Tuesday, June 30, 2009

Sun Belt SDPI

One of my favorite set of posts from the past two offseasons has been the SDPI recap/early preview. Don't know what SDPI is? It stands for Standard Deviation Power Index and is a tool Eddie Epstein used in his book Dominance to rate pro football's best teams. The basic idea is to look at how far above or below average (by standard deviations) a specific team is relative to their conference brethren. Since each team plays the same number of conference games, it can give us a good idea about who the best team was within the conference. However, it cannot tell us which conference is better. But the purpose of these posts is not to determine which conference is superior, but rather project ahead which teams in a conference will be contenders or also rans in the upcoming season. In the first post on SDPI two years ago, I calculated SDPI based on points scored and allowed within conference play. Last season I used points scored and allowed as well as yards gained and allowed. This season, I'm sticking with yards only. The yardage version of SDPI has a better correlation with future performance than points, and including both last season made the post seem (at least to me) quite muddled. Of course, this is by no means, the end all be all rating system, but it can give us an idea of which teams will improve and decline in 2009. This is our last stop on the SDPI train, wherein we will examine the youngest IA conference, the Sun Belt.

If you want the meat of the article, skip this next paragraph as it just gives an example of how SDPI is calculated. The mean yardage for and against for all Sun Belt teams in conference play was 2758.625 yards. The standard deviation for yards gained was 220.3. The standard deviation for yards allowed was 362.97. Florida International gained 2583 yards in conference play and allowed 2731. Their offensive SDPI was -0.80 = ([2583-2758.625]/220.3). Their defensive SDPI was 0.08 = ([2758.625-2731]/362.97). Their total SDPI was -0.72 which ranked 5th in the conference.

To refresh your memory, here are the 2008 Sun Belt Standings.

Now here are the 2008 SDPI Standings sorted by total SDPI, with conference rank in offense, defense, and total SDPI in parentheses.

The SDPI standings allign pretty well with the actual standings. Troy and Louisiana-Lafayette (now known as just Louisiana) were the cream of the crop, followed by Arkansas State and Florida Atlantic. North Texas was by far the worst team in the league.

Best Offense: Louisiana-Lafayette 1.81
The Ragin Cajuns had an eye-popping offensive game when they took on state rival Louisiana-Monroe. The Cajuns tabulated 728 total yards and averaged an unheard of 12.1 yards per play. Running back Tyrell Fenroy had nearly 300 yards on the ground (297), quarterback Michael Desormeaux added 149 yards rushing, and receiver Jason Cherry had 92 yards on just a pair of carries.

Worst Offense: Middle Tennessee State -1.49
Outside of an offensive showcase against the moribund defense of North Texas (489 total yards), the Blue Raiders averaged only 324 yards per game against their Sun Belt brethren.

Best Defense: Troy 1.41
No Sun Belt team gained more than 358 yards against the Trojans. The Trojans also proved their worth outside the league, holding Ohio State to 309 yards in Columbus and LSU to 340 in Baton Rouge.

Worst Defense: North Texas -1.59
Probably the worst defense in all the land last season. They did manage to hold Florida International to 347 yards in their best effort of the season, though they still gave up 42 points.

Looking ahead to next season, the prohibitive favorite should be...

Troy
The Trojans have either won outright or shared three consecutive Sun Belt titles, and they stand a good chance at making it four straight in 2008. The league's third ranked offense returns both a senior quarterback, Levi Brown, who took over in the sixth game last season and proceeded to throw 15 touchdowns against just three interceptions, and a running back who netted over 1000 yards on the ground, junior DuJuan Harris. Brown also has his top-target back, junior receiver Jerrel Jernigan, who paced the team with 868 yards last season. The offense should improve upon last season's performance. As a Troy fan, the only concern I would have for the offense is that other teams may leapfrog them. Troy was third in the conference in offense last season, but they only averaged about 11 yards more per conference game than the sixth ranked team (North Texas). Troy was good offensively, but their margin over the other Sun Belt teams is not exceptionally large. At least on offense. On defense, Troy was head and shoulders above everyone in the Sun Belt last season. However, that unit does lose six starters, and in particular, three from a secondary that was 29th in the nation in pass defense and 18th in pass efficiency defense. That side of the ball may experience some growing pains, especially considering the Trojans open with three of their first four games on the road. Troy is as good a bet as anyone to win the Sun Belt, but this is not Boise State we're talking about. It won't be a cakewalk.

The team(s) you should be buying are...

North Texas
The Mean Green suffered through an embarrassing 1-11 year in 2008, including a winless Sun Belt campaign. Their lone victory was against Western Kentucky (Sun Belt member starting this season). North Texas was quite pathetic last season (easily the worst team in the conference), but there is reason to expect improvement. For starters, the Mean Green boasted (?) a turnover margin of -17 (117th in the nation) and -13 in league play (last). The primary reason for the poor turnover margin was the 17 interceptions they threw (102nd in the nation) and the fact that they lost two-thirds of their own fumbles (16 of 24). This is the third year of head coach Todd Dodge's spread offense, so the quarterback (likely Dodge's son Riley) should be playing with a more knowledgeable cast, and the interceptions should drop. North Texas should also have better luck falling on their own fumbles, so their turnover margin should improve. The defense, statistically the worst in the nation last season in both yards and points allowed, returns nine starters, including the top seven tacklers, so its virtually impossible for that unit not to improve slightly. North Texas won't go bowling in 2008, but they have a realistic shot of matching their win total in coach Dodge's first two seasons combined (3).

The team(s) you should be selling are...

