Expectations

Week 11 Conference Wins Update

Preamble

“Break their hearts my pride and hope, break their hearts and have no mercy.”
    - Miss Havisham (Charles Dickens’ Great Expectations)

Week 11 means there are two games remaining, and the computations are now in the realm of solution using your favorite handheld device. The resulting charts will either be single or dual-moded in nature, with not much beyond. I’d previously considered making last week’s post the last installment of this diary, but then M fell victim to the trap game of all trap games, and failed to exorcise its demons in Kinnick Stadium. As a result, the charts have become moderately interesting. That said, this diary will be brief in that it skips over the B1GW results and the Win Differential Distributions from past diaries. However, in light of re-energized three-way race for the B1GE title, a new section has been added to analyze the probabilities and scenarios by which each of the three contenders can punch their tickets to the B1GCG.

Nonetheless, with all of its goals yet ahead, Team 137 continues its ascent toward the pinnacle of college football, the Rose Bowl College Football Playoff. As part of Harbaugh’s diabolical master plan, he has played the ultimate troll upon rival OSU by dropping the Iowa game, which has all but blocked the Buckeyes from getting to the B1GCG even if they win out. Speight being knocked out of the last two games also plays into that plan by incorporating a shift in the offensive threat matrix known as Harbaughffense just two weeks before the biggest collision to be seen since Football Armageddon. The threat posed by John O’Korn of running the ball - combined with the relative dearth of film-based documentation of JOK doing just that - adds yet another element to the ruthlessly efficient schemes that Harbaugh will have at the ready. Between now and Columbus, the challenge is that the Wolverines must avoid an untimely demise in yet another impending trap game with ChaosTeam 2.0.

Schedules, Margins, Probabilities & Distributions

B1G East Schedules & Margins Rundown

The table of schedules below shows the overall schedules for all seven teams in the B1G East based on the Bill Connelly’s S&P+ weekly ratings. The last table simply shows a rank-ordering of the B1GE teams based on their  expected in-conference win totals, it’s not a projection of divisional standings based on projected wins, losses, and tie-breakers.

S&P+

Michigan, by virtue of its loss to Iowa, slips from #1 to the #2 ranked team in all the land as per S&P+, with OSU right behind at #3. Penn State also slips in the the S&P+ ranks and now stands at #12 - two behind #10 Wisconsin.

Looking at the S&P+ probabilities, the Nittany Lions now lead the B1GE with nearly 7.9 expected B1G wins, ahead of the 2nd place Wolverines at 7.5 wins, with the Buckeyes less than 0.1 wins behind. This is all by virtue of PSU being strongly favored in its remaining games, and The Game looking to be a toss-up, with OSU only a ½-point underdog.

Indiana, despite losing to PSU, remains in the fourth spot. The Hoosiers, now with just under 4.0 expected wins and being favored in the last of their remaining games against Purdue, are marginally on track for bowl eligibility. Meanwhile, Maryland is also on the bowl-eligibility bubble looking for its sixth win. With LOLRutgerz still on the schedule, the Terps have an ace in the hole.

FPI

The FPI results differ ever so slightly. Here as well, Michigan comes in with the #2 ranking with OSU in the #3 spot. PSU tops all teams in the B1GE with just over 7.7 expected wins and is on track for a 10-2 season and have a fairly good chance of making it to Indy. The Nits lead second place OSU, who is a shade under 7.5 expected wins. U-M slips to #3 and is no longer favored in The Game. The margin now has the Buckeyes favored by 2.8 points. Indiana holds the #4 B1GE spot firmly at about 3.8 expected wins, and remains in the edge of bowl-eligibility. Likewise, Maryland is on the bowl-eligibility bubble, now at 3.0 expected wins and being a favorite in only one more game. Sparty closed on the last remaining game in which is was favored, capturing the Situation Trophy from arch-nemesis LOLRutgerz.

Here’s a link to a chart showing the results from the Power Rank-ings. The numbers here look similarly pessimistic.

B1G East Expected Conference Wins Distributions

The bar plots below show the expected total overall wins distributions for teams in the B1G East, in alphabetical order. Noted above each bar is the probability for that number of wins (you may need to click & embiggen to read it). The bar with the highest value is the most likely outcome (the mode). Also flagged on each plot is the expected overall win total (the mean). The last line plot is just an overlay of the same data from the other seven bar plots.

