Jump to content
Sign in to follow this  
  • entries
    78
  • comments
    48
  • views
    34,742

About this blog

A sports analysis blog hosted by lovable, furry old Duquesne. Come here for DUSHEE missives and other sundry wonkish thoughts.

Entries in this blog

 

SEC Fatigue

Let's get it out of the way. The SEC is the best. I'm not even going to try to dispute it as much as I'd like to. I am as sick of them as anybody else who doesn't live within 30 miles of a Civil War battlefield. The bona fides are not in question. Every Mythical National Championship since 2006. Only conference in history to win more than 3 in a row. By far the most NFL draft picks. And you can't avoid the superlatives. Steve Spurrier, albeit for transparently self-serving reasons, absurdly declared Alabama better than the Jacksonville Jaguars. ESPN, also for transparently self-serving reasons, touting the SEC even when it's ostensibly attacking the conference. As with any media darling, the SEC has legions of backlash haytas loaded for bare with reasons to diminish their success. They rarely play out-of-conference games outside the South. Outside the MNC and BCS games over the last 14 years, the SEC is a slightly more pedestrian 56-42. They haven't produced a decent NFL quarterback since Eli Manning. There are few quantitative measures by which one can conclude that the SEC has not been, on average, the best conference in college football. But the flip side of that coin is that the bias toward the SEC has become so strong that it seems as though the 110 college FBS teams who play in conferences not named the SEC have access to only one slot in the MNC, and even then, they still might not get the nod over the second place SEC team. That the SEC has won seven straight MNC games is a truly epic feat, but it is a feat whose accomplishment is owed as much to opportunity as it is superior play or talent. Getting to the game is the important part. Two of the best teams in the country winning a single game against the other is generally a statistical coin flip. Has the SEC won all these national championships because they clearly had the best team in the game or have they managed the statistically improbable feat of flipping heads 7 times in a row? Does the SEC Champion deserve a de facto berth into the MNC game? The question is probably moot until somebody beats them in the MNC. But the numbers, nonetheless, tell an interesting story. What Is the Best Conference? Before we dive into the details, how do we go about determining the best conference? The conference with the best singular team? The conference best at the top -- e.g., most teams in the top 25? The best conference top-to-bottom? The answer, usually, is whatever metric that supports whatever argument you want to support. Here at NMMH, we strive to avoid such tautologies. The first metric -- best singular team -- seems a poor metric for judging the strength of a conference. By that metric, the WAC was in the argument for best conference in the country during the Kellen Moore era at Boise State. But it's not a metric to be dismissed because it is, after all, the two "best" teams in the country who are supposed to be chosen for the MNC. And winning the MNC is the primary (but not the only) basis for extolling SEC dominance. As we discussed in the last blog post, the example of SMU and Georgia in the 1982 season showed that being undefeated in a "major" conference may not be sufficient justification for declaring a team the "best." A good, but not great, team can go undefeated based on the good fortune, or the uncanny capability if you choose to look at it that way, to put their best efforts out against the best opponents on the schedule. The turds that SMU laid against TCU and Texas Tech in 1982 are no less indicative of the kind of team they were than the impressive wins against Texas and Pitt. Had they executed against Texas and Pitt the way they executed against TCU and Tech, they not only would not have been undefeated, but they would have been blown out. Did they get lucky, or did they "play up" to their competition? So for the sake of argument, let's say that the top 5 teams in the DUSHEE poll are all teams that should have, at minimum, been in the conversation for best team in the country. Here are the DUSHEE top 5 for each year of the SEC's run, the teams in bold were selected to play in the MNC: 2006 LSU SEC 22.29 177.67 72.78 Ohio St. B10 24.69 133.22 71.60 Louisville BE 22.12 137.15 66.66 Florida SEC 20.10 160.33 65.64 BYU MWC 20.76 145.39 64.92 2007 West Virginia BE 25.04 167.29 77.17 LSU SEC 22.95 170.42 73.15 Oklahoma B12 24.60 135.54 71.72 Ohio St. B10 20.87 185.48 70.83 Missouri B12 22.46 148.11 68.94 2008 Florida SEC 35.59 184.94 102.21 USC P10 31.19 247.26 101.64 Oklahoma B12 32.67 207.85 99.21 Texas B12 28.67 157.01 83.46 Penn St. B10 25.52 178.00 79.72 2009 TCU MWC 27.12 222.03 89.36 Texas B12 26.45 205.00 85.52 Florida SEC 24.91 212.98 83.36 Alabama SEC 25.85 192.02 82.40 Boise St. WAC 24.97 157.39 75.63 2010 Boise St. WAC 32.41 251.22 104.80 TCU MWC 26.84 238.90 91.16 Ohio St. B10 25.02 209.34 83.09 Oregon P10 25.04 154.95 75.43 Auburn SEC 23.33 174.17 74.50 2011 Alabama SEC 31.50 269.74 105.48 LSU SEC 30.53 129.87 83.60 Boise St. MWC 25.52 163.45 77.65 Houston CUSA 22.46 182.74 73.86 Oregon P10 24.06 154.10 73.22 2012 Alabama SEC 31.13 217.62 97.30 Texas A&M SEC 26.30 201.68 84.73 Oregon P10 29.20 155.42 84.38 Florida St. ACC 19.33 178.66 66.59 Notre Dame Ind 17.89 144.12 58.62 Table 1. Top 5 DUSHEE teams from 2006-2012. From left to right: Team, Conference, PD, YD, and DUSHEE Score. DUSHEE Score is normalized against the average best team from 2000 through 2012. Bold teams played in MNC game. Ignore for a moment the DUSHEE WTF team of 2011 (I'd like to think Kevin Sumlin had Houston's DUSHEE ranking in hand when he interviewed for the A&M gig) and look at the conference affiliations of these 35 teams: SEC: 11 B12: 5 B10: 4 P12: 4 MWC: 4 BE: 2 WAC: 2 ACC: 1 CUSA: 1 Ind: 1 Table 2. The number of DUSHEE Top 5 teams per conference affiliation over the last 7 years The SEC has had 11 of the 35 Top 5 DUSHEE teams of the last 7 years, or 31%. Yet they have been given 8 of the 14 MNC game slots, or 57%. If we focus the analysis down to the top 2, the SEC has accounted for 7 of the 14 Top 2 DUSHEE teams, but 4 of those 7 came in the last two years. In the 5 years prior, the SEC accounted for 3 of 10 Top 2 teams. So, while the SEC has clearly had the most consistent presence in the Elite in college football over the last 7 years, it is also arguable that they have nonetheless gotten disproportionate opportunity to play in and win the MNC at the expense of other conferences. Notably, the MWC, which has had 4 teams in the top 5 (BYU, TCU (twice), and Boise) over the last 7 years, hasn't had a single opportunity in the MNC. The SEC has clearly deserved more opportunities, but one might argue that the growing bias toward them as somehow a "super" conference has given them even more opportunities than they've deserved. The Middle of the Road Perhaps the more relevant way to judge conference strength is to look at the conference from top-to-bottom. There are a few ways to assess this. The typical way a computer rating system does it is by averaging the scores of all the teams in the conference together and using a single number to determine conference strength. Let's begin with this customary conference rating approach. Again, over the last decade plus, here are the conferences ranked by average DUSHEE score of all teams in their conference: Figure 1. Three-Year moving average DUSHEE score per conference from 2001-2012. The moving average is plotted for the middle year. Figure 1 shows the 3-year moving average for each conference since 2001. A three-year moving average means that each point on the plot represents the average of the three years surrounding the point; for example, the 2011 SEC point represents the average DUSHEE score for 2010, 2011, and 2012 combined. The 2010 point combines 2009, 2010, and 2011. And so on. The reason for doing this is as busy as Figure 1 is, it's even busier if you plot each individual year by itself. Moving averages smooth out the data and make it a little easier to discern multi-year trends. Notice that the SEC curve remains on top all the way back to 2001. However, the AQ (automatic qualifier, in BCS parlance) conferences were much more tightly packed together in the early aughts. Based on the DUSHEE metric, the average team in the Big Six conferences were much more on par with each other in the early 2000's than they have been more recently. In fact, the average performance of all of the AQ conferences, after steadily rising for the most part in the early 2000's, have moved down over the last 4-5 years, EXCEPT for the SEC. This seems to indicate that the SEC dominance of the last few years hasn't been as much about their improvement as other conferences' decline. You'll also note our old stomping grounds, the MWC. Through 2008-2009, it was pretty close to on par with the ACC and in some years the highly volatile Big East. Not surprisingly, the MWC's performance has taken a significant dive after the loss of their big 3. The addition of Boise just hasn't been enough to replace three perennial top 25 teams. The MWC's decline has resulted in a convergence in the non-AQ conference performance, making the distinction between the haves and have-nots more obvious than ever. We're all happy to have been Indiana Jones rolling under the closing crypt door just in time but the downside may have truly been the death of the non-AQ conference as a serious BCS threat to the Elite. Only as Strong as Your Weakest Link Averaging conference performance is a perfectly fine way of assessing strength, but by boiling the data down to a single number, some information inevitably gets thrown away. For instance, say Conference A has 10 teams. Three teams have DUSHEE scores of 70, three are around 0, and the other four are all around -50. On average, Conference A has a DUSHEE score around 0 ... the definition of mediocre. Yet, they have three teams who are BCS, if not MNC worthy. Is that a mediocre conference? Meanwhile Conference B also has 10 teams, evenly distributed between 40 and 0. Nobody is terrible, and on average Conference B has a DUSHEE score around 20. Significantly higher than Conference A. Yet Conference A has three teams that could comfortably win Conference B. Which is the better conference? Conference B has the tougher "week in -- week out" grind. Yet none of those teams in Conference B play two games nearly as tough as the two games the top 3 of Conference A play against each other. This generalization, as I alluded to a few paragraphs above, somewhat describes the situation between the "Big 3" era MWC and the ACC and Big East. Figures 2, 3, 4, 5 and 6 show the 3-year moving average DUSHEE scores for the SEC, Big 12, Big East, ACC, and MWC, respectively. Figure 2. SEC 3-year moving average DUSHEE score. Note Missouri and A&M haven't been in the conference long enough to make a 3-year average. Figure 3. Big XII 3-year moving average DUSHEE score. Note WVU and TCU haven't been in the conference long enough to make a 3-year average. Figure 4. Big East 3-year moving average DUSHEE score. Note Temple hasn't been in the conference long enough the second go around to make a 3-year average. Figure 5. ACC 3-year moving average DUSHEE score. Figure 6. MWC 3-year moving average DUSHEE score. Note Boise, Fresno, Hawaii, and Nevada