Louisiana-Lafayette
If the Ragin Cajuns were going to find their way to the first bowl game in school history, last year was the year. A senior quarterback, running back, and wide receiver powered the Cajuns offense to the best mark in the conference. They even challenged BCS-member schools Illinois and Kansas State on the road, before losing by 3 and 8 respectively. The Cajuns did finish second in the conference and were bowl eligible for the third time in four seasons. However, Florida Atlantic was the second Sun Belt team selected for a bowl game, and the Cajuns had to spend the holidays at home. Now those three skill players are gone, and despite the fact that seven other starters return, including all five along the offensive line, it appears as if the offense will decline. The defense, far from a strength last season (sixth in the league), does return 9 starters, so they should see some improvement on that end. The Cajuns have been very competitive under head coach Ricky Bustle, and they may even flirt with bowl eligibility this season, but expecting them to match their five league wins from last season is foolhardy.

The team(s) you should be holding are...

Arkansas State and Western Kentucky
Last season started out looking as if it could be a great one for the Arkansas State Red Wolves. They upset Texas A&M (at College Station) in the opener and then waylayed IAA Texas Southern 83-10 in their second game. Unfortunately, the rest of the way, they managed only a 4-6 record, lost four of five road games (with the lone win coming against punching bag North Texas), and despite finishing bowl eligibile for the fourth time in head coach Steve Roberts' seven seasons, were left out of the postseason. The Red Wolves return a plethora of offensive talent from a unit that surprisingly under-achieved last season. The Red Wolves were fourth in the league in offense, but were pretty much on equal footing with North Texas in terms of moving the ball (averaged about 6 yards more per league game). Returning for 2009 are a senior quarterback (Corey Leonard), on track to become the school's all-time leading passer, a senior running back (Reggie Arnold), on track to become the school's all-time leading rusher, and last year's top two receivers (Brandon Thompkins and Jahbari McLennan). The Red Wolves do lose three starting offensive linemen, so improvement is not a given, but the glut of skill position players means its likely. On defense, the Red Wolves bring back 8 starters from a unit that was the second best in the league behind Troy. Led by defensive end Alex Carrington, who paced the team and the Sun Belt with 10.5 sacks last season, the Red Wolves should be strong on defense. The Red Wolves also get to host league favorite Troy, so why are they not a team to invest heavily in? Road games. The last five seasons, Arkansas State has posted a Sun Belt road record of 5-13, including 1-3 last season and 0-3 in 2007. If they can win on the road, this could be a special year for Arkansas State. 2009 marks Western Kentucky's maiden voyage in the Sun Belt and in IA football. They were a transitional member in 2007 and 2008, and the results thus far have not been so good. The Hilltoppers were 7-5 in 2007, but just 1-5 against IA foes. They were 2-10 last season, and did not beat a single IA school. The Hilltoppers played five Sun Belt opponents in 2008 and in those five games they averaged 300 yards per game and allowed an average of 393 yards per game. The offensive numbers would rank dead last in the Sun Belt, about 47 yards below Middle Tennessee State. The defensive numbers would have ranked a respectable sixth in the league, just behind Florida Atlantic. The offense, last season's weak link, returns the most starters (8), but must break in a new quarterback, while the defense returns only four starters. Western Kentucky has a history of succes at the IAA level, including winning the national title in 2002 and posting a winning record for 12 consecutive seasons prior to 2008. Whether they can continue that success in the Sun Belt conference remains to be seen.

Wednesday, June 24, 2009

MAC Review: 2005-2008

In addition to the SDPI posts, another offseason interest of mine has been to look at how each IA conference has shaken out since 2005. I chose 2005 as the starting point because that was the year the ACC expanded to 12 teams, the Big East added 3 new members and booted Temple, Conference USA added a championship game, the Mountain West added TCU, the WAC looted the Sun Belt after several of its members joined Conference USA, and the Sun Belt added 2 independents from Florida (FAU and FIU). Since 2005, there has only been one change in any IA conference. That was the expansion of the MAC by a single team with the addition of Temple. Only two more conferences left to analyze, and only about 10 more weeks until kickoff :) This week we move to the nation's heartland, and take a look at the Big 10's little brother, the Mid-American Conference. First here are the cummlative MAC standings since 2005.Not surprisingly, the team with two league championships in the past four seasons is number one overall. Butch Jones, who took over at Central Michigan, following Brian Kelly's departure to Cincinnati prior to the Chippewas bowl game in 2006, has actually exceeded Kelly's winning percentage in conference play. Kelly went 12-4 in the MAC in 2005 and 2006 (15-9 if we include his first year in 2004). Butch has guided the Chippewas to a 12-3 conference record the past two seasons. That brings me to my next point. You'll notice six teams have played 32 conference games and six have played 31 (Temple has only played 8 as they joined before the 2007 season). The MAC thought it would be a great idea to have some teams play seven conference games in 2007 and others play eight because the addition of Temple gave the league 13 teams. The league counted only games against division opponents when determining the representatives in the 2007 MAC Championship Game. As Murphy could have predicted, the MAC East champion in 2007 (Miami) was actually a half game behind Bowling Green because they played only 7 league games (5-2) while the Falcons played 8 (6-2). Each team went 4-2 within the division (as did Buffalo). Both schools beat the Bulls and Miami beat Bowling Green so they were crowned division champs. The league corrected itself in 2008, having each school play 8 total conference games and making life easier on statisticians everywhere. Another interesting fact is that Bowling Green, despite owning the fourth best record overall and top record in the MAC East over the past four seasons, did not 'win' the division a single time. The Falcons, in a move they will probably wind up regretting, fired their coach, Gregg Brandon, after last season. It just seems like a stupid thing to do, firing a coach who consistently had the team in contention to win the division (tied for 1st in 2005, tied for 1st in 2006, tied for 2nd in 2008). No, he wasn't as good as his predecessor (you may know him) but he only had one losing season in six years and the Falcons participated in three bowl games under his watch.

