--------------------------------------------------------------------------------
DR. BOB
3 Star Selection
***UTAH ST. 30 Fresno St. (-16.5) 34
Fresno State was expected to be a good team and were anointed as a possible BCS buster by media "experts". A win over Rutgers to start the season, followed by a close home loss to a Wisconsin team rated in the top 10 added to the misconception that the Bulldogs were a good team. Turns out that Rutgers is a bad team and that Wisconsin is now 3-4 and falling fast. Fresno's other wins have been an overtime win against 2-5 Toledo, a 5 point win against a mediocre 3-4 UCLA team and a 46-32 win as a 34 point favorite against an Idaho team that may be the worst in Division 1A. Fresno State is 0-5 ATS since their season opening win over Rutgers and the Bulldogs' very good offense (6.6 yards per play against teams that would allow 5.6 yppl to an average team) is balanced out by a horrible defense that has allowed an average of 37 points and 6.3 yppl in 4 games since losing run-stuffing defensive lineman Jon Monga early in their week 4 game against Toledo ? to 4 bad offensive teams that would combine to average just 4.8 yppl against an average defensive team. Fresno generally has good talent among their starters, but their lack of depth along the defensive line has hurt them for a second consecutive season (they went from decent to horrible defensively last season after losing some defensive line starters in week 5). Utah State is a bad offensive team, averaging just 4.7 yppl against teams that would allow 5.6 yppl against an average team, but they are considerably better than Fresno State's defense and Aggie's running quarterback Diondre Borel and backs Turbin and Marsh (444 combined yards at 4.7 ypr) should feast on a Fresno defensive front that can't stop the run ? the Bulldogs have allowed 237 yards at 6.3 yards per rushing play since losing Monga despite facing teams that would combine to average just 4.4 yprp against an average team. Fresno will certainly score a good number of points too against a sub-par Utah State defense that rates at 0.5 yppl worse than average, but my math model favors Fresno by just 7 ? points in this game and the Bulldogs are not to be trusted laying points when they are not playing well. The Bulldogs are just 11-25-2 ATS under Pat Hill as a favorite of 2 points or more following a point spread loss, including 0-4 ATS this season and 4-19-1 ATS since 2002. Fresno also applies to a negative 34-83-2 ATS big road favorite off a bye week angle that has won 9 consecutive times for the home dog. I'll take Utah State in a 3-Star Best Bet at +14 points or more and for 2-Stars from +13 ? to +10 ? points.
3 Star Selection
***GEORGIA TECH (-13.0) 33 Virginia 10
Virginia appears to be getting better and enters this game riding a 3 game win steak despite being an underdog in each game. I've been on the Cavaliers in recent weeks at home, but their recent improvement is most likely a function of playing 3 consecutive home game rather than Virginia being a good team. Virginia has always played better at home under Al Groh (32-15 ATS, 29-12 not favored by more than 23 points) than they've played on the road (14-27 ATS). The Cavaliers have also been better with revenge (22-6 ATS) under Groh than they have been when the other team has the revenge motive (16-25 ATS in those games). And, Groh's team has been much better after a loss (23-12 ATS) than they have after a good game (15-22 ATS after a win and spread win). History is certainly not on the side of Virginia today, as they are on the road after 3 consecutive win and covers against a superior team that they happened to have beaten last season. Virginia is just 6-20-1 ATS on the road against a team with revenge and a horrendous 1-18 ATS on the road after a game in which they won and covered the spread (1-16 ATS under Groh). Virginia's only two road games this season resulted in a blowout losses against mediocre teams Connecticut (10-45) and Duke (3-31) and Georgia Tech is much better than either of those teams. Virginia does look like an improved team since getting blown out by USC to start the season and they are certainly better offensively now that top back Cedric Peerman is healthy. Peerman has rushed for 327 yards the last 3 weeks and is averaging 6.0 ypr for the season, but I still rate Virginia's offense at 0.5 yards per play worse than average and the Cavaliers were 0.2 yppl worse than average offensively in their 3 game home winning streak. That attack doesn't figure to move the ball very well against a Georgia Tech defense that has allowed just 12.0 points per game and 4.1 yppl in 5 games against Division 1A teams that would combine to average 23 points and 5.1 yppl against an average team. Georgia Tech's triple option offense has also been very good, as the Yellow Jackets have averaged 6.1 yppl in 5 Division 1A games against teams that would allow 5.3 yppl to an average attack. Georgia Tech's offense may be underrated because they struggled to move the ball in a 10-7 win over Gardner-Webb a couple of weeks ago, but the Yellow Jackets' top 2 quarterbacks were held out of that game and the option attack was run by a traditional drop back passer with limited mobility and no experience running the option. Georgia Tech has also lost 8 fumbles in 5 1A games, which has kept them from scoring as much as they should be scoring given how well they move the ball (although they've still averaged 24.4 points in those games against teams that would allow just 19.7 points per game to an average team). Virginia Tech's defense was overwhelmed in their opener against USC and I decided to throw those stats out. Since that game the Cavaliers have been 0.6 yppl better than average defensively (4.8 yppl allowed to teams that would combine to average 5.4 yppl against an average team), but they have been just average defending the run (4.8 yprp allowed to teams that would average 4.8 yprp) and that is a problem in this game. Georgia Tech has faced just one 1A team that is not good defending the run and the Jackets ran for 439 yards at 8.3 yprp in that game against a Mississippi State defense that is actually slightly better defending the run than Virginia is. The line on this game should be much higher and my math model gives the Yellow Jackets a 59.6% chance of covering at -11 ? points. That doesn't include how poorly Virginia plays on the road after a win and cover, which makes Georgia Tech an even higher percentage play. I'll take Georgia Tech in a 3-Star Best Bet at -14 points or less and 2-Stars from -14 ? to -16 points.