haven't been in the conference long enough to make a 3-year average. The tops of the SEC (Fig. 2) and Big XII (Fig. 3) don't look qualitatively different. For the most part, both conferences have been dominated by 2-3 teams (SEC -- Alabama, LSU and Florida; Big XII -- Oklahoma and Texas, with brief challenges to supremacy by Kansas State and Oklahoma State). Moving down to that next tier, a little separation begins to emerge. The SEC has continually had a dense grouping of teams in the 20-40 range, keeping in mind that a DUSHEE score of around 30 means a team is at least on the cusp of being a Top 25 team. The Big XII's densest grouping of teams typically appears in the 0-20 range. And then at the very bottom, the distinction is even more clear. While both conferences have typically had 4-or-so below average teams (usually Kentucky, Vanderbilt, Mississippi State and Ole Miss in the SEC; Baylor, Kansas, Colorado and Iowa State in the Big XII), the SEC have not had what might be technically called God-awful teams as Baylor and Kansas were for parts of the last decade. The worst SEC team has been typically at about a DUSHEE score of -40. Baylor and Kansas have both spent time in the -60 to -70 range. The comparison between the ACC, Big East, and MWC also sheds some light on the ambiguity in assessing conference strength. Going back to my Conference A vs. Conference B scenario, the MWC (Fig. 6) for most of the last decade was a pretty good analog for Conference A. TCU, Utah, and BYU never dipped below a DUSHEE score of 20 and all three were typically at least flirting with 40. TCU's moving average during that time places it squarely in the Alabama/Oklahoma stratosphere. At the beginning of the last decade, before TCU joined, the MWC had a healthy grouping in that same 0-20 category that a lot of Big XII teams have held. However, both Utah and BYU were bunched there too; the MWC didn't start producing consistent Top 25 level teams until 2004, when Utah began to emerge ahead of the peloton. At about that same time, the steady decline of the bottom of the MWC begins. Colorado State, New Mexico, UNLV, and San Diego State all get bad, really bad, coincidental to the emergence of Utah, the re-emergence of BYU, and the entry of TCU into the league. By 2009-2010, the MWC has the rare distinction of containing three consistent Top 25 teams along with three consistent Bottom 25 teams, all within a 9 team conference. Compare that to the Big East and ACC during this time. Prior to leaving the Big East (Fig. 4), Miami and Virginia Tech were miles above the rest of the conference in terms of performance. Miami was winning MNC's and Virginia Tech was playing in BCS games; nobody else in the conference was in the Top 25. However, the decline that they had begun to show in the early part of the decade continued after they joined the ACC (Fig. 5), where they joined a declining Florida State in a conference whose champion generally found itself well outside the Top 10 and sometimes toward the bottom of the Top 25. Outside of the emergence of West Virginia and the initial success of Louisville in the few years immediately after they moved from CUSA (somehow managing to thrive in the week-in-week-out grind of an AQ conference), the Big East champion also tended to sit in that 15-25 spot in the AP (and DUSHEE) poll. Utah and BYU could have regularly challenged for championships in either of those two conferences and the Dalton-era TCU teams would have most likely won either of those conferences going away (much as they did in the MWC). The difference is that the bottom of both the Big East and ACC were not nearly as bad as the bottom of the MWC. The ACC has had only one perpetual resident in the less than -40 realm of the God-awful, Duke. The Big East had two roaming that wasteland, albeit at different times; Rutgers at the beginning of the decade and Syracuse at the end. The end of the Big East found the entire conference tightly bunched between -10 and 20, making it by far the most evenly balanced conference during the DUSHEE-analyzed era. None of the teams are terrible. And none of the teams are really good. The ACC in recent years has told a similar story. Most of their teams have been bunched between -20 and 40. One or two decent teams, one or two pretty bad teams, and everyone else varying shades of mediocre. So which was the "better" conference of the three? Based on average performance, the ACC and Big East were a little better, largely because their bottom halves weren't nearly as low. Conversely, their top 2-3 teams weren't nearly as good. Is it better to be consistently mediocre or have a few great teams along with several legitimately terrible teams? The Numbers, Unfortunately, Tell You What You Already Know The SEC has been the best conference in college football over the last decade-plus, based on almost any quantitative measure you wish to throw at it. But the growing notion that the SEC is somehow in a league of their own, a league that should receive an automatic berth into the MNC game by virtue of winning the conference is not supported by the numbers. The SEC isn't THAT much better, and particularly not at the beginning of this run. The Cam Newton Auburn team of 2010 was a really good team, but there were 4-5 other teams who had a claim to be worthy of an opportunity to play in a MNC. As many on this board feel, I'd have loved to have seen the Frogs get an opportunity to play them. Thank the Football Gods we're finally going to get to see some more teams with an opportunity to play. For as long as the SEC champion gets an automatic bid into the MNC, they're going to continue to win lots of championships. At least now, they're going to have to get that coin to flip heads at least twice as many times.