Now here is each team's home record in conference play since 2005.Western Michigan and Temple!? are tied for the best home league mark since 2005. I'll give the nod to the Broncos since they have played twice as many games. The biggest discrepancy between home and road record belongs to the Temple Owls and Akron Zips. Temple is a robust 6-2 at home against MAC foes since joining the league in 2007, but only 2-6 on the road for a difference of four games. Akron is a solid 9-7 at home since 2005, but only 5-11 on the road. The most interesting home/road split probably belongs to Ball State. The Cardinals are a ho-hum 9-7 at home, but a league-best 13-2 on the road since 2005 (difference of 4.5 games).

Now here is how homefield advantage shakes out in the MAC (in conference play only) with respect to the nation at large (with rank out of the 11 IA conferences in parentheses).With the exception of 2007, the MAC has found itself in the bottom half of the nation in terms of the value of homefield advantage. Twice the conference's home teams have finished with losing records (2005 and 2008). Overall, homefield has meant next to nothing in the aggregate in terms of wins and losses within the league.

Next up is how each MAC team stacks up offensively for each season. This is the ranking of yards per game in conference play.If there is one thing that is constant in the MAC, its change. Outside of Central Michigan, which has finished either first or second in offense each of the past four seasons, no other team has been consistent. Temple has finished dead last in both of their seasons in the league, but every other full-time member has seen their fortune shift dramatically at least once. Toledo and Northern Illinois are the prime examples of this phenomenon. Toledo owned the league's best offense in 2005, drooped to eighth in 2006, rebounded to second in 2007, and fell back to second-to-last in 2008. Northern Illinois was third in 2005, tops in the league in 2006, fell to ninth in 2007, and then dropped one spot further to tenth in 2008. If you look at Ball State's numbers, its pretty easy to tell when Nate Davis became the starting quarterback.

And finally, here are the defensive rankings for each team. This is yards allowed per game in conference play. Change has been the story on defense as well. Here are just a few examples. Akron had the best defense in the league during their title-winning 2005 season. They dropped to the middle of the pack in 2006, and fell even further in 2007. Western Michigan and Ohio featured the two worst defenses in the league in 2005. They had the two best in 2006. Ball State and Buffalo had the two worst defenses in 2006. In 2007, they were both in the top five. Miami had the best defense in 2007. They had the third worst in 2008. Northern Illinois had the second worst defense in 2007. They had the league's best in 2008. What kind of shenanigans are in store for 2009? Will Eastern Michigan become the standard against which all MAC defenses are measured? Will Northern Illinois go back in the crapper? Stay tuned.

Tuesday, June 16, 2009

MAC SDPI

One of my favorite set of posts from the past two offseasons has been the SDPI recap/early preview. Don't know what SDPI is? It stands for Standard Deviation Power Index and is a tool Eddie Epstein used in his book Dominance to rate pro football's best teams. The basic idea is to look at how far above or below average (by standard deviations) a specific team is relative to their conference brethren. Since each team plays the same number of conference games, it can give us a good idea about who the best team was within the conference. However, it cannot tell us which conference is better. But the purpose of these posts is not to determine which conference is superior, but rather project ahead which teams in a conference will be contenders or also rans in the upcoming season. In the first post on SDPI two years ago, I calculated SDPI based on points scored and allowed within conference play. Last season I used points scored and allowed as well as yards gained and allowed. This season, I'm sticking with yards only. The yardage version of SDPI has a better correlation with future performance than points, and including both last season made the post seem (at least to me) quite muddled. Of course, this is by no means, the end all be all rating system, but it can give us an idea of which teams will improve and decline in 2009. Only two more conferences left. Our penultimate review will be the Mid-American Conference.

If you want the meat of the article, skip this next paragraph as it just gives an example of how SDPI is calculated. The mean yardage for and against for all MAC teams in conference play (championship game excluded) was 30777.46 yards. The standard deviation for yards gained was 376.11. The standard deviation for yards allowed was 302.15. Kent State gained 3344 yards in conference play and allowed 3019. Their offensive SDPI was 0.71 = ([3344-3077.46]/376.11). Their defensive SDPI was 0.19 = ([3077.46-23019]/302.15). Their total SDPI was 0.90 which ranked 4th in the conference.

To refresh your memory, here are the 2008 MAC Standings.

Now here are the 2008 SDPI Standings sorted by total SDPI, with conference rank in offense, defense, and total SDPI in parentheses.
The big head-scratcher here involves the eventual league champion Buffalo Bulls. On a down-to-down basis, the Bulls were about the fourth best team in the MAC East. However, they created a ton of turnovers (MAC best turnover margin of +10 in conference play), completed a Hail Mary to beat Temple, and outlasted Bowling Green in overtime to win the division. In the MAC Championship Game, they capitalized on a spate of Ball State fumbleitis (ran two fumbles back for touchdowns) and won their first conference title as a IA school. Speaking of Ball State, they were clearly the class of the conference, riding the top offense and third best defense to an undefeated regular season, including a 22-point beating of the Big 10's Indiana.

Best Offense: Ball State 1.59
Prior to their bowl game meltodown against Tulsa, Ball State failed to gain at least 400 yards in any game just once, when Toledo held them to 355. Heck, they gouged Buffalo for 503 in their MAC Championship Game loss, but could not overcome four lost fumbles and an interception.

