After spending hours and hours over weeks and weeks, I got my model ready and got data entered for all 128 FBS teams. I then tested out the variables trying to quantify my gut. I usually pick the 5 teams I know the most intimately and benchmark my model's results against what I think about their season. I tweak the model until it matches for at least 4 out of the 5 teams. Last season, I managed to hit all 5 with the exact win-loss records I felt were appropriate for their season. This season, I could only manage to hit 4 of them. No matter how I altered my model, I could never get BYU to hit the 9-10 win mark, which is where I think they will probably land. Perhaps my model knows something that I don't.
Here is a look at BYU's opponents and how they fare in the model, from toughest game to easiest game.
1. Texas
Rating: 10
Location: Road
Biggest weakness: OL
Greatest strength: 2013 Record
Game result: Texas by 7
2. UCF
Rating: 44
Location: Road
Biggest weakness: QB
Greatest strength: 2013 Record
Game result: BYU by 8
3. Boise State
Rating: 47
Location: Road
Biggest weakness: QB/OL
Greatest strength: 2010-2013 Results
Game result: BYU by 10
4. Houston
Rating: 33
Location: Home
Biggest weakness: DL
Greatest strength: # of defensive starters returning
Game result: BYU by 13
5. Utah State
Rating: 55
Location: Home
Biggest weakness: # of offensive starters returning
Greatest strength: QB
Game result: BYU by 23
6. Connecticut
Rating: 82
Location: Road
Biggest weakness: 2013 Record
Greatest strength: DL
Game result: BYU by 25
7. Middle Tennessee
Rating: 85
Location: Road
Biggest weakness: QB
Greatest strength: # of defensive starters returning
Game result: BYU by 27
8. California
Rating: 90
Location: Road
Biggest weakness: # of projected senior starters
Greatest strength: recruiting
Game result: BYU by 29
9. Nevada
Rating: 74
Location: Home
Biggest weakness: 2013 Record
Greatest strength: # of defensive starters returning
Game result: BYU by 31
10. Virginia
Rating: 86
Location: Home
Biggest weakness: 2013 Conference Record
Greatest strength: # of defensive starters returning
Game result: BYU by 36
11. UNLV
Rating: 88
Location: Home
Biggest weakness: Coach win %
Greatest strength: OL
Game result: BYU by 37
12. Savannah State
Rating: FCS
Game result: BYU by 55
So, there it is, my model predicts BYU to go 11-1, playing Texas reasonably close on the road and beating every body else without too much difficulty. I couldn't move the needle no matter how I played with the model. I think that is partially a reflection of BYU and partially a reflection of the opponents. The model likes BYU. They have won a fair amount of games over the past 4 seasons, they have an experienced QB behind an experienced OL, they have a lot of starters coming back on both sides of the ball, they have a lot of seniors expected to contribute, and Bronco Mendenhall has a reasonably high career winning percentage.
I thought the model would like Houston, UCF, and Boise State a bit more than it does. I think Houston is going to be absolutely explosive on offense. UCF's defense may not be statistically as good as last season's dominant group b/c their offense won't help them out as much (BYU fans can relate to this), but they will probably be an all-around better defense and BYU may find it tough to drive down the field against them and will be hoping for big plays.
I think Boise State has been trending downward for 2 years now, Chris Peterson got out while he still could and I think we'll see one or two more seasons of 10 wins before Boise falls down to that 7-10 wins/season range. It will be tough to win on the blue turf, but my model doesn't think as much of Boise's team this particular season: there certainly are questions on offense and the undersized front 7 might not stand up for the duration of this season against a very run-heavy schedule. Boise State will face 7 consecutive "run-first" teams, with only one bye, before meeting BYU. BYU also gets them on a short week following the Broncos' rivalry game against Fresno State where my model projects BYU will be coming off a 31-point victory over Nevada.
I'm not sure I 100% like the results for games listed as 2-4 above, but I can wrap my ahead around all of the projections and feel good about #1 and 5-12. Is it Friday yet?