Duquesne Frog

Duquesne Frog

 

Makin' Your Way in the World Today, Takes Everything You've Got ...

Bill Clements' Pride and Joy and Must See TV Yesterday, some network (TVLand?) ran a Cheers! marathon starting with the first ever episode which aired on September 29, 1982. On that Wednesday, the American TV audience, one limited to an unthinkable half dozen TV channels from which to choose, stood up from their couches, walked over to the TV, and turned a knob to whatever number corresponded to their NBC affiliate station, and "Must See TV" was born. On the Saturday prior, SMU's Pony Express, ranked 6th in the country at the time and on their way to a SWC title and number 2 ranking in the AP at season's end, beat TCU in Texas Stadium by a score of 16-13. This was a TCU team still a few years away from the Unbelieeeeevable! Jim Wacker-led team of 1984; it was in fact, what would be 3-8 TCU team, in F.A. Dry's last season as coach, with wins over Utah State, Rice and Baylor. This wasn't SMU's only less-than-impressive victory on the year. SMU only beat Baylor by 3, a team TCU beat by 24. They beat a .500 Houston team by 6 at home. They beat a 4-7 Texas Tech team by 7. Yes, they won, but these were mediocre-to-bad teams in a mediocre conference. The only genuinely good teams in the conference and on SMU's schedule (their non-conference schedule consisted of Tulane, UTEP, and North Texas State, all of whom were about as good then as they are now) were Texas and Arkansas. And SMU did beat Texas by two scores, 31-17, (although they were outgained by Texas slightly) and tied Arkansas, 17-17 (again, outgained slightly). Cause They're Still Preoccupied With 19 ... 19 ... 1982 This was a pretty fascinating year in college football, and the stuff of a lot of future ESPN 30-for-30 documentaries. This was the season before the famous Elway-Marino draft of 1983. Herschel Walker won the Heisman at Georgia. Auburn had some freshman kid named Bo lead their team in rushing. Howard Schnellenberger, in his 4th year at Miami, was one year away from launching the next college football dynasty. Marcus Dupree was leading college football in yards per carry at OU and would blow out his knee the next season. Schnellenberger's starting quarterback was current Georgia coach Mark Richt (Jim Kelly spent much of the year hurt and Vinny Testaverde was 4th string). Fresno State's starting quarterback was recently fired Cal coach Jeff Tedford who was throwing to Henry Ellard. Rick Neuheisel was backing up at UCLA. Steve Young was calling the signals in Provo. Doug Flutie was starting for Boston College, still two years away from throwing the most famous Hail Mary in college football history. Randall Cunningham and Boomer Esiason were calling the plays for UNLV and Maryland, respectively. The game was played quite a bit differently then as well. Despite the much vaunted class of quarterbacks who were drafted that following spring, only 5 threw for more than 3000 yards in 1982: Todd Dillon, Long Beach State, 3517 (yes, they were D1-A then); Tony Eason, Illinois, 3248; John Elway, Stanford, 3242; Steve Young, BYU, 3100; and Ben Bennett, Duke, 3033 yards. Only three receivers had 1000 yards: Ellard, 1510; Mark Clayton, Louisville, 1112; and Darral Hambrick of UNLV, 1060. In 2012, thirty-eight quarterbacks threw for over 3000 yards and thirty-six receivers had 1000 or more yards. The landscape was quite a bit different as well. Obviously the SWC existed and Arkansas was still there. There were twenty-five Independent D1-A programs including most of what would become the Big East and CUSA. The ACC only had 7 teams. The Big 8 had, well, 8. The Big 10 had ... 10. This was the era when the number in an official conference designation actually meant something. The SEC only had 10 teams. And the "minor" conferences were completely different. The MAC still existed, albeit with only 10 teams. The WAC was the original WAC before it sorta became the Mountain West by a process that most of us remember all to well. The other two conferences were the Missouri Valley Conference, dominated by Tulsa, but also including Wichita State, West Texas A&M, and New Mexico State, among others, and the Pacific Coast Athletic Association, which would morph into the Big West Conference and included at the time Fresno, the Jack Elway led San Jose State (who beat his son's Stanford team that year), UNLV, and others. Evaluating 1982 At the end of the season, the nation rejoiced as living legend and man of unassailable character, Joe