Worst Offense: Temple -1.58
The Temple offense was bad, but without quarterback Adam DiMichele it was awful. In the five conference games DiMichele played in, the Owls averaged 338 yards of offense (would have ranked 11th in the conference), but in the three he missed, they averaged 264 yards.

Best Defense: Northern Illinois 1.76
Ball State put a hurting on the Huskies (529 yards of offense) and Central Michigan moved the ball reasonably well (430 yards), but the Huskies held five of their six remaining conference opponents to under 300 yards.

Worst Defense: Eastern Michigan -1.48
This is sad since Eastern Michigan had an offense capable of getting them to their first bowl game since 1987 (fourth best in the league). No surprise that when they matched up with Central Michigan (second best offense and second worst defense in the conference) the final score was 56-52. The victor may have been a little surprising though.

Hardest Schedule (based on cumulative SDPI of opponents): Miami 2.70
The Redhawks faced the four best statistical teams in the MAC (Ball State, Bowling Green, Northern Illinois, and Kent State) along with the eventual league champion (Buffalo).

Easiest Schedule (based on cumulative SDPI of opponents): Bowling Green -2.95
The Falcons avoided the league's best team (Ball State) and had the good fortune of facing three of the bottom four teams (Eastern Michigan, Toledo, and Miami).

Entire Schedule Strength (hardest to easiest)
Miami 2.70
Toledo 2.21
Eastern Michigan 2.06
Kent State 1.11
Akron 0.81
Central Michigan 0.60
Northern Illinois 0.48
Western Michigan -0.22
Buffalo -1.11
Ohio -1.52
Temple -1.67
Ball State -2.50
Bowling Green -2.95

Looking ahead to next season, the prohibitive favorite should be...

East: Look to the bottom of last year's standings
Ohio won the MAC East in 2006 one season after finishing with a 3-5 conference record. Miami won the division in 2007 after a 2-6 campaign in 2006. Buffalo, last year's East champ, did have a winning record in the league in 2007, but prior to that, they had not won more than two conference games since joining the MAC in 1999. Keep an eye out for some combination of Akron, Ohio, Kent State, or Miami. I'd be inclined to give the nod to Ohio.

West: Central Michigan
After two straight MAC West titles (and two wins in the MAC Championship Game), the Chippewas ceded control of the division to Ball State last year. Those Cardinals lost their coach, quarterback, and four starters along the offensive line. Without those key pieces, its hard to envision a repeat. On the other hand, Central Michigan returns a three-year starter at quarterback, senior Dan LeFevour, from the league's second most prolific offense (Ball State was first). LeFevour has already thrown for the most yards in school history, and with a monstrous season, he could pass Timmy Chang for most yards of total offense in NCAA history (a little over 5200 behind). LeFevour is a dynamic dual-threat quarterback (a Tim Tebow light), having thrown for over 9000 yards and 74 touchdowns in his career to go along with over 2200 yards and 32 touchdowns on the ground. LeFevour has his top three receivers from last season (Antonio Brown, Bryan Anderson, and Kito Poblah) back to throw to. Neither topped 1000 yards, but each had at least 532. The only concern any Chippewa fan should have about the offense is that fact that three starters depart along the offensive line. Even if the line takes some time to develop, the plethora of talented skill position players should keep the Chippewa offense near the top of the league. The other side of the ball is where Central Michigan must improve if they have designs on winning the division. The Chippewas had the second worst defense in the league last year, and they ranked 106th nationally in pass efficiency defense. Of course, that kind of defense does not necessarily preclude a conference title as they finished 106th in the same category in 2007. The major difference was that in 2007, they grabbed 18 interceptions and last season they only managed 8. As mentioned ad nauseum on this blog, extreme performances are unlikely to be repeated, so there's a great chance Central Michigan manages more than 8 interceptions in 2009. The defense should grab more interceptions and should also be a little better as 10 starters (including the top-13 tacklers) return. If the defense can just become mediocre, Central Michigan should roll to the MAC title.

The team(s) you should be buying are...

Eastern Michigan and Ohio
Last season, perennial MAC East doormat Buffalo rose up and won the whole conference. Can the MAC West doormat (Eastern Michigan) do the same this season? Probably not, but there is reason to believe Eastern Michigan could be much improved and perhaps challenge for a bowl game in 2009. For starters, the offense, which was the fourth best in the conference returns nine starters, including four starters along the offensive line, last season's leading receiver (Jacory Stone), last season's leading rusher (Terrence Blevins), and quarterback Andy Schmitt. Schmitt had a pair of historic games to close 2008, throwing a combined 156 passes against Temple and Central Michigan for 1000 yards and 8 touchdowns. Eastern Michigan scored 108 combined points in those games and only won one! I wouldn't expect Schmitt to post such amazing numbers over the course of 2009, but the offense should once again be one of the best in the conference. Improvement on defense will determine how successful Eastern Michigan is. Last season, they were dead last in the league in yards allowed within the conference. The good news is seven starters return in 2009. The bad news is the four leading tacklers from 2008 are gone. The other good news is new head coach Ron English has a solid track record of coordinating good defenses (we'll give him a mulligan on last season's unit at Louisville thanks to all the injuries suffered by the Cardinals) and like Central Michigan, they should come up with a few more interceptions on defense. The Eagles nabbed only 8 interceptions last year and should see that number improve. The Eagles are a long way from fielding a good defense (even by MAC standards), but with a little luck could find themselves in a bowl game for the first time since 1987. The Ohio Bobcats followed up consecutive non-losing seasons under Frank Solich with a 4-8 campaign in 2008. They finished 4-8 despite the fact that they outgained their opponents (both in conference play and overall). The primary culprit in their poor record was a 2-4 mark in one-score games (though they were 2-2 in league play) and a turnover margin of -13 which ranked 113th in the nation. Extreme turnover margins are rarely duplicated so the Bobcats should see their margin trend back toward zero in 2009. They should also experience better fortune in one-score games. But most importantly, the Bobcats have a number of key contributors returning in 2009 that should have them in contention for a MAC East title. Senior Boo Jackson is the best quarterback the Bobcats have had under Solich. Jackson threw 19 touchdown passes last season (a school record) against just 12 interceptions. In Solich's first three seasons as coach (2005-2007) the team had combined to throw 31 touchdowns and 40 interceptions. If we only consider conference games, Jackson was even better, throwing 16 touchdowns and 7 interceptions. He also showed solid athleticism, rushing for 333 yards and two scores on the ground. Joining Jackson in the backfield are last season's two leading rushers Chris Garrett and Donte Harden. The pair combined for nearly 1000 yards on the ground last season (983) and with three starters back along the offensive line, should continue their solid production. Last season's two leading receivers are also back (Taylor Price and LaVon Brazill), meaning the offense should break into the top half of the league (they were eighth last season). The defense also brings back seven starters from a unit that ranked fourth in the league in yards allowed, including linebacker and leading tackler Noah Keller, a 2nd Team All-MAC selection in 2008. The defense should remain in the top half of the league, and coupled with improvements on offense, and a better turnover margin, the Bobcats should return to a bowl game for the second time under Solich.

