-back losses. Detroit, however, may be the best 2-7 team in memory, as the Lions are 8-1 ATS this season with the only spread loss being by just 2 points. They've been consistently good enough to barely lose games this season, as 5 of their 7 losses have been by 5 points or less. Detroit's offense has averaged 24 points per game and they are average on a compensated yards per play basis when Shaun Hill is at quarterback. Hill has been better than his numbers make him look, as his 5.9 yards per pass play has come against teams that would combine to allow just 5.7 yppp to an average quarterback. The Lions call passing plays 45 times per game and that should work against a leaky Cowboys' secondary that has allowed 7.1 yppp this season to quarterbacks that would average 6.7 yppp against an average team. Overall, the Cowboys defense is 0.2 yards per play worse than average and the Lions have a slight edge when they have the ball.
The Cowboys' offense took flight last week and Kitna is now solidly better than average for the season on a compensated yards per pass play basis (7.1 yppp against teams that would allow 6.4 yppp to an average QB), but he also is interception prone (3.7% career interception rate and 7 picks in 4 games this season) and the Cowboys' rushing attack has been bad (3.7 ypr against teams that would allow 4.1 ypr). Dallas has been 0.4 yppl better than average offensively in 4 games with Kitna at quarterback, but last week's 9.0 yppl has skewed that average and 3 of those 4 games have been worse than average offensive performances. Detroit's defense is just 0.2 yppl worse than average and 0.2 yppl worse than average overall from the line of scrimmage, but the Lions are actually a slightly better than average team thanks to their good special teams. Dallas, meanwhile, is not a better than average team and my math model favors the Cowboys by only ½ a point in this game at home.
I mentioned earlier how last week the Cowboys were in a great situation and this week they are not. Teams that win straight up as an underdog of 5 points or more against a division foe are just 22-59-1 ATS the next week if facing a team that is 2 games below .500 or worse. Detroit, meanwhile, applies to a very strong 44-3-1 ATS subset of a 120-63-2 ATS road underdog situation. Detroit has been out-scored by an average of just 1.5 points per game this season while playing a schedule that is 1.8 points tough than average (so they are 0.3 points better than average based on points) and the Lions are better with Shaun Hill at quarterback. Dallas, meanwhile, has been out-scored by an average of 6.4 points per game playing a schedule that is 1.3 points tougher than average (so they're -5.1 points) and they're worse with Jon Kitna at quarterback. I suppose that the coaching change could make a difference and it probably did in getting the Cowboys fired up last week, but new coach Jason Garrett didn't fix the defensive problems and it's unlikely Jon Kitna will have the game of his life every week.
Syracuse (-3 ½) 27 Connecticut 17
Syracuse has struggled offensively against Division 1A competition (4.7 yards per play against teams that would allow 5.5 yppl to an average team), but the Orange have a good defense that is 0.6 yppl better than average (4.6 yppl allowed to 1A teams that would combine to average 5.2 yppl against an average team, adjusted for facing Cincy and Louisville backup QB's). So, overall the Orange are just 0.2 yppl worse than average from the line of scrimmage. Connecticut, meanwhile, is 0.3 yppl worse than average offensively with Zach Frazer at quarterback and the Huskies' defense is 0.2 yppl worse than average (5.5 yppl allowed to teams that would combine to average 5.3 yppl), so U Conn is 0.5 yppl worse than average from the line of scrimmage. Along with the advantage from the line of scrimmage, Syracuse is also projected to run 7 more plays than U Conn and my math model projects the Orange with 363 yards at 5.1 yppl to 281 yards at 4.4 yppl for the Huskies. Syracuse is also 1.2 points better in special teams and my math model gives the Orange a solid 54.4% chance of covering at -4 points based solely on the math. That percentage is enhanced by a negative 10-54 ATS road letdown situation that applies to Connecticut that is based on last week's upset win over Pitt that was aided by a +2 in turnover margin. Syracuse, meanwhile, applies to a solid 178-87-2 ATS late season situation and the decent line value combined with the technical analysis makes the Orange a solid play.