Paterno, won his first national championship in his 17th season at 55 years of age. Going into January 1 (back in the stone ages when the season ended on New Years Day), undefeated and #1 Georgia was playing 1-loss #2 Penn State in the Sugar Bowl. #3 Nebraska, who Penn State had beaten 27-24 in State College in week 4, was 10-1 and in the Orange Bowl against 13th ranked 8-2-1 LSU. Penn State went on to beat Georgia 27-23 to claim the MNC. However, based on the DUSHEE metric, Nebraska was far and away the best team in college football that year. Here is the DUSHEE top-25 for 1982: Rk Team PD YD Score 1 Nebraska 29.87 253.14 100.00 2 Penn St. 23.93 103.87 67.21 3 North Carolina 18.11 202.49 67.02 4 Florida St. 19.95 149.10 64.18 5 Pittsburgh 17.87 149.38 59.56 6 Texas 17.91 139.10 58.31 7 UCLA 17.52 115.82 54.39 8 USC 17.88 108.35 54.23 9 LSU 16.42 124.80 53.09 10 Maryland 15.76 116.93 50.58 11 Oklahoma 13.76 138.22 48.87 12 Arkansas 14.40 125.99 48.73 13 Arizona St. 14.68 118.28 48.35 14 BYU 14.35 122.58 48.16 15 Alabama 14.60 106.68 46.64 16 Georgia 17.04 33.05 42.52 17 Ohio St. 13.93 76.66 41.23 18 SMU 13.54 66.86 39.08 19 Southern Miss 12.93 77.34 39.07 20 Washington 12.87 76.73 38.86 21 Michigan 14.33 49.06 38.53 22 West Virginia 12.35 76.24 37.64 23 Florida 10.87 83.26 35.24 24 Illinois 9.88 81.26 32.75 25 Clemson 13.26 20.42 32.39 Table 1: Top 25 in DUSHEE score in 1982 A few things jump out about these numbers that we'll discuss further: 1) Why is Nebraska so much higher than Penn State when Penn State beat them? 2) 4-loss North Carolina? WTF, DUSHEE? 3) Why are Georgia and SMU so low? 1) Nebraska-Penn State As mentioned above, Penn State and Nebraska played on the field and Penn State won. That's all that needs to be said right? Penn State was better. This is sort of the paradox of using a scheme like this to evaluate performance. Ultimately, the win is all that matters. And in determining who should play for a national championship, what happened on the field probably is all that matters. But the question of "who was better" is more nuanced. Both teams played 10 other games prior to their bowl game. For each team, their game against each was only roughly 8% of what they did that season. All their other games contributed the other 92%. Penn State lost a game as well ... to Alabama. Alabama wasn't a bad team in 1982; they wound up 8-4 and unranked (although 15 in DUSHEE). But Penn State lost 42-21 to a team not nearly as good by every quantitative standard you wish to use as Nebraska's loss. And then there was the rest of the season. Seven of Nebraska's 10 wins on the year were among the top 5 DUSHEE rated performances for the week in which the game was played. Penn State had 2 wins (including Nebraska) rank that high in a given week. In other words, on a week-to-week basis, Nebraska was playing on a more consistent, high level than Penn State was. Figure 1: Histogram of week-to-week DUSHEE performance scores for Nebraska, Penn State, North Carolina, SMU, and Georgia during the 1982 season. On the y-axis label, "wins" should be "games" So you can see that from this chart, Penn State only had 2 games where they scored out better than 100, including the Nebraska game. Nebraska, on the other hand, had 7 games where they scored out 100 or better. The Penn State game for them was one of the games in the 0-50 block. The Alabama game is the game in the < 0 block for Penn State. So was Penn State the better team? Maybe. Were they a little fortunate to have played one of their best games against Nebraska at the same time that Nebraska played one of their worst? Maybe. Or was there just something about the matchup between the two where Penn State had an intangible advantage that doesn't necessarily show up in a box score? Hard to say definitively. But if I were a betting man and the two played 100 times that season, my money would have been on Nebraska to win more than they lost. 2) 4-Loss North Carolina As I mentioned in the last entry, every year there is at least 1 team that DUSHEE places in significantly higher esteem than pretty much anybody else in the world. In 1982, that team was North Carolina. UNC's 4 losses were: 1) a 1-point loss to Pitt (#5 DUSHEE), 2) a 7-point loss to Boomer Esiason's Maryland team (#10 DUSHEE), 3) a 3-point loss to Clemson (#25 DUSHEE) and 4) a 6-point loss to Duke (#47 DUSHEE). So all of their losses were close games and, aside from Duke, to high quality opponents. UNC also outgained Pitt (by 50) and Clemson (by 123). They pounded everybody else