The team(s) you should be selling are...

Buffalo
Let me preface this by saying that the job Turner Gil has done at Buffalo in three short seasons is phenomenal. However, a ton of good fortune went into last season's MAC title campaign. For starters, the Bulls recovered an amazing 2/3rds of their opponents fumbles (26 of 39), aiding them in their turnover margin of +20 (3rd best in the nation). Forcing fumbles is a skill, and the Bulls clearly swarmed to the ball in 2008. However, recovering fumbles is not a skill. Case in point, the Bulls recovered only 8 of their opponents 23 fumbles in 2007. There's a great chance the Bulls won't force as many turnovers in 2009, which means their defense will have to improve in forcing 3 and outs in order to successfully defend their conference title. The Bulls were ninth of 13 teams in yards allowed last season, hardly numbers befitting a league champion. Fortunately for the defense, the top-eight tacklers are back from last season, so some improvement is likely imminent. On the other side of the ball, the offense did not fare much better, finishing seventh in the league despite the presence of Drew Willy, the school's all-time leading passer, a four-year starter, and a free agent signee of the Baltimore Ravens. The Bulls still have plenty of talent on offense, including the leading rusher in school history (James Starks) and the second-leading receiver in school history who also happens to be a likely NFL draft choice (Naaman Roosevelt). However, with 3/5ths of the offensive line gone and a new starter at quarterback, the offense will be hard pressed to improve. Gil has done great things here, and the future is much brighter than it was when he was first hired. However, 2009 looks to be a season when the Bulls have to pay the piper for all the good fortune they experienced last year.

The team(s) you should be holding are...

Bowling Green and Western Michigan
Last season Bowling Green was statistically the best team in the MAC East, and second best team in the league overall. However, they were snakebitten in close games, going 1-4 in games decided by one score. Those four close losses each occurred in conference play, including one to Buffalo, which caused them to finish tied for second in the division. After the disappoinment of not winning the division, the administration got rid of head coach Gregg Brandon, who was 44-30 in six seasons at Bowling Green with only a single losing season. They then hired Tennesee offensive coordinator Dave Clawson to replace him. Keep in mind, Tennessee had one of the worst offenses in college football last season (116th in total offense and 111th in scoring offense). To be fair, Clawson has had success as a head coach before at Fordham and Richmond. He lead Fordham to one playoff appearance and three winning records in five seasons and he got Richmond into the playoffs twice in four seasons. Still, schools like Bowling Green should work hard to keep coaches like Gregg Brandon, not let them go. That being said, Bowling Green has some conflicting indicators going forward, I already mentoned their poor record in close games which saddled one of the league's better teams with a .500 record. However, the strength of last year's team was the defense, which ranked second in the conference. That unit returns only three starters in 2009 and loses the best pass rusher on the team, defensive end Diyral Briggs, who lead the team with 9.5 sacks and 13.5 tackles for loss last season. On offense, the team does bring back a senior quarterback (Tyler Sheehan), who has thrown 43 touchdowns against just 20 interceptions the past two seasons. However, the leading rusher (Anthony Turner) and leading receiver (Corey Partridge) from last year's team are gone. It's hard to get a good read on this team going into 2009. The coaching change and the loss of so many starters from the defense may make this a rebuilding year despite the likely improvement in luck. While Bowling Green could not buy a win in a close game, Western Michigan seemed unable to lose one. The Broncos won three MAC games by six points or less and also knocked off Illinois from the Big 10 by six. Those close wins pushed Western Michigan to a 9-4 record despite the fat that they did not do any one thing particularly well. The Broncos were fifth in the MAC in offense and seventh in defense. The offense returns a senior quarterback (Tim Hiller) who threw 36 touchdowns last season and has a shot at becoming the school's all-time leading passer with another good year. Four starters also return along the offensive line, but Hiller will be without the services of his two leading receivers from last season (Jamarko Simmons and Schneider Julien) who combined for 163 catches and nearly 2000 yards. The defense loses eight starters including seven of the top-ten tacklers, and will be breaking in an entirely new secondary. Bill Cubit is an outstanding head coach, Tim Hiller is one of the best quarterbacks in the league, and the Broncos host MAC West favorite Central Michigan, however, I'd be hesistant to predict them to win the division with all the losses on defense and the likely reversal in good fortune. In addition, a bowl game is not a certain proposition with games at Michigan, Indiana, and Michigan State adorning the non-conference ledger.

