Against All Odds, Week 14: The Turkey Illusion

In 1912, British philosopher and mathematician Bertrand Russell introduced the world to a story about a turkey. 

The turkey in question was fed and cared for every day. Each day, he would stick his head through a hole and on the other side he would received a veritable feast, at least from the perspective of a bird. Summer turned slowly into fall and the turkey grew steadily larger and stronger. It was a good life for the turkey and he saw no end in sight to his good fortune.

But one day in late November, something different happened. When our plucky protagonist stuck his head through the hole, instead of a hearty mean he instead met his fate by the means of a very sharp knife. A little while later, he was the guest of honor at the farmer's Thanksgiving feast.

This classic allegory is meant to illustrate the dangers of inductive reasoning. In other words, just because a certain trend has happened in the past does not mean that it will continue indefinitely. One has to dig deeper to understand the full picture. In this case, the turkey did not understand the true motivation of the farmer, which led to an inaccurate assumption about the future.

In this final weekend of the regular season, there were several practical examples of this so-called "turkey illusion."

For example, there were several turkeys who root for a certain school down the road who seemed to believe that just because their team had beaten their rivals four straight times, that trend was going to continue. This belief was held despite the broader view that the team was a double-digit underdog to the defending National Champions and perhaps the team in question had benefited during the winning streak due to an illegal sign stealing scheme.

Closer to home, some Spartans fans likely believed that because the Michigan State football team had lost eight games in a row that they would lose the final game of the season to Maryland. This belief was held despite the boarder view that the Spartans had been playing much better football over the past few weeks and (most importantly) had continue to play hard, fight, and compete in every game.

Both trends came to an abrupt end on Saturday. Most notably, Michigan State was able to stave off a second half surge from the Maryland Terrapins and scored a late touchdown on fourth down to secure the Spartans' first Big Ten win of the season.

A potential third variation of the turkey illusion may have also been in play on Sunday. This version said that because Jonathan Smith was not able to produce a winning season in either of his first two years that he was not a good fit for the head coaching position at Michigan State. The theory in this example is that Smith was never going to work out in East Lansing.

Illusions or not, we will never get a chance to learn whether Smith could have turned the Spartans around. On Sunday, the ax fell on Smith and his staff. On Monday afternoon, former Northwestern Coach Pat Fitzgerald was named as the next head coach of the Spartans.

I went on record last week with my support in Jonathan Smith getting another year in East Lansing. I still believe that was the best course of action. But, I also did not have all the facts or the broader perspective that Michigan State athletic director J Batt has. For this reason, I will reserve judgment on his decision.

I have always respected Fitzgerald as a coach and as a leader. He always fielded tough and competitive teams at a place with limited resources and recruiting pull. Fitzgerald seems to share a lot of personality traits with former Spartan coach Mark Dantonio. That is going to play well in East Lansing.

That said, his record in his final few years in Evanston were not spectacular. His Wildcats went just 1-8 in Big Ten play in three of his final four years. More concerning, he was fired from Northwestern due to a disgraceful scandal and has been out of coaching for the last three years as a result. This certainly calls into question my previous statement about his leadership abilities.

But for better or for worse, Pat Fitzgerald is now the 27th coach of the Michigan State Spartans. Maybe the next few years will result in a bounteous future for Michigan State football. Or maybe the Spartans are simply getting Northwestern's cold Thanksgiving leftovers. Only time will tell. 

Week 14 Betting Results

Let's take a look at the performance of last week's picks from my Bad Betting Advice article, starting with the overview summary shown below in Figure 1. More information about how to read this figure can be found in the Week One edition of Against All Odds.

Figure 1: Results of Week 14 showing the actual point differentials relative to the opening spread.

A total of 19 teams overachieved in Week 14 by beating the spread by more than 14 points including Indiana, Texas Tech, James Madison, South Florida, Louisville, East Carolina, Miami, Iowa, TCU, Florida, and Arizona. No team this week failed to beat the spread by 14 points, yet still won.

