One more domino fell last night, and with it, Utah bit the dust. The Utes now have less than a 0.1% likelihood of making the playoff according to our model, and that’s more a sign that our model’s error margin is too wide than anything else. If the Pac-12 wants to make the playoff, it needs Oregon’s 1.5% chance to bear fruit.
That, of course, would require a lot of help. How much, though? We took our model’s 4,000 simulations from last night and sifted through them to find out. Here’s who makes the playoff in each of the 16 possible ways today’s Power Five Conference Championships can go.
Please note that in some cases, probabilities do not sum to a perfect 400.0% due to rounding.
Oklahoma, LSU, Clemson, and Ohio State win (occurs in 31.3% of our model’s 4,000 simulations)
Ohio State – 100.0% Playoff Probability
LSU – 100.0%
Clemson – 100.0%
Oklahoma – 87.1%
Georgia – 11.5%
Oregon – 1.4%
It would be shocking if Oregon were to pass Oklahoma should the Sooners win today, and noteworthy but not drop-your-coffee shocking should Georgia hold on to the fourth spot with a loss and an Oklahoma victory. Our model does believe it’s possible, but overall, in the most likely scenario (less than one-in-three likely, however), the end result is straightforward.
Oklahoma, Georgia, Clemson, and Ohio State win (21.1%)
Ohio State – 100.0%
Clemson – 100.0%
LSU – 99.9%
Georgia – 98.6%
Oklahoma – 1.5%
It would take a one-in-one-thousand decision by the committee to keep LSU out should they lose to Georgia today. Again, straightforward, barring a dramatic reversal of course regarding Georgia’s loss to South Carolina, which to this point has been largely forgiven.
Baylor, LSU, Clemson, and Ohio State win (14.9%)
Ohio State – 100.0%
LSU – 100.0%
Clemson – 100.0%
Baylor – 77.6%
Georgia – 18.7%
Oregon – 3.4%
Oklahoma – 0.2%
Sometimes our model spits out weird things, such as Oklahoma’s 0.2% chance of making the playoff in this scenario. We’d prefer it doesn’t do that, but it’s the balance we strike between rigidity and precision—we’d rather say impossible things are 0.2% likely than accidentally saying possible things are impossible.
Baylor, Georgia, Clemson, and Ohio State win (11.4%)
Ohio State – 100.0%
Clemson – 100.0%
LSU – 99.8%
Georgia – 99.3%
Baylor – 0.9%
The argument here is not what will happen, it’s how a bad loss should be weighted compared to strength of schedule and strength of victory, and whether the committee treated it correctly (for the record, I think that answer is up to individual taste).
Oklahoma, LSU, Clemson, and Wisconsin win (6.5%)
Ohio State – 100.0%
LSU – 100.0%
Clemson – 100.0%
Wisconsin – 53.3%
Oklahoma – 42.1%
Georgia – 4.6%
In the event Wisconsin pulls off a shocker in the nightcap, our model likes the Badgers’ chances, barring Georgia and Clemson both winning too. They’re still only roughly a coin flip if it comes down to them and Oklahoma, though.
Oklahoma, Georgia, Clemson, and Wisconsin win (4.0%)
Clemson – 100.0%
Ohio State – 99.4%
LSU – 98.8%
Georgia – 97.5%
Wisconsin – 3.7%
Here we uncover the scenario in which Ohio State and LSU are each in the most trouble. It isn’t much trouble.
Baylor, LSU, Clemson, and Wisconsin win (3.5%)
Ohio State – 100.0%
LSU – 100.0%
Clemson – 100.0%
Wisconsin – 59.0%
Baylor – 34.5%
Georgia – 4.3%
Oregon – 2.2%
We probably all knew Wisconsin’s chances are better against Baylor than they are against Oklahoma. Still nice to put a number on it, though.