Tuesday, June 09, 2009

WAC Review: 2005-2008

In addition to the SDPI posts, another offseason interest of mine has been to look at how each IA conference has shaken out since 2005. I chose 2005 as the starting point because that was the year the ACC expanded to 12 teams, the Big East added 3 new members and booted Temple, Conference USA added a championship game, the Mountain West added TCU, the WAC looted the Sun Belt after several of its members joined Conference USA, and the Sun Belt added 2 independents from Florida (FAU and FIU). Since 2005, there has only been one change in any IA conference. That was the expansion of the MAC by a single team with the addition of Temple. This week, we'll go west and meet the Gonzaga of college football, Boise State. First here are the cummlative Western Athletic Conference standings since 2005.There's domination, and then there's what Boise State has done to the rest of the WAC over the past four seasons (heck since they've been in the league). The table below lists each team's cummulative scoring margin in conference play since 2005. That +789 number means Boise has outscored their WAC opponents by more than 24 and a half points per game over the past four season. You can see, both by the record and scoring margin, that the Fresno State program has taken a fall over the past four seasons. They have been usurped as second banana to Boise by both Hawaii and Nevada. Louisiana Tech has had an interesting roller coaster ride over the past four seasons. In 2005, they finished 6-2 in the league (7-4) overall, but did not receive a postseason invite. The following season, the wheels came off and Tech finished 1-7 in the league (tied for last with Utah State). Head coach Jack Bicknell was given a pink slip and the progeny of Vince Dooley (Derek) was hired to coach the team. Tech improved to 4-4 in WAC play in 2007. In their first conference game, they nearly derailed Hawaii's dream season before it got started, falling to the Warriors 45-44 in OT. Tech went 5-3 this past season and won their first bowl game since 1977.

Now here is each team's home record in conference play since 2005.Again, no surprise at the very top. Boise has not lost a home game to a WAC foe since joining the conference in 2001. In fact, Boise has not lost a conference home game since 1998, when they fell to fellow Big West member North Texas on October 10th. Elsewhere in the WAC, Hawaii comes in with the second best homefield advantage. This won't surprise many observers, as Hawaii has a reputation as a notoriously tough place to play. However, what may surprise folks, is that Hawaii has been nearly just as good away from the islands. Their 11-5 road record in the WAC is the second best mark over the past four seasons (behind only Boise's 14-2 record). One unique aspect of the WAC since 2005 is that no team has a better road record than home record. New Mexico State and Idaho both have the same putrid record in Las Cruces and Moscow as they do away from home (2-14 for the Aggies and 3-13 for the Vandals). The biggest discrepancy between home and road record belongs to the San Jose State Spartans. The Spartans are 9-7 at home and 6-10 on the road against league foes since 2005. Of course, this is all relative as no WAC team has performed significantly better at home as compared with the other conferences we've examined.

Now here is how homefield advantage shakes out in the WAC (in conference play only) with respect to the nation at large (with rank out of the 11 IA conferences in parentheses).After two seasons of slightly above average home performances, the WAC has slipped toward the bottom of IA in homefield advantage. Overall, the league ranks eighth since 2005. This is a sign that the better teams tend to win no matter where the game is played,

Next up is how each WAC team stacks up offensively for each season. This is the ranking of yards per game in conference play.Until 2008, the most consistent offense in the WAC belonged to the Hawaii Warriors. Think the departure of June Jones may have had something to do with their slide? On the mainland, the most consistently good offenses have been those at Boise and Reno. It's a little surprising that Boise has never lead the league in total offense. The most consistently bad offense has been that of Idaho. The Vandals have finished second to last in total offense each of the past four seasons. If you had taken the pulse of the New Mexico State program after the 2006 season, one would have thought it was quickening. Hal Mumme had just completed his second season in Las Cruces and the Aggies had finished behind only Hawaii and Boise in offense. Surely his air raid attack would wreck havoc on the rest of the conference. However, his charges dipped to the middle of the league in 2007, and then bottomed out in 2008, finishing ahead of only offensive luminaries Idaho and San Jose State. Not surprsingly, Mumme ball has been laid to rest. One can only hope Brendan Frasier can revive it.

And finally, here are the defensive rankings for each team. This is yards allowed per game in conference play.This is where Boise State has made their mark in the WAC. The national media likes to emphasize the skill position players and the high-octane offense Boise has run as the key to their success (not to mention their numerous gadget plays like those employed in the Fiesta Bowl win over Oklahoma), but the defense has been the dominant force in their foray through the WAC. They have boasted the league's best defense three years running, with a silver medal thrown in for good measure in 2005. They are the only team to have been in the top half of the league's defensive rankings each season. What's happened to Fresno State? After owning the league's best defense in 2005 (the year they gave Southern Cal all they could handle), the Bulldogs have been either mediocre or awful defensively. Dick Tomey, who popularized the 'Desert Swarm' defense during his time at Arizona, still knows a thing or two about that side of the ball. After a rough go of it in his first season (2005), his Spartans have been in the top four defensively each of the past three seasons. Also note the uptick in Hawaii's defensive ranking when Greg McMackin returned to the coordinator duties in 2007 (he became the head coach in 2008).