on their schedule, including a bunch of so-so teams, but also including three good to really good teams: 1) A 24-point win (313 YM) over Vandy (#40 DUSHEE), 2) a 19-point, 261-yard win over Bowling Green (#31 DUSHEE) and a 16-point, 125-yard win over Texas (#6 DUSHEE) in the Sun Bowl. In their 8 wins, they beat their opponents by an average of 36-9. So North Carolina was a case of a good team doing what it should against bad teams (beating the hell out of them), beating the hell out of most of the mediocre teams it faced, and going .500 against the good teams, blowing half out and losing to the others in squeakers. If a team plays a good team closely, the difference between a 1-point win and a 1-point loss is the difference between a PD of, say, 16 rather than 14. It has very little overall impact on the DUSHEE score the team earns on that day. A fatal flaw in the system, some might say. I say that the system is measuring average capability in a noisy data set. A one-point loss today could have easily been a 6-point win if played the next week. 3) Whither Georgia and SMU? How can 4-loss UNC be so high when one-loss Georgia and one-tie SMU were so low? Well, I've already spoiled the SMU analysis. Unlike North Carolina, SMU struggled to beat some bad teams in 1982. Here is how DUSHEE scored the Southwest Conference that year: Rk Team PD YD Score 6 Tex 17.91 139.10 58.31 12 Ark 14.40 125.99 48.73 18 SMU 13.54 66.86 39.08 65 Bay -3.96 -30.84 -12.91 66 Hou -7.53 24.95 -13.63 70 Tech -4.92 -51.01 -17.68 85 TCU -9.47 -71.74 -30.60 86 A&M -8.97 -97.44 -32.83 104 Rice -20.03 -89.80 -56.62 Table 2: DUSHEE ranking of the SWC in 1982 The SWC was very Mountain West-esque in 1982. In fact, on average, the WAC was a slightly better conference in 1982 than the SWC was. The average opponent beat Baylor by 4, Tech by 5, Houston by over a TD, TCU and A&M by 9, and Rice by 20. SMU managed to outdo the average opponent against A&M and Rice but underperformed compared to Baylor's, Houston's, and TCU's average opponent. SMU was outgained by Tech by 72 yards. SMU had three games with negative DUSHEE scores, UTEP, Baylor, and Texas Tech So compared to UNC, SMU did not do what good teams should do ... beat the hell out of bad and mediocre opponents. And they didn't have enough good opponents to make up the difference on that end. Georgia's lower than expected score comes from their YD shortfall. Georgia was outgained by their opponent in 5 of their 11 wins. Georgia was 9th in the country in PD, 39th in YD. Nonetheless, even if one dismisses the low YD, Georgia was still far from the top in PD. Georgia also failed to blow out several weak opponents including Kentucky (13-point win against a team that got beat by an average of 18 points against everybody else) and Mississippi State. Georgia just kinda plodded along, good but not great, but winning every game. The exception to that was a 44-0 blowout of a good Florida team. That was the only game where Georgia's DUSHEE score exceeded 65. Compare that to how Penn State and Nebraska did in Figure 2. Figure 2. Figure 1 with North Carolina, SMU, and Georgia added And in the end ... So what does all this mean? The results of the 1982 season are bogus? Penn State should return their MNC trophy and Herschel Walker should give his Heisman to Mike Rozier? Probably not. Penn State beat Nebraska. They might have won on "puncher's chance" type odds, but they won nonetheless. Was Georgia clearly the best team in the country going into the bowls by virtue of being undefeated? Maybe not. College football teams are erratic. Their performance varies greatly from week to week. If a lowly team has the good fortune to pair its best game of the season against the powerhouse's worst game, then upsets happen. Play that game again and the lowly underdog probably gets treated like the lowly underdog. In a league where a lot of teams play such a small schedule, you have to live with the variability. If a team gets lucky and wins an improbable game, you live with the result. And with such a short schedule, the odds of that improbable result affecting rankings and standings are very high. In major league baseball, the random improbable win over C.C. Sabathia is irrelevant over a 162 game season. Or as Coach would say, "I'm not a rich man, I'm not a young man, I'm not a handsome man, I'm not a tall man, I'm not a strong man, I'm not a talented man, I'm not a well travelled man, I'm not a smart man, I'm not a milk man, I'm not a fat man, I'm not a gingerbread man, I'm not a..."

