In 2011 you had 11-1 Oklahoma State and 11-1 Stanford, who were edged out by 11-1 non-champ Alabama.
You can make a claim that Alabama (having only lost to an undefeated LSU team) was the 2nd-best team in the nation. But I'm not sure you can argue that OkSU or Stanford were necessarily "weak" unless by strength/weakness you're comparing the shine of their helmet. They weren't 2-loss champs like PSU in 2016 and OSU or USC in 2017. They were 1-loss conference champs. Shouldn't that mean something?
Either way, however, the BCS was essentially a mathematical formula. It was the polls and the computer rankings. The human polls differ from the committee in that the individual human voters are independent and don't collectively agree on who the representatives in the title game should be.
The committee does, however. And the committee has several stated rationales as far as what should carry weight. 2016 PSU lost in weeks 2 and 4, and then basically rolled roughshod through the rest of the season, but were passed over for a team they beat head to head. And I don't want to argue from results, but that team got completely decimated by Clemson in the semis. 2017 OSU did what they were supposed to do (schedule tough and win their conference) and had a pretty good resume, beating two teams that finished in the top 10, but based on losing to a team that finished 2nd in the standings and having one "WTF?" game were jumped by a non-champion who didn't schedule particularly tough and didn't really have any "signature wins".
I think you can easily argue the 2016 PSU team was tough, and and beat the one-loss non-champ that was selected over them head to head. I think you can easily argue the 2017 OSU team was tough, had some great wins, and 9 times out of 10 beats that Iowa team to a pulp. Is their resume truly worse than an Alabama team with no signature wins that lost to a 10-3 Auburn team?
I want to address a few things in here that haven't already been covered by others:
First, I strongly disagreed with the BCS taking Bama after Bama had lost H2H to LSU. However, OkSU's OOC that year was:
- UL-Lafayette
- Arizona (finished 4-8)
- Tulsa
That is hardly a murder's row and they got all their best B12 opponents at home (OU and KSU both came to Stillwater so their best road win was at Mizzou.
Stanford's OOC was:
- SJSU
- Dook (finished 3-9)
- Notre Dame (finished 8-5)
Stanford also lost to Oregon by 23 points which I think hurt them just like Ohio State losing badly to Iowa hurt the Buckeyes this year.
Alabama's OOC was:
- Kent
- Penn State (finished 9-4)
- N. Texas
- Georgia Southern
Penn State was far better than any OOC game that either Stanford or OkSU played and Bama's only loss was by a FG to the #1 team.
A lot of people keep bring up PSU last year as a comparison and I view that completely differently. I was happy with the decision last year not only because my team got in but also because I thought the committee sent a message that SoS matters. That is a message that I want the committee to send because it will improve the viewing experience for all of us as fans.
I agree with a statement made above that Ohio State got in last year largely because of their blowout OOC win over B12 Champion Oklahoma. I liked that because I figured it would encourage other teams to schedule difficult OOC games because the potential reward made it worth the risk. However, on further review, I think it really isn't and that scheduling cupcakes is probably the best bet for getting to the CFP.
These last two years the best OOC match-up has probably been tOSU/OU both years. Those two games included three P5 Champions (OU both years, tOSU this year), two teams that finished the regular season with one loss (OU this year, tOSU last year) and two that finished with two losses (OU last year, tOSU this year). All of that said, I think the net result of that series was -1 playoff berths. Imagine if the Buckeyes had hosted Tulsa both years and the Sooners had hosted Ohio U both years. Assume that the Buckeyes and Sooners easily beat the Golden Hurricanes and Bobcats, what would have been different?
2016:
- Ohio State with one loss, no title, and no big OOC wins probably gets left out. -1
- Oklahoma with one loss, a B12 title, and no OOC losses probably get in. +1
2017:
- Oklahoma with one loss, a B12 title, and no big OOC wins probably gets in anyway. even
- Ohio State with one loss, a B1G title, and no OOC losses probably gets in. +1
Collectively, the Buckeyes and Sooners probably lose one CFP berth by playing each other the last two years. So the message sent by the committee is to schedule cupcakes because at the end of the day "# of losses" is a higher ranking criteria than "SoS".