Tuesday, June 02, 2009

WAC SDPI

One of my favorite set of posts from the past two offseasons has been the SDPI recap/early preview. Don't know what SDPI is? It stands for Standard Deviation Power Index and is a tool Eddie Epstein used in his book Dominance to rate pro football's best teams. The basic idea is to look at how far above or below average (by standard deviations) a specific team is relative to their conference brethren. Since each team plays the same number of conference games, it can give us a good idea about who the best team was within the conference. However, it cannot tell us which conference is better. But the purpose of these posts is not to determine which conference is superior, but rather project ahead which teams in a conference will be contenders or also rans in the upcoming season. In the first post on SDPI two years ago, I calculated SDPI based on points scored and allowed within conference play. Last season I used points scored and allowed as well as yards gained and allowed. This season, I'm sticking with yards only. The yardage version of SDPI has a better correlation with future performance than points, and including both last season made the post seem (at least to me) quite muddled. Of course, this is by no means, the end all be all rating system, but it can give us an idea of which teams will improve and decline in 2009. This week, we'll examine the Western Athletic Conference, home to the Gonzaga of football, the Boise State Broncos.

If you want the meat of the article, skip this next paragraph as it just gives an example of how SDPI is calculated. The mean yardage for and against for all WAC teams in conference play was 2986.44 yards. The standard deviation for yards gained was 610. The standard deviation for yards allowed was 384.39. Louisiana Tech gained 3091 yards in conference play and allowed 3002. Their offensive SDPI was 0.17 = ([3091-2986.44]/610). Their defensive SDPI was -0.04 = ([2986.44-3002]/384.39). Their total SDPI was 0.13 which ranked 3rd in the conference.

To refresh your memory, here are the 2008 WAC Standings.


Now here are the 2008 SDPI Standings sorted by total SDPI, with conference rank in offense, defense, and total SDPI in parentheses.


As you can see, Boise State thoroughly dominated the league. The difference in their SDPI and that of second place Nevada (1.64) was greater than the difference between third place Louisiana Tech and eighth place Nex Mexico State (1.07). Outside of the top of the league (Boise and Nevada) and the very bottom (Idaho), the league's members were bunched very tightly last season. That means minor improvement from teams like New Mexico State or Utah State, coupled with some regression from teams like Louisiana Tech or Hawaii could turn the conference upside down in 2009.

Best Offense: Nevada 1.71
The 'Pistol' offense ran roughshod over the WAC last season. Nevada gained at least 500 yards of offense against six of their eight conference foes, failing to do so only against Boise State (385) and Hawaii (481).

Worst Offense: San Jose State -1.50
The worst part for the Spartans is that the terrible offense hamstrung the second best defensive unit in the WAC. San Jose State managed to finish 6-6, despite featuring an offense that was held under 300 yards in six of their eight conference games. If not for Herculean efforts (by their standards) against Utah State and Idaho (802 combined yards), the Spartans would have barely averaged over 200 yards of offense in the conference.

Best Defense: Boise State 1.84
The Broncos boasted the league's best offense by a mile, finishing a whole standard deviation above second place San Jose State. No WAC foe gained more than 385 yards against the Broncos, and the only teams to top 400 yards were a pair of top-10 finishers (Oregon and TCU).

Worst Defense: Idaho -1.93
Idaho was more than one and a half standard deviations below the second worst WAC defense (New Mexico State). Yardage wise, they allowed about 80 more yards per game in conference play than the Aggies.

Looking ahead to next season, the prohibitive favorite should be...

Boise State
No surprises here. Boise has won or shared six of the eight WAC titles that have been awarded since joining the league in 2001. They have never lost more than two conference games in any season, and have never fallen to a conference foe at home. Quarterback Kellen Moore, an inexperienced freshman last season, is now a seasoned veteran with a win at Oregon and an undefeated WAC championship on his resume. The backfield does lose Fiesta Bowl hero Ian Johnson, but Johnson split time last season with Jeremy Avery in a committee running game, so there should be no considerable dropoff there. The Broncos will be without the services of receiver Jeremy Childs who entered the NFL draft with one year of eligibility remaining, but was not selected. The defense also loses two of its top-three tacklers in defensive back Ellis Powers and linebacker Kyle Gingg. Unfortunately though, if you're looking for intrigue in the WAC in 2009, it won't be coming at the top of the conference. Look to the middle and you'll be in luck.

The team(s) you should be buying are...