Duquesne Frog

Duquesne Frog

 

An introduction to DUSHEE for the uninitiated ...

I'm a little uncertain how this blog thing works, but here's mud in your eye ... Why DUSHEE? A few years ago, on some random college football Saturday, I decided that it wasn't enough for me to just waste the day on the couch watching football ... that I needed to incorporate my two loves, college football and Excel spreadsheets, into one glorious symbiotic whole. And on that day, the Duquesne Ultimate Statistical Heuristic Evaluation of Excellence (I don't remember if that was the original non-sense acronym or not, just go with it) was invented. I did it for a couple of reasons. One, I love putzing around with numbers. It makes me happy. Second, amid the hullabaloo about BCS computer polls, I wanted to better understand how they might be used to analyze performance and what their value was. Using just a few simple and basic performance measures, could I create a ranking system that paralleled both the "human" polls and the other computer ranking systems? Third, being a SABR nerd, I'm fascinated by comparitive techniques that allow you to compare teams of different years and eras. And I think DUSHEE actually does a pretty fair job of doing it. In the following weeks in the lead up to the next season, I'll go into some of the interesting things the numbers tell you. Before we go there, though, I thought it best to start off with a description of the methodology. Many on here have seen and understand it (not that it's all that complicated) but this will serve as a reference for future discussions. Plus, one of the reasons why so many people have a negative predisposition to "computer polls" is that the authors of those polls don't reveal their methodology. I know why they don't; their efforts get them column inches and salaries and name recognition in USA Today and other places and if they told you how they got their numbers, no one would need them to do it. But as we may discuss down the road, I'm not too far off of Sagarin and many of the other polls. He tends to have many of the same "Whaaa?!?!" teams that show up in DUSHEE as appearing to be way better than anyone else thinks they are. Which means either they are way better, or Jeff and I have the same GIGO problems with our models. That's for you to decide. You can criticize the methodology, you can take it and tweak it, you can print it out and use it to line the cat's litter box. I'll take it all into consideration although I've invested way too much time in this to greatly change the way I'm doing it and go back to tweak all the stuff I've already done. So feel free to comment, but don't expect a massive revision of the method. Not until USA Today starts sending me a check too. Keep It Simple, Stupid DUSHEE looks at two numbers for each game between two FBS opponents, margin of victory (i.e., point margin) and yardage margin (i.e., yardage margin). Those two numbers are then compared to how a team's opponent's opponents did against them. For example, let's look at last years' TCU-Baylor game. Why that game? Because we beat their ass. TCU won the game 49-21 (PM = 28) and outgained Baylor 508 to 431 (YM = 77). Baylor, over the course of their season and removing the TCU game from their totals, beat their FBS opponents by an average of 8.7 points and outgained their opponents by an average of 78.5 yards. The difference between how Team A does against Team B in a particular game and how Team B does, on average, in their other 11 (or however many) games is what I'm defining as the Point Differential (PD) and the Yardage Differential (YD). So in this particular game, TCU scored a PD = 28 + 8.7 = 36.7 and a YD = 77 + 78.5 = 155.5. On the other side of the coin, we can look at how Baylor fared in this game. Outside of the Baylor game, TCU was outscored their FBS opponents by 0.9 points per game and outgained their opponents by an average of 29.8 yards per game. So Baylor's PD for the game was (-28 - 0.9 = -28.9) and their YD was (-77 + 29.8 = -47.2). This calulation is made each week for each game and the total PD and YD for each team is summed over the course of the season and averaged. TCU's totals for last year looked like this: Opp: Kan Uva SMU | ISU Bay Ttech | OkSt WVU KSU | Tex OU MichSt PD: -7.40 12.00 9.09 | -17.64 36.73 0.00 | -10.91 0.00 1.82 | 14.58 2.27 2.92 YD: -9.00 119.10 -93.64 | -5.27 155.45 258.18 | -43.45 82.73 40.18 | 29.67 3.91 156.42 Table 1: TCU's point differential (PD) and yardage differential (YD) for each game during the 2013 season TCU's average PD and YD per game for the year was 3.6 and 57.8, meaning we were 3.6 points better against our opponents than their opponents were against them. And we outgained them by 57.8 more yards. The same thing is done for all the other now 124 FBS teams. To this point, everything is a completely objective, and really very simple, statistical analysis. If the NCAA wanted to, they could report a team's PD and YD on their website. But then we get to the slightly subjective part ... There's Room for Me, Sagarin The team with the highest PD and YD (not necessarily the same team) is used as a reference and