Following a team like the 2005 Texas Longhorns will naturally cause a fan to question their place in the history of college football. Having generated ratings for each season of college football since the beginning in 1869 allowed me to see a quick and dirty comparison to great teams from the past. Long story short, that team ranks as one of the top ten in the history of college football and as the best of the BCS era as determined strictly by my algorithm. Obviously there is room for much debate beyond a simply computer program, but subjectivity will always pop up in such arguments. What can be shown objectively, though, is that every BCS computer system except for Massey agrees that 2005 Texas was the best team since the BCS was enacted. Massey places them #2 for that time period behind only the 2001 Miami Hurricanes. With the objective side of that matter settled from my perspective I recently set out to determine the best Longhorn season performances. If you are either familiar or completely uninterested in the theory behind the rankings go ahead and skip the next three paragraphs. For the other two of you, read on.
Ranking results from one season against even the previous year can be tricky. Rules changes, differences in number of games for each school, and the changing number of schools in the top division of NCAA football are but some of the problems facing any analysis of this type. These obstacles apply whether you are comparing individual statistics, team statistics, or the teams themselves. When the comparisons are extended to results from completely different eras the disconnects obviously become even larger, to the point of being insurmountable for direct comparisons.
The obvious and perhaps only solution to this problem is comparing each result within the context of the season in which it was attained. How good was the performance compared to others from the same year? With this approach decided upon the next step is to evaluate each season on its own. Team performances are fortunately the simplest results to evaluate given that there are fewer total team performances than any other useful metric associated with the sport. Each team plays a limited number of games and each game has two measurable results - the final score for each team. My ratings system is based on the late David Rothman's FACT ratings set as far as final calculations are concerned. However I assign different game scores than his system (still compiled by Peter Wolfe) does. Essentially my ratings make an adjustment for the ratio of points scored by the winning team, not just the margin. For example, a 30-7 victory is a more compelling result in my system than a 54-31 win while the two are equal in FACT's calculations. After considering each result in the season a maximum likelihood estimate is computed that produces the final ratings for each team that played during the season. The results are scored on a logarithmic scale. The next step, readjusting the scale to a standardized mean and standard deviation, is what allows comparisons between seasons. Essentially each team is assigned a sigma rating that reports their strength in terms of how many standard deviations their performance was either above or below the mean performance for that season.
Everything is purely statistically driven to this point. But I felt it appropriate to add further considerations to this particular rankings set. While the calculations discussed above included a minimum game score for winning the contest, I felt it important to consider the level of dominance displayed by the team week in and week out. To accomplish this I calculated the Pythagorean expected value for each season on a game-by-game basis. This adjusted pythagorean score has been shown to be slightly more accurate to actual results than a standard calculation based on the season point totals as a whole. This result was added to each team's sigma rating. The next factor considered was each team's final ranking for the year under consideration. This ranking was subtracted from the total after being divided by ten. Finally the number of losses and half the number of ties were subtracted after being divided by four. Certainly the factors were decided upon haphazardly, but once the decision was made to go beyond the straight ratings calculations there simply had to be some level of subjectivity introduced.
Now that the explanation is out of the way, here are the results with a short summary for each season performance:
1. 2005 - Final Rating Value 3.57
Fortunately for my sanity the system agreed with common sense. I find it hard to believe that Texas has ever fielded any team approaching the 2005 version and if my numbers are to be believed, it's unlikely that I'll ever see another this good in my lifetime. The gap between #1 and #2 is larger than the gap between #2 and #7. The two closest contests this team won were both road games, one at Ohio State (a Top 5 team for the season) and one at a rival that presents a difficult atmosphere. Despite what was in my opinion the team's worst performance of the year thanks to a down game from Vince Young this team still scored 40 at Kyle Field. Obviously having played another great team in the title game helps the ratings but that Rose Bowl is just one more feather in the cap. I once tried to put a value on the greatness of individual games and the Rose Bowl win over USC was only challenged by Nebraska's 1971 win at Oklahoma.
2. 1963 - Final Rating Value 2.90
Texas' first National Championship team scored well across the board aided by their destruction of Roger Staubach's Navy squad in the Cotton Bowl. When combined with the win over Oklahoma the 1963 Longhorns outscored two top ten caliber teams by a combined 56-13 in Dallas that season. There were some close calls along the way but at the end of the year Memphis was #2 in the ratings and they weren't close to Texas.