S&P+

So here in final week probability distributions, a sea change has taken place. The once proud Michigan spike at the top end of the overlay has been supplanted by … Penn State? Yes indeed, it is the Nittany Lions who have the greatest likelihood to win out at 87.4%, and a mode of 8 wins. Yet, Michigan also has a mode of 8 wins, but with only a 50.1% likelihood, it’s nearly perfectly balanced against 7 wins. This is of course primarily because of The Game, which is a near toss-up as, noted above. Thus, OSU has a mode of 7 wins, with a strong lean toward 8 wins.

Still, the cluster at the 8-win mode has grown larger by virtue of M’s loss to Iowa, however, the only way a 3-way tie can occur now is if all 3 teams lose one and only one more game - the likelihood of which is vanishingly small due to the covariance between M & OSU. That said, PSU can now easily advance if M drops one more game than PSU from here on out. Yet at this point, it still remains that the B1GE title is Michigan’s to lose. It’s just that the likelihood of not winning out has registered a significant uptick this past week, which as noted above is about 50/50. Meanwhile, the only way that OSU can claim the B1GE title is to beat Michigan and have PSU drop one more game than OSU from here on out.
Indiana and Maryland now sit firmly at 4 wins and 3 wins, respectively. MSU and LOLRutgerz are looking like locks for single-win and winless B1G seasons, respectively.

FPI

The FPI results show a similar profile, but with a slightly different clustering at the top. Here PSU and OSU show 8-win modes, with PSU’s mode being the strongest, while OSU shows a bias toward the 7-win mode. Conversely, M registers a 7-win mode with a bias toward 8 wins. As such, PSU has the highest probability to win out at 74.8%. OSU has a 52.2% chance, and  UM stands at a 42.0% chance to win out.  As in the S&P+ results, the remaining four teams occupy distinct modes with weak probabilities to break out of them.

Here’s a link to a chart showing the results from the Power Rank-ings.

B1GE FTW

With the various point spreads and win probabilities in mind, and with only two games remaining in the season, it becomes straightforward to run through the various scenarios by which the remaining contenders - Michigan, Ohio State and Penn State - can become the B1GE divisional champ. What is evident is that all three teams have paths to Indianapolis not only by winning out along with a little help, but that all three teams have paths to Indianapolis even if they should lose one game. That said, the only team that has the freedom to lose either of its last two games is Penn State. However, it’s not necessarily true that PSU is the most likely B1GE champ. So without further ado, here are the possible scenarios:

Following are pie charts depicting the likelihoods of each of the three teams based on the individual win probabilities derived from S&P+, FPI and the Power Rank.

So as you can see, Ohio State is not nearly out of it, however, the Buckeyes do have the lowest probability of advancing in all three ratings. Michigan has the greatest probabilities in S&P+ and the Power Rank, while Penn State has the best probability based on FPI. Michigan receives the only probability greater than 50% in the S&P+ breakdown. PSU’s probabilities have the tightest spread of only 8.6 points with an average of 40.7%. Michigan has the highest average of 44.7%, with a spread of 8.8 points. Ohio State has a spread of 17.4 points with an average of 14.7%.

Wrap-up

So there you have it. As you can see, the distributions are looking much more deterministic than earlier in the season. That said, Michigan remains in the catbird seat looking ahead to Columbus the last Saturday of the month. Regardless of what happens this week, both teams will still have a chance to advance by winning The Game. In the meantime nonetheless, Michigan will need to remain focused on the intervening game while breaking in a new Quarterback.

It’s undeniable that the prospects for Michigan football to play in the Big Ten Championship Game have definitely taken a hit with the loss to Iowa. Most if not all great teams have lapses over the course attaining a championship. Most if not all great teams have the benefit of luck on their side as well. To that end, Team 137 is fortunate to still be in mix after the gaffe that occurred last week, and if it has the character and leadership of a championship team, it will come together and set things right again. OSU is always a tough out when facing Michigan, and that’s why so many look forward to The Game with such great anticipation each year regardless of the stakes. In all, Team 137 has exhibited all the qualities one expects to see in a team that has both great expectations as well as a great capacity to realize those expectations.

Yours in football, and Go Blue!