3 Star Selection
***CALIFORNIA (-17.0) 38 UCLA 10
Cal struggled last week against a good Arizona squad in Tucson, but the Bears should bounce back from their one bad defensive game of the year by shutting down a sub-par UCLA offense today. Prior to last week, the Golden Bears' top-notch defensive unit had not allowed more than 5.1 yards per play to any team (they allowed 6.3 yppl to Arizona) and they have still yielded just 4.4 yppl for the season against a schedule of teams that would combine to average 5.4 yppl against an average defense. Cal has faced just 3 below average offensive teams in Washington State and Colorado State and the Bears gave up an average of just 5 points in those two games ? so I don't expect a bad UCLA offense averaging just 4.6 yppl (against teams that would allow 5.5 yppl to an average team) to have much success today against a proud Bears' defense looking to atone for last week. Cal's offense has averaged 6.2 yppl this season against teams that would allow 5.5 yppl to an average team, but I rate the Bears' attack at +0.5 yppl for the season after adjusting for the fact that they have 3 runs for 80 yards or more. Cal is likely to have more breakout TD runs, but it's doubtful that they'll come from the 20 yard line or further (I adjust my stats for long plays, which tend to skew the data). Cal's offense is better with Nate Longshore starting at quarterback, as Longshore has averaged 6.6 yards per pass play against teams that would allow 5.5 yppp to an average quarterback, while Kevin Riley has been 0.2 yppp worse than average (5.9 yppp against teams that would allow 6.1 yppp). Longshore is going to throw more interceptions (3.6% lifetime percentage is a bit higher than the national average of 3.1%), but he is still a better overall quarterback than Riley has shown to be this year. Star RB Jahvid Best was taken out of last week's game after running for 107 yards on 16 carries, but Best should play this week after having another week to recover from his dislocated elbow, which kept him out of the ASU game a few weeks ago. Backup Shane Vereen has run for 395 ards at 6.0 ypr, so he's more than capable of playing well if Best is limited. Cal's outstanding offensive line has paved the way for 6.4 yards per rushing play and the Bears should be able to run the ball pretty easily against a soft UCLA defensive front that has given up an average of 274 rushing yards at 6.8 yprp to the 3 good rushing teams that they have faced (Fresno, Oregon, and Stanford). The Bruins' pass defense has been good (5.3 yppp allowed to quarterbacks that would combine to average 5.9 yppp against an average team), but Longshore is 1.1 yppp better than average so the Bears should be able to throw when they need to ? although they may not need to given how well they should run the ball. The big advantage is Cal's defense, which is 1.9 yppl better than UCLA's offense. My math model favors California by 25 points and the Bears should be favored by more than 21 points in this game. In addition to the line value, the Bears apply to a very strong 77-17-1 ATS statistical match-up indicator. Cal is only 5-18 ATS in recent seasons as a favorite of 11 points or more, but they are 2-1 ATS in that role this season thanks to their stingy defense and won by 28 points as a 16 ? point favorite the last time they applied to that 77-17-1 ATS indicator. I'll take California in a 3-Star Best Bet at -20 points or less and for 2-Stars at -20 ? or -21 points.