Monday, August 25, 2014
Thursday, August 21, 2014
Last and Least: Mo's 2014 Sun Belt Conference Predictions
| ULL | 18 |
| South Alabama | 92 |
| ULM | 97 |
| Arkansas St | 100 |
| Appalachian St | 101 |
| Texas St | 105 |
| Troy | 112 |
| Georgia Southern | 113 |
| Idaho | 117 |
| NMSU | 125 |
| Georgia St | 128 |
| ULL | 8-0 |
| Arkansas State | 6-2 |
| Appalachian State | 6-2 |
| ULM | 5-3 |
| South Alabama | 5-3 |
| Texas State | 5-3 |
| Georgia Southern | 4-4 |
| Idaho | 2-6 |
| Troy | 2-6 |
| Georgia State | 1-7 |
| NMSU | 0-8 |
My model likes Louisiana Lafayette. Road games at Ole Miss and Boise State will be a challenge but I don't see anyone else in the conference with the combination of experience and experience at winning (a proxy for talent, or as much as can be said for a Sun Belt team). Arkansas State and Appalachian State are the only other two teams my model predicts to win 7+ games (an unwritten rule for Sun Belt teams to go bowling).
Independents: Mo's 2014 Predictions
Ratings:
Records:
Navy takes on a relatively easy schedule: neutral site games against Ohio State and Notre Dame, as well as Rutgers at home. Notre Dame's is decidedly more difficult with road contests against Florida State, Arizona State, and USC, as well as a difficult October test at home against Stanford. Notre Dame does look poised to beat Navy in the head-to-head matchup, but even that won't be a gimme for the Irish as they rebuild a bit in 2014. Looking ahead to 2015, I really like Notre Dame's team. They are a year too young right now to face such a daunting 2014 schedule.
| Notre Dame | 19 |
| Navy | 29 |
| Army | 110 |
Records:
| Navy | 10-2 |
| Notre Dame | 8-4 |
| Army | 3-9 |
Navy takes on a relatively easy schedule: neutral site games against Ohio State and Notre Dame, as well as Rutgers at home. Notre Dame's is decidedly more difficult with road contests against Florida State, Arizona State, and USC, as well as a difficult October test at home against Stanford. Notre Dame does look poised to beat Navy in the head-to-head matchup, but even that won't be a gimme for the Irish as they rebuild a bit in 2014. Looking ahead to 2015, I really like Notre Dame's team. They are a year too young right now to face such a daunting 2014 schedule.
SEC: Mo's 2014 Prediction
It's going to be a fun season to watch the SEC. I don't believe top-to-bottom the SEC has been far and away the best conference in America for years and years like ESPN has been touting for 8 years now. Their champion has been the best team in America over that time period. But until this year, I haven't seen the middle of the SEC as strong as they like to think it is. There really are 10 solid teams this year. I think the 8th best team in the SEC could compete for a conference title in every other league in America. I don't think they'd win it, but they'd compete.
Here is how my model rates the SEC teams, followed by conference standings.
Auburn beats South Carolina for the SEC Title. It'll be interesting to see how the College Football Playoff Selection Committee treats a 1-loss SEC champion. It seems almost a certainty that the SEC champ will have one loss. South Carolina has the best chance to go undefeated, but would have to win a game at Auburn to do it. Auburn plays 5 games that my model says will be decided by a touchdown or less. South Carolina only plays 2 such games (but one of them is at Auburn). I guess you could say my model doesn't like either of their chances to go undefeated.
Here is how my model rates the SEC teams, followed by conference standings.
| Auburn | 1 |
| South Carolina | 6 |
| Alabama | 8 |
| Georgia | 12 |
| Ole Miss | 13 |
| LSU | 17 |
| Missouri | 21 |
| Texas A&M | 24 |
| Mississippi St | 26 |
| Florida | 36 |
| Arkansas | 52 |
| Vanderbilt | 57 |
| Kentucky | 79 |
| Tennessee | 83 |
| West | |
| Auburn | 7-1 |
| LSU | 6-2 |
| Ole Miss | 6-2 |
| Alabama | 6-2 |
| Mississippi State | 4-4 |
| Texas A&M | 3-5 |
| Arkansas | 0-8 |
| East | |
| South Carolina | 7-1 |
| Missouri | 6-2 |
| Georgia | 5-3 |
| Florida | 3-5 |
| Vanderbilt | 2-6 |
| Tennessee | 1-7 |
| Kentucky | 0-8 |
Auburn beats South Carolina for the SEC Title. It'll be interesting to see how the College Football Playoff Selection Committee treats a 1-loss SEC champion. It seems almost a certainty that the SEC champ will have one loss. South Carolina has the best chance to go undefeated, but would have to win a game at Auburn to do it. Auburn plays 5 games that my model says will be decided by a touchdown or less. South Carolina only plays 2 such games (but one of them is at Auburn). I guess you could say my model doesn't like either of their chances to go undefeated.