Nevada and Utah State
Statistically, Nevada was the second best team in the WAC last season. By a large margin. Unfortunately, they lost some some close games (1-3 in one-score games against conference foes), had a hard time falling on loose balls (recovered only a third of all fumbles in their games) and played pretty poorly on special teams (63rd nationally in opponents punt return average and 115th in opponents kickoff return average). Those three factors conspired to cause the Wolfpack to finish tied for second in the league standings with two other teams (Hawaii and Louisiana Tech) that they significantly outplayed on a down to down basis. For 2009, Nevada returns the bulk of their arsenal that led the conference in offense. Quarterback Colin Kaepernick maintained his stellar play in his sophomore campaign, throwing for nearly 3000 yards and 22 touchdowns, while adding over 1100 yards and 17 touchdowns on the ground. Joining Kaepernick in the 'Pistol' backfield is the best running back you've never heard of, Vai Taua. The vowel-intensive Tuau rushed for over 1500 yards last season and averaged nearly six and a half yards per carry. Kaepernick will be without his favorite target, receiver Marko Mitchell, who gained 1141 yards last season. However, the 'Pistol' offense is predicated on the run, and despite the loss of center Dominic Green, the tandem of Kaepernick and Taua should keep the chains moving. The real question for the Wolfpack is whether they can stop anyone on defense. Last season Nevada finished sixth in the WAC in defense, but remember this is the WAC, not the SEC. Nationally, Nevada was 86th in pass efficiency defense and 120th (or dead last) in passing yards allowed per game. The defense returns eight starters, so the hope is that with experience comes at least marginal improvement. The Pack must play Boise on the Smurf Turf, but their other conference road games are against Utah State, San Jose State, and New Mexico State. I don't think Nevada can win the conference crown, but the potential exists for them to win their seven conference games not involving Boise. Nevada was a little on the unlucky side last season with the close losses and fumble recovery issues, so if the defense improves, and their luck reverts to normal, a double-digit win season could be in the coffers. Meanwhile, in Logan, Utah, 2009 is a season of new beginnings. The Aggies have not participated in a bowl game since December of 1997 when they lost the Humanitarian Bowl to Cincinnati. New coach Gary Andersen comes to the Aggies after being the defensive coordinator of the Utah Utes for the past five seasons. Utah State was far from a good team last season, but statistically they were better than Fresno State, a team that appeared in a bowl game, and San Jose State, a team that finished bowl-eligible. The road to the postseason for Utah State may not be a pipe dream in 2009. Utah State does not have the offensive talent to match the upper-eschelon WAC schools like Boise State and Nevada, but they do have a player to build around in junior quarterback Diondre Borel. In his first season as a starter, Borel led the team in both passing and rushing yards. He didn't put up great numbers, but he did throw more touchdowns (11) than interceptions (10) and added five scores on the ground. Outside of leading receiver Otis Nelson (only 455 receiving yards) every other significant contributor in the passing game is back. It wouldn't be a total shock if Borel improved in his second season in the lineup, and keep in mind, the Aggies were only a shade below average offensively (in the conference) last season. The defense, Andersen's area of expertise, gets eight starters back, including leading tackler, linebacker Jake Hutton. Any marginal improvement on that side of the ball will give the Aggies an above-average WAC defense, something that has been unheard of for a long time. The non-conference schedule is loaded with road trips to Utah, BYU, and Texas A&M, so the heavy lifting to get to six wins and bowl eligibility will have to be done in conference play. The Aggies are certainly not favored to go bowling in 2009, but I think their odds are better than most prognosticators would give them.

The team(s) you should be selling are...

Louisiana Tech
2008 was a banner year for the Bulldogs. Under the guidance of second-year head coach Derek Dooley, Tech finished tied for second in the WAC and won their first bowl game since 1977. And while statistically the Bulldogs were the third best team in the WAC, they were much closer to the eighth best team (New Mexico State) in terms of performance than they were to the second best team (Nevada). A lot of the good that befell Tech in 2008 is unlikely to be repeated in 2009. Tech was 5-2 in one-score games (3-1 against conference foes), which is a big reason they were able to win eight games despite outscoring their opponents by only 12 points over the course of the season. They also recovered a high percentage of fumbles in their games (58%) and scored five more non-offensive touchdowns (scores by special teams or defense) than their opponents. Tech brings back a host of starters (ten by most estimations), with the lone loss receiver Philip Beck, from an offense that ranked third in the WAC last season. However, that ranking is a little misleading, as the four teams bunched around the middle offensively (Louisiana Tech, Fresno State, Utah State, and Hawaii) were only separated by about 25 yards per game. The Tech offense may improve, but that improvement will likely come on the ground. Quarterback Ross Jenkins, who took over for Taylor Bennett a little less than halfway through the season completed only 52.9% of his passes (98th nationally among qualifying players) and finished with a passer rating of 118.46 (77th nationally). This despite playing in a league not known for its defense. On defense, the Bulldogs bring back seven starters and could improve upon their fourth place finish in that category last season. However, once again those teams that finished between third and eighth defensively were bunched very closely together (separated by about 16 yards per game). Louisiana Tech may improve upon their performance from last season, but their luck will probably be a little worse, and a non-conference schedule that includes trips to Auburn, Navy, and LSU will likely keep them out of a bowl game this season.

The team(s) you should be holding are...

Fresno State
2008 was supposed to be the year Fresno State broke through and won the WAC. They brought back a senior quarterback and tight end (Tom Brandstater and Bear Pascoe respectively) who ended up being taken in the 2009 NFL draft. They also brought back eight starters on defense. The lone fly in the ointment was a season-ending trip to Boise State, where the Broncos have yet to lose to a WAC foe. And yet it was not to be. The season began well enough with an upset win at Rutgers on Labor Day. They then gave Wisconsin all they could handle in a 13-10 loss at home. The next week showed the first real signs that maybe Fresno was not all they were cracked up to be. It took the Bulldogs two overtime periods to knock off a Toledo team that would finish 3-9. Worse yet, that Toledo team that averaged 334 yards per game on the season (84th nationally), gouged the Bulldogs for 598 total yards. Fresno also had a closer than expected tussle with UCLA the following week (won 36-31), but entered WAC play with a 3-1 non-conference record. Alas, Fresno would go on to lose four more times in the league. Some were close (Hawaii and Louisiana Tech edged the Bulldogs by three points apiece), and some were not (Boise State ripped the Bulldogs by 51). Fresno then went on to lose the New Mexico Bowl to Colorado State. Statistically, the team thought by many to be one of the best in the conference, was ranked sixth (below Utah State). Now Brandstater and Pascoe are gone from an offense that was already below average by WAC standards. The defense does return nine starters, so there is hope that for improvement there. Unfortunately, the non-conference slate once again includes three road dates with likely bowl teams from BCS leagues (Wisconsin, Cincinnati, and Illinois). If the Bulldogs have a notion of returning to the postseason, they must sweep every home game save the once against Boise.