3. 1977 - Final Rating Value 2.79
Obviously this one will rank as a surprise on this list. The lasting memory of this outfit was a demoralizing 38-10 defeat at the hands of hated Notre Dame. But the objective evaluation shows that this team defeated two top 5 opponents and neither was at home. Aside from the three games against the elite teams of the year there wasn't a game that was even close. Further benefiting this team's result is that no team finished undefeated in the nation that year. Notre Dame lost to a weak Ole Miss squad that finished the year with a losing record. Arkansas helped this team's final rating immensely with their famous destruction of Oklahoma in the Orange Bowl. Texas handed Arkansas their only defeat and the Sooners their only other defeat of the year. Even with the adjustment for the loss this team still finished ahead of the team at #4.
4. 1969 - Final Rating Value 2.67
Combined with the 1977 team above, this placement speaks to the power of subjectivity in some cases over objectivity. Whereas in 1977 Arkansas' bowl result seriously boosted Texas' final rating, in 1969 the Hogs did quite the opposite. A common sense look back will show that the Razorbacks were likely on a serious down after the disappointment of their 15-14 loss at home to the Longhorns. But the fact of the matter is that they went on to lose to a very subpar Ole Miss team that they should have handled. As a result the 1969 Texas team finished behind Penn State in the computer ratings. Once again any amount of subjective analysis will prove that Penn State shouldn't hold that spot as they refused to meet Texas at the Cotton Bowl. But facts is facts. What this endeavor did show me, though, is that while the 1969 squad was the gold standard for Texas teams throughout most of my lifetime, that was a result of chronological luck moreseo than performance. The 1963 team was superior in most every aspect.
5. 1961 - Final Rating Value 2.50
The season ruined by the cockroaches from TCU. I haven't run any kind of numbers because frankly not even I am interested in ridiculously nerdy analysis of devastating losses, but I'd have to think the 6-0 home loss to the losing record Horned Frogs is the worst upset loss in the history of the program. The team recovered enough to paste A&M in the regular season finale and then defeat a strong Ole Miss team in the Cotton Bowl. Alabama was the only undefeated in the nation and their 10-3 win over Arkansas in the Sugar Bowl didn't quite stack up to the 33-7 beatdown that this Texas team put on the same Razorbacks. What could have been.
6. 1970 - Final Rating Value 2.36
Another chronological fluke enables this team to be claimed as our other National Champions while the 1977 squad is barely remembered despite their season-long dominance. Back when the final UPI poll was still taken before the bowl games this team also ran through the regular season before a bowl game loss to the Fighting Irish. The thirty game winning streak came to a crashing halt but in all honesty it was a fantastic year. There were no undefeated and untied teams that year, and Notre Dame's 10-point defeat at a 4-loss Southern Cal enabled the 1970 Longhorns to finish the year #1 in the computer ratings.
7. 1947 - Final Rating Value 2.26
Blair Cherry's first year in charge was very nearly Texas' first national championship squad. Somewhat of a forgotten team, at least to me, this version of the Longhorns handled everyone in their path with the exception of a Top 5 SMU team on the road. If the ratings are adjusted to change that one-point loss into a one-point win, the 1947 team becomes the first in school history to finish atop the ratings. It should come as no surprise, then, that this team was led by the only QB other than Vince Young with an argument as best in school history. Bobby Layne went on to an NFL Hall of Fame career and along with Young without question completes the 1st and 2nd team all-time Texas QB depth chart.
8. 1972 - Final Rating Value 2.14
The last great season of the Darrell Royal era this edition suffered its only loss to Oklahoma in a 27-0 blowout. They regrouped well and ran the SWC table. The season culminated in a 17-13 win over Alabama in the Cotton Bowl. Aside from the loss to Oklahoma the closest Texas came to defeat was a 25-20 road win against Texas Tech. Oklahoma's only loss was at Colorado and the Sooners finished #2 in the nation behind undefeated Southern Cal.
9. 1964 - Final Rating Value 1.92
Coming off the school's first National Championship, Texas followed up with another great year. The only loss was a one-point heartbreaker at home to Arkansas. The Razorbacks would end the bowl season as the only major undefeated left in the nation. The reason? Young Tommy Nobis led #5 Texas over Joe Namath and the previously undefeated #1 Alabama Crimson Tide 21-17 in the Orange Bowl.
10. 1983 - Final Rating Value 1.81
Longhorn fans still argue about who exactly was at fault for the muffed punt. Craig Curry muffed it but should he have ever even been in that position? The way the defense was playing against Georgia it seems almost certain that all he had to do was get out of the way and Texas would have finished the year undefeated. The Orange Bowl result later that night made it all the more painful as Miami defeated Nebraska. Texas would have won the title as the only undefeated team left and Miami would have had to wait for their first championship.
Rest of the Top Twenty:
11. 1968 - 1.78
12. 1941 - 1.63
13. 1975 - 1.54
14. 2001 - 1.51
15. 2004 - 1.50
16. 1918 - 1.47
17. 1981 - 1.38
18. 1914 - 1.35
19. 1959 - 1.30
20. 1990 - 1.09