every other team's PD and YD is "normalized" relative to it. This is a fancy way of saying that you divide each team's score by the highest team's score. Thus the team with the highest PD will get a "normalized" PD of 1. And it follows that the team with the highest YD ... you get the picture. The subjective part comes in when establishing the relative importance of PD vs. YD when establishing a ranking metric. Some would argue that YD is meaningless and PD is the only thing that matters. That sort of analysis is basically what basketball and baseball do with the RPI calculation. As I'll argue in a follow up entry at some point, I think margin of victory is made "noisier" by the randomness of turnovers and other intangible factors that aren't necessarily controlled by the performance of a team. Looking at TCU's numbers from last year in Table 1 shows just how "noisy" a team's performance can be from week to week. Without that noise, a slightly above average team like TCU was last year should play other slightly above average teams to a near tie, lose to better teams by a steadily increasing number as the opponent gets better, and beat worse teams by an increasingly bigger number as the opponent gets weaker. That is rarely the case, but that is not the subject of this entry. So I view yardage margin as a damper on the noise of college football performance. Touchdowns are scored on freak plays, but good teams are going to outgain bad teams on a more consistent basis than they outscore them. I'm not sure the numbers support this theory entirely (hey, a subject for a future post!) but it's my story and I'm sticking to it. So to rank teams, I take the normalized PD and multiply it by 67 and I take the normalized YD and multiply it by 33. This weights points over yards 2:1 and gives a team who leads FBS in both PD and YD a perfect score of 100. I chose 100 because it's 100, the most kickass of all round numbers. And I chose a 2:1 weighting largely because it used to be 3:2 and Uniballer kvetched and moaned that DUSHEE was slandering his poor Kansas State team who had a great PD and a not-so-great YD one year. Completely arbitrary and subjective, but there it is, the basis for the DUSHEE ranking system. The best team in all the land will have a DUSHEE score at or near 100. A middle-of-the-road, mediocre team will be at or near 0, and while I do not normalize the low end to force a team who would be worst in the country in both categories to a score of -100, that is typically pretty close to where the worst team ends up. An Ass out of U and me There are many underlying assumptions that DUSHEE makes as a model, but probably the most important one is that the teams have to be interconnected enough to make this comparison valid. Ideally, the method takes strength of schedule into account implicitly. If you are playing a bad team and barely beat them, you will get a bad score. If you play a bad team and beat them about as badly as another team has, you will get roughly the same score as that team. And DUSHEE will reward "moral" victories. Lose closely to a good team, you will have a high YD and PD. But for a "bad" or "good" team to be established, a team has to play enough teams over some range of oppositional "quality" to make the evaluation. Over the full course of a season, that should be the case. Even a team that plays in a weak conference should have enough basis for comparison if you establish their comparative performance to teams separated by one degree from them. By this method, a team is not only linked to their opponents, but also to their opponent's opponents as a basis for evaluation. So if every team plays 10 FBS opponents and all of those opponents have played 10 opponents, each team is getting compared to 100 teams (minus, of course any repeat teams in the opponent's opponents schedule). That should represent enough connectedness that even teams with really weak or strong schedules are evaluated fairly. If All That Doesn't Have You Positively Orgasmic ... That's probably enough minutia to digest for now, but to tease future entries ... I've run DUSHEE for every season dating back to 2000. Starting in 1999, it seems like the centralized storage of box scores on a single intraweb site vanishes. A few months ago, I discovered that the NCAA.org website mysteriously has hand-written official box scores saved as pdf's from the 1982 season, but nothing between then and 2000. So I ran DUSHEE for 1982. DUSHEE also generates a strength-of-schedule number, a conference strength number, can be used to select the best and worst single game performances in a particular season or week, can be used to evaluate the historical strength of teams and conferences (well, "historically" back to 2000, at least) and all kinds of other cool stuff that I know will keep this audience in rapt attention for the months ahead. So we'll discuss all that, what "computer" polls really tell you, and what the numbers tell you about the performance of college football teams as a whole. And then when the 2013 season rolls around, we'll start to look at those numbers as well.

Duquesne Frog

Duquesne Frog

Sign in to follow this  
×