Pac 12: Mo's 2014 Prediction
My model places the Pac 12 as the #2 conference in America behind the SEC. How tough will the Pac 12 be this season? My model has Stanford as the #5 team in America and my model predicts them to go 5-4 in Pac 12 play. If Colorado or California can find a way to win 4-6 of their combined non-conference games, or steal a couple in conference play, according to my model, the Pac 12 would jump the SEC as the best conference in America.
I do not believe Colorado will go winless. I do not see Utah getting all of the 3 Pac 12 wins my model predicts (though I think 3 is a good number, one upset and two of the three they "should" win). Stanford plays at Washington, at Arizona State, at Oregon, and at UCLA. They'll probably pull off one of those 4, but my model predicts 4 losses there, all by less than a TD. UCLA hosts Oregon in the Pac 12 Title game (a rematch of an earlier game won by UCLA): UCLA wins and heads on to represent the Pac 12 in the Inaugural College Football Playoff as a 2/3 seed, my guess is against the SEC champion. I've got to think Florida State runs the table and gets the #1 seed, in spite of a much easier schedule than every one else in the playoff. The SEC champ, whether that's Auburn, Georgia, Ole Miss, Alabama, LSU, South Carolina, or Missouri, isn't going to go 13-0.
| Oregon | 3 |
| UCLA | 4 |
| Stanford | 5 |
| Arizona St | 9 |
| Washington | 15 |
| USC | 16 |
| Oregon St | 30 |
| Utah | 51 |
| Washington St | 56 |
| Arizona | 61 |
| Colorado | 87 |
| California | 90 |
| North | |
| Oregon | 8-1 |
| Washington | 7-2 |
| Stanford | 5-4 |
| Oregon State | 4-5 |
| Washington State | 2-7 |
| California | 1-8 |
| South | |
| UCLA | 8-1 |
| USC | 7-2 |
| Arizona State | 7-2 |
| Utah | 3-6 |
| Arizona | 2-7 |
| Colorado | 0-9 |
I do not believe Colorado will go winless. I do not see Utah getting all of the 3 Pac 12 wins my model predicts (though I think 3 is a good number, one upset and two of the three they "should" win). Stanford plays at Washington, at Arizona State, at Oregon, and at UCLA. They'll probably pull off one of those 4, but my model predicts 4 losses there, all by less than a TD. UCLA hosts Oregon in the Pac 12 Title game (a rematch of an earlier game won by UCLA): UCLA wins and heads on to represent the Pac 12 in the Inaugural College Football Playoff as a 2/3 seed, my guess is against the SEC champion. I've got to think Florida State runs the table and gets the #1 seed, in spite of a much easier schedule than every one else in the playoff. The SEC champ, whether that's Auburn, Georgia, Ole Miss, Alabama, LSU, South Carolina, or Missouri, isn't going to go 13-0.
MWC: Mo's 2014 Prediction
The MWC sure has gone downhill lately. There is talk of Hawaii dissolving football altogether. After turning down multiple suitors for the last 5 years, Chris Peterson finally left Boise State: did he see the writing on the wall? Boise was trending slowly downward and the MWC was doing so more quickly. My model has them sliding below CUSA as the 8th best conference (or 3rd worst), with half the conference settling in the bottom 30 of all college football teams. Still, there is always some intrigue in the MWC with the triumvirate of Fresno State, Boise State, and Utah State. My model predicts the winner of the final regular season game between Utah State and Boise State to win the Mountain Division. The home field advantage gives Boise the close win. Fresno State, by virtue of higher BCS ranking, hosts the championship game where the home field gives them a win and a berth in the Vegas Bowl.
