Perhaps the BEST B1G Forum anywhere, here at College Football Fan Site, CFB51!!!

Over in @ELA 's simulation thread it has become plausible that there will be a three-way tie for third place.  Given that there are only two slots available in the CG one team would obviously be left out.  As it stands now the determinative factor would end up being the cumulative winning percentage of conference opponents (ie, SoS) and the team that HAS a H2H win among the tied teams would likely be left out while the team that they beat would play the team that didn't play either of the others in the CG.  

Here are the [72].pdf"]current tiebreakers
[72].pdf"]current tiebreakers for a multi-team tie.  If you don't want to read all of that here is a shorthand version:

• H2H but only if they all played or if one defeated each of the others.  
• Record against common conference opponents.  
• Record against the best common opponent, then the next, etc.  
• SoS:  Cumulative conference record of conference opponents.  
• SportSource Analytics Ranking.  
• Random draw.  

I voted something like current. In fact, I don't see anything worth changing in the first four tiebreakers. 

I do think H2H should be the first tiebreaker (as it currently is). A team should not go to the CCG over a team it lost to unless H2H doesn't produce a CLEAR outcome. But it's true that if the teams did not all play, or if they (assuming three) played and each team went 1-1 against the other two so there wasn't a clear outcome, or there wasn't one team that defeated all of the others, that you need something else. This is directly comparable outcomes against each other... Unless it doesn't yield a clear and direct "winner" of the tiebreaker.

After that, record against common opponents, then against best common opponent, makes sense. Again, you're comparing performances that are directly comparable because it's teams that each tied team played in common. 

Only after that do you get IMHO to SoS. Because you've run out of directly comparable events, and now you're looking at more nebulous things like "who has the best resume?" 

I prefer "who has the best resume?" to "who is the best team?", which might come from point 5, or poll ranking, or some other analytic like FPI. Because again, it's rating you based on how strong your competition was. But I at least prefer "who is the best team?" to a random draw. 

So... I don't see any reason to change the existing system...

Instead of random draw, just employ the old Rose Bowl rule at that point. The team with the longest ccg draught. 

If it's for the last spot in a bowl game, it matters.

I voted something like current. In fact, I don't see anything worth changing in the first four tiebreakers.

I do think H2H should be the first tiebreaker (as it currently is). A team should not go to the CCG over a team it lost to unless H2H doesn't produce a CLEAR outcome. But it's true that if the teams did not all play, or if they (assuming three) played and each team went 1-1 against the other two so there wasn't a clear outcome, or there wasn't one team that defeated all of the others, that you need something else. This is directly comparable outcomes against each other... Unless it doesn't yield a clear and direct "winner" of the tiebreaker.

After that, record against common opponents, then against best common opponent, makes sense. Again, you're comparing performances that are directly comparable because it's teams that each tied team played in common.

Only after that do you get IMHO to SoS. Because you've run out of directly comparable events, and now you're looking at more nebulous things like "who has the best resume?"

I prefer "who has the best resume?" to "who is the best team?", which might come from point 5, or poll ranking, or some other analytic like FPI. Because again, it's rating you based on how strong your competition was. But I at least prefer "who is the best team?" to a random draw.

So... I don't see any reason to change the existing system...

I generally agree but, and this is mostly devil's advocate, where I don't like SoS is when the difference is insignificant.  For example:
In @ELA 's scenario the plausible three-way tie is at 8-1 between Oregon (lost at PSU), Michigan (lost at UNL), and Ohio State (lost at Michigan).  The only opponents that all three have in common are the two UW's, Wisconsin and Washington.  Running the tiebreakers:

• H2H2H:  Does not apply because they didn't all play and no team defeated each of the others.  
• Record against common conference opponents:  All three are 2-0 with wins over the Badgers and Huskies.  
• Record against the best common opponent, then the next.  They are each 1-0 against both common opponents so this does not help.  
• Cumulative conference winning percentage of all conference opponents:  At this point in the scenario Ohio State leads because their opponents are 26-30 vs 25-31 for Oregon's opponents and 22-34 for Michigan's opponents.  

I'm fine with that for breaking a tie between Ohio State and Michigan because the gap is 4 games.  It isn't huge but it is significant.  Ohio State played four teams that have two or less losses (MN, M, IL, PSU) and went 3-1 while Michigan only played three (tOSU, UNL, USC) and went 2-1.  Ohio State played more teams that would have a plausible chance against a league title contender.  

If we were breaking a tie between Ohio State and Oregon where the difference is only one game, this seems unfair.  The SoS difference between tOSU and Ore is that tOSU played 1-6 Purdue while Oregon played 0-6 Northwestern.  That isn't significant.  Either the Ducks or the Bucks should easily beat either the Boilermakers or the Wildcats so penalizing the Ducks because the Wildcats are marginally worse seems silly.  

My problem is that the above is very subjective and it would be difficult for me to write it into an objective rule.  

I generally agree but, and this is mostly devil's advocate, where I don't like SoS is when the difference is insignificant.  For example:
In @ELA 's scenario the plausible three-way tie is at 8-1 between Oregon (lost at PSU), Michigan (lost at UNL), and Ohio State (lost at Michigan).  The only opponents that all three have in common are the two UW's, Wisconsin and Washington.  Running the tiebreakers:

• H2H2H:  Does not apply because they didn't all play and no team defeated each of the others. 
• Record against common conference opponents:  All three are 2-0 with wins over the Badgers and Huskies. 
• Record against the best common opponent, then the next.  They are each 1-0 against both common opponents so this does not help. 
• Cumulative conference winning percentage of all conference opponents:  At this point in the scenario Ohio State leads because their opponents are 26-30 vs 25-31 for Oregon's opponents and 22-34 for Michigan's opponents. 

I'm fine with that for breaking a tie between Ohio State and Michigan because the gap is 4 games.  It isn't huge but it is significant.  Ohio State played four teams that have two or less losses (MN, M, IL, PSU) and went 3-1 while Michigan only played three (tOSU, UNL, USC) and went 2-1.  Ohio State played more teams that would have a plausible chance against a league title contender. 

If we were breaking a tie between Ohio State and Oregon where the difference is only one game, this seems unfair.  The SoS difference between tOSU and Ore is that tOSU played 1-6 Purdue while Oregon played 0-6 Northwestern.  That isn't significant.  Either the Ducks or the Bucks should easily beat either the Boilermakers or the Wildcats so penalizing the Ducks because the Wildcats are marginally worse seems silly. 

My problem is that the above is very subjective and it would be difficult for me to write it into an objective rule. 

The thing is, you're going to have narrow gaps sometimes. 

I still prefer SoS, even if the gaps are narrow, to subjective analysis like human poll rankings, or to objective but potentially inaccurate things like an analytic. 

Also note that as I understand the tiebreakers, you get to #4, you see that Ohio State has the best SoS, and now they've "won" the tiebreaker. But you then go back and re-run it with Oregon and Michigan, excluding Ohio State. 

In this case, it doesn't help, because Oregon and Michigan didn't play H2H, and the two teams that Oregon and Michigan lost to (PSU & UNL, respectively), are not common opponents. But in many cases, it actually would help and thus Oregon vs Michigan wouldn't come down to SoS. 

But now you return to SoS and Oregon goes to the CCG because they've got a 3 game advantage in SoS, not a 1 game advantage. So Oregon isn't being materially harmed here. 

I mean, maybe it would suck if Oregon's opponents were 23-33 and Michigan's 22-34... But them's the breaks. There's a reason it's the 4th tiebreaker, not the 1st. 

How do you fix a tie for 9th using those rankings?

😂😂