3 Star Selection
***Mississippi (-5.0) 37 ARKANSAS 23
Mississippi may be 3-4 straight up but the Rebels are a very good team that has out-gained their opponents 6.1 yards per play to 5.1 yppl and out-scored a schedule that is 4.7 points tougher than average by 3.3 points. Arkansas has played an even tougher schedule (7.0 points tougher than average), but the Razorbacks have been out-scored by an average margin of 14.4 points. Arkansas is actually a pretty decent team from the line of scrimmage, rating at 0.3 yppl better than average with Casey Dick at quarterback (5.5 yppl against teams that would allow 5.2 yppl to an average team) and average defensively (5.6 yppl allowed to teams that would average 5.6 yppl against an average team). Ole' Miss, meanwhile, is 1.0 yppl better than average offensively (6.1 yppl against teams that would allow 5.1 yppl) and the Rebels are 0.4 yppl better than average defensively (5.0 yppl allowed to teams that would combine to average 5.4 yppl against an average team) and potentially are going to be better with star DE Greg Hardy rounding into form (2.5 sacks in 3 games) after missing the first 4 games of the season (Hardy was 1st team All-SEC last season). Mississippi's losses this season have all been by 7 points or less to good teams Wake Forest, Vanderbilt, South Carolina, and Vanderbilt and the Rebels would have won at least a couple of those games if not for being a combined -9 in turnover margin in those 4 losses (and at least -1 in each game). Mississippi is good enough to beat Florida on the road when turnovers aren't hurting them and today they face an Arkansas team that is just as turnover prone, as the Razorbacks are -8 in turnover margin in 7 games while Ole Miss is -7 in TO margin in 7 games. My math model actually projects Arkansas with a 0.2 edge in turnovers, but Mississippi is significantly better from the line of scrimmage and in special teams and the Rebels should be favored by more than 7 points in this game. In addition to the line value, Ole Miss applies to a very strong 53-7-1 ATS statistical match-up indicator and there is a chance that Arkansas star RB Michael Smith may not play after suffering a concussion last week (although I expect him to play). I'll take Mississippi in a 3-Star Best Bet at -6 ? points or less and for 2-Stars at -7 points.
2 Star Selection
**TENNESSEE 21 Alabama (-6.0) 20
Hidden by Tennessee's 3-4 record is the strength of their defense, which has given up just 4.2 yards per play this season to a schedule of teams that would combine to averaged 5.5 yppl against an average team. The Vols' defense ranks 6th best in the nation in my ratings, which places them ahead of Alabama's defense, which ranks 7th at 1.2 yppl better than average. Tennessee's stout defensive front, which has yielded just 3.4 yards per rushing play (to teams that would average 5.1 yprp against an average team), has an advantage over Alabama's great rushing attack (5.8 yprp against teams that would allow 4.5 yprp to an average team) and the Crimson Tide have been held in check by the 3 other good run defenses that they have faced. In fact, Alabama has averaged just 4.2 yprp against Clemson, Georgia, and Ole Miss (who would combine to allow 3.8 yprp to an average team) ? so Alabama's rushing attack has been just 0.4 yprp better than average against good defensive teams. John Parker Wilson has been just mediocre throwing the ball, averaging 6.1 yards per pass play against teams that would allow 6.2 yppp to an average quarterback, but I decided to throw out his horrible game against Tulane, when the Tide were without All-American candidate left tackle Andre Smith (Wilson was sacked 4 times and pressured constantly in that game). In the 6 games when Smith was protecting his back Wilson averaged 7.0 yppp (against teams that would allow 6. 2yppp). Tennessee's pass defense is 0.8 yppp better than average, so Wilson's passing numbers will probably be worse than average since this game is on the road. Tennessee's offense was a major problem early in the season, but the insertion of Nick Stephens as the starting quarterback has helped the Vols' attack, as Stephens has been much better throwing the ball in his 3 starts (6.2 yppp against teams that would allow 5.8 yppp to an average QB) than former starter Jonathan Crompton was in the first 4 games of the season (5.0 yppp against teams that would allow 6.0 yppp). Tennessee is still only about average offensively (on a national scale) with Stephens at quarterback, but my math model projects 19 points for the Volunteers in this game, which should be enough to cover given how well their defense is playing and given the negative situation that unbeaten Alabama applies to. Alabama applies to a very negative 21-80-4 ATS situation as well as a negative 23-59-1 ATS situation that plays against unbeaten teams. Tennessee also is above average in special teams with punter Britton Colquitt back after serving a 6 game suspension. Tennessee's net punting was horrible (29.0 yds) without Colquitt, but he averaged 36.7 net yards per punt last year and field position will be important is a low scoring game between two good defensive teams. On the injury front, Alabama will be without run-stuffing star DT Terence Cody for a few weeks while Tennessee will miss star G Anthony Parker and WR and returnman Gerald Jones. Cody and Parker probably cancel each other out, but losing Jones will hurt the Vols' special teams a bit (although not as much as having Colquitt has helped their special teams rating). My math model favors Alabama by just 4 ? points and the situation merits a play against the Tide in this game. I'll take Tennessee in a 2-Star Best Bet at +4 points or more.