| Fresno St | 39 |
| Boise St | 47 |
| Utah St | 55 |
| San Diego St | 72 |
| Nevada | 74 |
| UNLV | 88 |
| Wyoming | 102 |
| Colorado St | 104 |
| New Mexico | 107 |
| San Jose St | 108 |
| Air Force | 116 |
| Hawaii | 122 |
| West | |
| Fresno State | 7-1 |
| Nevada | 6-2 |
| SDSU | 5-3 |
| UNLV | 4-4 |
| SJSU | 2-6 |
| Hawaii | 0-8 |
| Mountain | |
| Boise State | 8-0 |
| Utah State | 7-1 |
| Colorado State | 4-4 |
| Wyoming | 3-5 |
| Air Force | 1-7 |
| New Mexico | 1-7 |
CUSA and MAC: Mo's 2014 Predictions
Conference USA:
UTSA over Marshall in title game. UTSA looks like a legitimately good team this season. On paper they might have what it takes to go 12-0, even with games at Houston and OK State in addition to hosting Arizona. Model has them 11-1. Old Dominion also looks strong on paper. Those two teams are both somewhat new to FBS level, so perhaps it is only on paper that they look so good...
MAC:
NIU looks like the team to beat again. This is a very bottom heavy conference, so most of the half-decent teams end up with 7+ wins in an 8-game conference schedule. I look for Ohio to rebuild a little this season, come on strong at the end and spring themselves into a successful 2015 season. You heard it hear first, Bobcat fans, make your 2015 MAC Championship plans now!
| UTSA | 23 |
| Marshall | 31 |
| ODU | 42 |
| Rice | 69 |
| North Texas | 84 |
| MTSU | 85 |
| WKU | 94 |
| Louisiana Tech | 96 |
| FIU | 109 |
| UTEP | 111 |
| UAB | 118 |
| FAU | 119 |
| So Miss | 121 |
| CUSA | |
| East | |
| Marshall | 8-0 |
| ODU | 7-1 |
| MTSU | 6-2 |
| WKU | 3-5 |
| UAB | 1-7 |
| FIU | 1-7 |
| FAU | 1-7 |
| West | |
| UTSA | 8-0 |
| North Texas | 6-2 |
| Rice | 5-3 |
| Louisiana Tech | 4-4 |
| UTEP | 1-7 |
| Southern Miss | 1-7 |
UTSA over Marshall in title game. UTSA looks like a legitimately good team this season. On paper they might have what it takes to go 12-0, even with games at Houston and OK State in addition to hosting Arizona. Model has them 11-1. Old Dominion also looks strong on paper. Those two teams are both somewhat new to FBS level, so perhaps it is only on paper that they look so good...
MAC:
| NIU | 27 |
| Toledo | 58 |
| Buffalo | 66 |
| CMU | 68 |
| Bowling Green | 76 |
| Ball St | 91 |
| Akron | 95 |
| Ohio | 103 |
| EMU | 114 |
| Kent St | 120 |
| Miami (OH) | 123 |
| Massachusetts | 126 |
| WMU | 127 |
| MAC | |
| East | |
| Bowling Green | 7-1 |
| Buffalo | 7-1 |
| Ohio | 4-4 |
| Akron | 4-4 |
| Miami (OH) | 3-5 |
| Kent State | 1-7 |
| Massachusetts | 0-8 |
| West | |
| NIU | 8-0 |
| Toledo | 7-1 |
| CMU | 5-3 |
| Ball State | 4-4 |
| EMU | 2-6 |
| WMU | 0-8 |
NIU looks like the team to beat again. This is a very bottom heavy conference, so most of the half-decent teams end up with 7+ wins in an 8-game conference schedule. I look for Ohio to rebuild a little this season, come on strong at the end and spring themselves into a successful 2015 season. You heard it hear first, Bobcat fans, make your 2015 MAC Championship plans now!
Big XII: Mo's 2014 Predictions
For the 2nd straight year, my model predicts a straight line down the Big 12 Standings. OU goes 9-0, Iowa State 0-9, with someone occupying every spot in between. The Big 12 is rated as the 3rd best conference, with #3-8 in the conference really only separated by home field advantage.
