This SOS Thing
By Woody Brammel, 10/29/99

Is it just me or does anyone else think this "strength of schedule" thing is getting blown WAY out of proportion? You can almost bet that any article you read these days about Virginia Tech's bowl chances will contain at least a half-dozen paragraphs trashing their weak schedule and suggesting that VT's appearance in the national title game would represent a sacrilege, a travesty - the end of college football as we know it. Obviously, a team's SOS affects its number of wins and losses. Schedule strength should be considered in the rankings. And it is - in spades!

First of all, there is absolutely no question that both the writers and coaches adjust subjectively for strength of schedule in their respective polls. How all this works out - too much or too little adjustment for SOS - is anybody's guess, but it surely is done. In fact, I think a lot of what we see in the AP and ESPN polls as voting on the "tradition" of a program could be explained equally well as voting on a team's traditional SOS.

Every computer ranking system also adjusts mathematically for strength of a team's opposition, of the opposition's opposition, of the opposition to the opposition's opposition, ad infinitum, to the limit of the ties between teams. And some computer systems may arbitrarily include even more weight to SOS. The Seattle Times poll, for example, includes "strength of conference" in some unknown way (the contribution of "conference strength" to a team's rating, over and above that reflected in the games played against conference opponents, is an elusive concept at best). This undoubtedly was responsible for Tech's ranking of 10th in the Seattle Times Rantings (oops, ratings) following the Syracuse game, while their average rank in the other seven BCS computer polls was 1.29.

Lastly, a team's SOS is considered AGAIN when the BCS ratings combine results of the subjective and computer rankings. The SOS is incorporated in the BCS ratings by rank quartiles, which logically should reduce somewhat the SOS impact, but it's there. Again.

So, why all the hub-bub about SOS from internet posters and sports commentators? The evidence I see suggests "uninformed stupidity" for the former and "selective statistics to prove the point of my story" from the latter.

I read an internet post on an ESPN site during the week following the Syracuse game, in which a fan from Texas indicated he would become physically ill if Virginia Tech should somehow play for the national championship. His problem? Tech's "absurdly weak SOS" which is "not accounted for" in the ratings.

"Not accounted for"?? Tech fell like the proverbial rock in all the computer rankings - specifically from 1st to 10th in the New York Times poll - after beating Rutgers 58-20 (no typo here; a nine position drop after WINNING by 38).

"Absurdly weak schedule"?? Jeff Sagarin's SOS ratings for the week following the Syracuse game showed SOS for Texas I-A programs (our poster's home state) to rank from 38th (Texas) to 92nd (Texas A&M) compared with VT's rank of 36. Some other interesting SOS numbers that week: Florida State - 31st; Penn State - 65th; Nebraska - 18th; UVa - 7th.

Just how much SOS impact are we looking for here? The argument seems to be that all a team need do is to schedule tough opponents - that actually winning a game against a "quality opponent", even occasionally, is far too much to ask. So is the suggestion to have Dueling AD's decide the national championship in making the schedules? I rest my case on the posters - uninformed stupidity.

The media commentators, however, should know better and I think they do. A group of print and broadcast sports journalists has been ragging on the Tech schedule all year, but have you noticed that before the Rutgers game (when Tech's SOS rank was in the mid-thirties) they never mentioned the actual SOS numbers. Problem was, the actual numbers didn't support the notion being offered for sale. On the other hand, following the Rutgers game, when Tech SOS ranking fell to the mid-seventies, Tech's SOS ranking of 75th or 79th or whatever was widely reported. Now, after the Syracuse game (SOS back to mid-thirties), the numerical moratorium has returned. Never fear, however, these guys will be back in force after the Temple game.

Occasionally, this "selection of statistics" has gotten almost laughably transparent. A Chicago Tribune writer praising the Big 10, and especially Penn State, recently wrote that VT's win over Syracuse was "tarnished by the fact that Syracuse barely defeated Pitt" (7 points), but saw no similar interpretations from the Penn State-Pitt (3 points) or Michigan-Syracuse (5 points) games. Another writer classified UVa as "a threat" to an undefeated FSU season two paragraphs after describing UVa as part of Tech's "cupcake" non-conference schedule. It all leaves me wondering whether these Tech detractors in the media are claiming some constitutional right to have it both ways, or counting on short-term memory problems among their readers?

The biggest hoax perpetrated by the media, however, concerns the strength of Tech's non-conference schedule. How many stories have you read/heard which began with reference to games against JMU or UAB as illustration of the evil conspiracy perpetrated by VT to steal the national championship? Tech can't be allowed to get away with this 'cause it just ain't fair.

If you want the real story, however, run the numbers on some non-conference schedules. What you find - for example, comparing Tech with the higher ranked and highly praised Penn State program - is that Tech's OOC opponents, ranked by Sagarin through games of October 25, have an average rank of 40th. Those for Penn State? A sparkling 50th (precisely, 49.5). As it happens, Tech non-conference opponents JMU, UAB, UVa and Clemson (ranked from worst to best) are, each and every one, ranked above corresponding Penn State opponents Akron, Arizona, Pitt, and Miami (also odd that Big East "cupcakes" Pitt and Miami - which Tech must schedule - are Penn State opponents by choice).

And if there is anything that's unfair in all this, it's that the BCS includes Akron's 6-2 record in computing Penn State's SOS (75% winning percentage for an opponent) but not JMU's 6-1 record for Tech's SOS, even though JMU is ranked higher and is a top-ten Division I-AA team (#7). For the purpose of computing Tech's SOS, JMU's record is 0-1, with the lone loss coming against - you guessed it - Virginia Tech.

Enough numbers and belly-aching. Here's an opinion. The BCS rules are THE rules. If Tech finishes 11-0 and doesn't make the national championship game due its relatively weaker SOS, so be it. We don't complain and neither does Penn State (Yea, right!) nor any other "deserving" team who is "left out". We know the system and we adjust accordingly in the future.

Personally, I began the season hoping Tech would play its best in every game and end the year with a seventh straight bowl appearance. I still feel that way. Is it just me, or does anyone else think this "We're Number One" thing is getting blown WAY out of proportion?

          

HC Voice of the Fan Archives

HC Home

HokieCentral.com is an independent publication and is not affiliated with or endorsed by Virginia Tech or the Virginia Tech Athletic Department. All material is Copyright ©1996-2000 by HokieCentral.com, all rights reserved.