Duke (+11) 28 Georgia Tech 32
Duke matches up well with Georgia Tech, whose option offense is not going to fully exploit the Blue Devils only major weakness - their horrendous pass defense. Duke has given up 8.2 yards per pass play to quarterbacks that would combine to average 6.5 yppp against an average defense, but Georgia Tech only throws the ball 15 times per game. Duke's run defense is actually good since LB Kelby Brown joined the starting lineup in week 4 after missing the first two games and coming off the bench in week 3. Brown has been Duke's leading tackler since he joined the starting lineup and he gets into the backfield with regularity, which is good against an option offense. Since week 4 the Blue Devils have yielded just 4.5 yards per rushing play to teams that would combine to average 5.0 yprp against an average defense and they've done particularly well defending the option. Duke held Army and Navy to a combined 4.2 yprp, which is very good considering those two option teams would combine to average 5.0 yprp against an average run defense. While Georgia Tech is playing into Duke's defensive strength, the Blue Devils' decent attack (389 yards at 5.2 yards per play against teams that would allow 5.4 yppl to an average team) will move the ball pretty well against a sub-par Georgia Tech defense that's given up 5.8 yppl to teams that would combine to average 5.4 yppl against an average defense. My math model only gives Georgia Tech a 434 yards to 395 yards advantage in this game and while Georgia Tech has an edge in projected turnovers Duke has the advantage in special teams. My math model gives Duke a very profitable 56% chance of covering at +11 ½ points.
Cincinnati (-13) 34 Rutgers 17
Cincinnati has been a major disappointment, but the Bearcats still have the makings of a good team. Cincy has averaged 6.0 yppl on offense this season despite star quarterback Zach Collaros missing some time and the Bearcats rate at 1.1 yppl better than average offensively with Collaros under center. Collaros struggled last week in his return to the lineup but Rutgers' defense isn't nearly as good at West Virginia's defense, as the Scarlet Knights are just average defensively (5.4 yppl allowed to teams that would combine to average 5.4 yppl) and worse than average against the pass (6.4 yards per pass play allowed to quarterbacks that would average 6.1 yppp). Cincinnati will be able to move the ball, but the Bearcats have had trouble turning yards into points thanks to their -10 turnover margin in 8 games against Division 1A opponents.
Cincy's defense is very good against the run (4.0 yards per rushing play allowed to teams that would combine to average 4.9 yprp against an average defense), but the Bearcats' secondary gives up too many big plays and the defense is just average overall. An average defense is more than good enough to contain a Rutgers' attack that has been 1.1 yppl worse than average so far this season (4.4 yppl against teams that would allow 5.5 yppl to an average team). Rutgers' offense looks to be improved with the wildcat formation taking the place of quarterback handoffs. Wide receiver Jeremy Deering ran for 166 yards on 29 carries out of the Wildcat formation last week and my math now rates Rutgers as a slightly better than average rushing team if Deering and Mohamed Sanu (if he plays) run the Wildcat on about 80% of the running plays. That's a big improvement from a season rating of 0.8 yprp worse than average running the ball (4.1 yprp against teams that would allow 4.9 yprp to an average team). Even with Deering's great numbers Rutgers still only averaged a modest 4.6 yprp last week and their offense managed just 3.8 yppl and 10 points against Syracuse. They should do better than that today against a mediocre Cincy defense but my math model projects the Bearcats with a 470 yards at 6.8 yppl to 287 yards at 4.6 yppl advantage over the Scarlet Knights in this game even with the huge 1.0 yprp upgrade for Rutgers. Cincinnati's very good stats have not translated into points because of their -1.1 turnover margin per game but it is highly unlikely that the Bearcats will continue to average -1.1 in turnovers going forward and my math model gives Cincinnati a profitable 56.4% chance of covering at -13 points based on the projected stats in this game.
Now on to my picks...
Detroit +6.5 @ Dallas
I don't really believe that Dallas has suddenly become a good team. Yes they are talented but this line is classic overreaction to last week's performance.
Cincy -5.5 vs. Buffalo
Cincy played hard in their loss to Indy and if not for their turnovers could have won that game outright. Buffalo finally won a game and could have a letdown this week.
St. Louis +3 vs. Atlanta
People are beginning to anoint the Falcons as the best team in the NFL. The last team to carry this moniker was the Giants. We saw what happened to them. Atlanta is overrated.
NY Giants +3 @ Philadelphia
Bounce back game for the Giants whose offensive balance and ability to rush the QB make this game at best a tossup. Take the points.