I don't really like the symmetry of this, especially with so many close games being predicted. Every team seems to play at home against the team just worse than them and on the road against the team just better than them, which is what causes the symmetry. Iowa State is predicted to have 3 losses by less than 3 points and another one less than a TD. I'd be shocked to see them 0-9, especially with Kansas and West Virginia...
| Oklahoma | 7 |
| Texas | 10 |
| Baylor | 32 |
| TCU | 45 |
| Kansas St | 46 |
| OK State | 54 |
| Texas Tech | 65 |
| West Virginia | 67 |
| Iowa St | 80 |
| Kansas | 89 |
| Oklahoma | 9-0 |
| Texas | 8-1 |
| Baylor | 7-2 |
| TCU | 6-3 |
| Kansas State | 5-4 |
| OK State | 4-5 |
| Texas Tech | 3-6 |
| WVU | 2-7 |
| Kansas | 1-8 |
| Iowa State | 0-9 |
I don't really like the symmetry of this, especially with so many close games being predicted. Every team seems to play at home against the team just worse than them and on the road against the team just better than them, which is what causes the symmetry. Iowa State is predicted to have 3 losses by less than 3 points and another one less than a TD. I'd be shocked to see them 0-9, especially with Kansas and West Virginia...
ACC: Mo's 2014 Predictions
I'm starting to run out of time to give a brief run-down on every team, so I'll just post the ratings and conference record of every team, as based on my model, starting with the ACC.
I don't agree with the Georgia Tech record necessarily. It figures to be somewhat of a rebuilding season and they have 3 games on the road against ACC teams they would beat on a neutral field. My model has those as close games, but all losses. In reality, I think they'll pull off one or two of those along with another "upset" but some teams never figure out how to win on the road: if that is GT's lot this season then it could be a long one. Duke isn't the best team in the ACC Coastal, but they have the most favorable schedule: they avoid Florida State, Louisville, and Clemson from the Atlantic Division and get Virginia Tech at home. If Louisville can start the season off with a win against Miami, I like that 7-1 conference record as they play 4 straight easy games after that and can gain confidence for a young QB and young defense. If they lose that, then I'd flip them with Clemson.
On a side note, Notre Dame goes 3-1 against the ACC, losing at Florida State while pulling off wins at home against North Carolina and Louisville while beating Syracuse on a neutral field.
| Florida State | 2 |
| Louisville | 22 |
| Miami | 28 |
| Duke | 38 |
| Clemson | 40 |
| Virginia Tech | 43 |
| Syracuse | 50 |
| North Carolina | 63 |
| Georgia Tech | 70 |
| Boston College | 71 |
| NC State | 73 |
| Pittsburgh | 75 |
| Virginia | 86 |
| Wake Forest | 115 |
| Atlantic | |
| Florida State | 8-0 |
| Louisville | 7-1 |
| Clemson | 6-2 |
| Syracuse | 4-4 |
| NC State | 3-5 |
| Boston College | 2-6 |
| Wake Forest | 0-8 |
| Coastal | |
| Duke | 7-1 |
| Miami | 6-2 |
| Virginia Tech | 6-2 |
| North Carolina | 4-4 |
| Pitt | 2-6 |
| Georgia Tech | 1-7 |
| Virginia | 0-8 |
I don't agree with the Georgia Tech record necessarily. It figures to be somewhat of a rebuilding season and they have 3 games on the road against ACC teams they would beat on a neutral field. My model has those as close games, but all losses. In reality, I think they'll pull off one or two of those along with another "upset" but some teams never figure out how to win on the road: if that is GT's lot this season then it could be a long one. Duke isn't the best team in the ACC Coastal, but they have the most favorable schedule: they avoid Florida State, Louisville, and Clemson from the Atlantic Division and get Virginia Tech at home. If Louisville can start the season off with a win against Miami, I like that 7-1 conference record as they play 4 straight easy games after that and can gain confidence for a young QB and young defense. If they lose that, then I'd flip them with Clemson.
On a side note, Notre Dame goes 3-1 against the ACC, losing at Florida State while pulling off wins at home against North Carolina and Louisville while beating Syracuse on a neutral field.
Subscribe to:
Posts (Atom)