Baylor, Georgia, Clemson, and Wisconsin win (1.9%, or 75 of the 4,000 simulations)
Ohio State – 75 of 75 simulations
LSU – 75 of 75
Clemson – 75 of 75
Georgia – 73 of 75
Wisconsin – 1 of 75
Baylor – 1 of 75
We’re at a low enough probability now that we’re going to stop listing percentages by each team, since the error margins are too wide for that to turn out very accurately. Instead, we’re listing it as fractions, which should be taken with the understanding that the margin for error is quite wide, and if one of these unlikely scenarios does come to pass, we’ll run enough simulations tonight to take care of it.
Anyway, reminder that Georgia and Clemson both winning spells doom for Wisconsin regardless of their own result.
Oklahoma, LSU, Virginia, and Ohio State win (1.8%, or 71 simulations)
Ohio State – 71 of 71
LSU – 71 of 71
Oklahoma – 64 of 71
Clemson – 34 of 71
Georgia – 34 of 71
Oregon – 10 of 71
A one-loss team with no top 25 wins and a loss to a top-20 team, or a two-loss team with three top 25 wins (one on the road, one at a neutral site) and a loss to a 4-8 team who played five top 25 opponents. Our model says it’s somewhere near a tossup, with Oregon sneaking into the back end of the picture.
Oklahoma, Georgia, Virginia, and Ohio State win (1.2%, or 46 simulations)
Ohio State – 46 of 46
Georgia – 46 of 46
LSU – 46 of 46
Oklahoma – 43 of 46
Clemson – 2 of 46
Oregon – 1 of 46
This scenario is fun in that we either get an immediate LSU/Georgia rematch right away in the Fiesta Bowl or quite the performance of logical gymnastics by the committee.
Baylor, LSU, Virginia, and Ohio State win (0.8%, or 31 simulations)
Ohio State – 31 of 31
LSU – 31 of 31
Baylor – 30 of 31
Clemson – 13 of 31
Georgia – 13 of 31
Oregon – 6 of 31
Another Georgia vs. Clemson vs. Oregon debate.
Baylor, Georgia, Virginia, and Ohio State win (0.6%, or 24 simulations)
Ohio State – 24 of 24
Georgia – 24 of 24
LSU – 24 of 24
Baylor – 19 of 24
Clemson – 3 of 24
Oregon – 2 of 24
Another LSU vs. Georgia Fiesta Bowl, most likely.
Oklahoma, LSU, Virginia, and Wisconsin win (0.5%, or 19 simulations)
Ohio State – 19 of 19
LSU – 19 of 19
Wisconsin – 17 of 19
Oklahoma – 14 of 19
Georgia – 5 of 19
Clemson – 1 of 19
Oregon – 1 of 19
Would Wisconsin outrank Oklahoma? Signs point to yes.
Oklahoma, Georgia, Virginia, and Wisconsin win (0.3%, or 12 simulations)
Ohio State – 12 of 12
Georgia – 12 of 12
LSU – 12 of 12
Oklahoma – 7 of 12
Wisconsin – 4 of 12
Clemson – 1 of 12
Example of error margins being too wide at these small samples: Oklahoma outranks Wisconsin here but nowhere else when they both win. Still, we’re including this readout for the sake of your entertainment, because if you’re still reading, you really want to know what’s up in every possible situation. It’s worth noting here that in a Wisconsin vs. the Big 12 scenario, it may come down to margin/manner of victory.
Baylor, LSU, Virginia, and Wisconsin win (0.3%, or 10 simulations)
Ohio State – 10 of 10
LSU – 10 of 10
Wisconsin – 8 of 10
Baylor – 6 of 10
Georgia – 5 of 10
Clemson – 1 of 10
A familiar Baylor vs. Georgia debate from above, with Wisconsin added to the mix.
Baylor, Georgia, Virginia, and Wisconsin win (0.2%, or 7 simulations)
Ohio State – 7 of 7
Georgia – 7 of 7
LSU – 7 of 7
Wisconsin – 5 of 7
Baylor – 2 of 7
The most surprising individual scenario leaves us with a fairly plain overall debate: two losses, one of which was avenged and one of which was bad, four top 25 wins, one of which was the most impressive of any team this season; against one loss, which was avenged, only two top 25 wins, and a lot of narrow escapes.
Actually, that would still be pretty interesting.