There were a total of 16 upsets in Week 12, which was a little below the expected value of 18.8 from my weekly simulation. Table 1 below summarizes those upsets and compares them to the picks made last week.

Table 1: Upsets in Week 14 based on the opening Vegas line compared to the upset projections from last week.

This week's biggest upset based on the opening spread was California over SMU (-11.5) which cost the Mustangs a shot at both the ACC Title and a potential playoff berth. Other notable upsets include Navy over Memphis (-6), Houston over Baylor (-3), Vanderbilt over Tennessee (-3), Clemson over South Carolina (-1.5), and New Mexico over San Diego State (-1.5).

My computer went 3-5 (37.5%) on upset picks which brings the year-to-date performance to 43-53 (45%). This is a solid performance and well above my benchmark target of 40%. But, it is not quite as good as the 2023 season (49%) or the 2024 season (51%).

The FPI went 4-5 (44%) this week which brings the year-to-date total to 24-35 (41%). This performance is also above expectations, but not as good as the last two years (44% and 47%).

There were no recommended picks against the spread by either computer this week. For the full slate of games, my algorithm went 36-31 (54%). The year-to-date performance on suggested bets remains at 32-29 (53%) and 381-371 (51%) overall. 

My FPI-based suggested bets finish the regular season at 20-16 (56%). The full set of FPI picks went 29-38 (43%) this week. This brings the overall year-to-date performance for the FPI to 355-397 (47%), which is significantly worse than my algorithm.

Table 3 below gives the results of the point total bets for Week 14.

Table 3: Results of the recommended point total bets ("over/under") for Week 14.

My point total bets have been a real turkey for most of the year, but they finished the regular season strong. All three picks were correct, two of which were locks. This brings the year-to-date totals to 26-31 (46%) for the locks and 82-89 (48%) for the full set of suggested bets.

Updated Big Ten Post-Season Outlook

Unlike last year, the final weekend of the Big Ten regular season was fairly uneventful with the favored teams winning all nine contests. As a result, Ohio State and Indiana will meet next week in the Big Ten Championship Game. Both teams are a lock for the playoffs and the Big Ten Champion will almost certainly earn the No. 1 seed.

Oregon defeated Washington to move to 11-1 overall. The Ducks are also a lock for the playoffs. 

Nine additional Big Ten teams finished with between six and nine total wins and will therefore be placed into non-playoff, traditional bowls.

Michigan (9-3) will almost certainly be placed in the Cheez-It Citrus Bowl against which ever SEC is the last team out of the playoffs. USC (9-3) will be placed in a legacy Pac-12 bowl, such as the Alamo Bowl against a Big 12 opponent.

Iowa and Illinois both finished at 8-4 and will be placed in likely the ReliaQuest Bowl (i.e. the former Outback Bowl in Tampa) against an SEC opponent or the Las Vegas Bowl against a legacy Pac-12 opponent (such as Utah).  

I would personally send Iowa to the ReliaQuest Bowl as the Hawkeyes last appeared there in 2019 while Illinois was there in 2023.

Washington also finished at 8-4 and is likely headed to the Sun, Holiday, or the L.A. Bowl. 

Minnesota and Nebraska both finished at 7-5 and seem likely headed to the Music City Bowl and the Rate Bowl in Phoenix. Minnesota hasn't played in the Music City Bowl since 2005, so that seems like the logical placement.

Penn State and Northwestern are the final two Big Ten bowl teams with 6-6 records. I would personally place Penn State in the Pinstripe Bowl in New York City and Northwestern in the GameAbove Sports Bowl at Ford Field in Detroit.

That said, Penn State and/or Nebraska might be a bigger draw for bowl organizers and could leapfrog other Big Ten teams in the overall bowl pecking order. 

National Overview

In SEC action, Texas upset Texas A&M and as a result knocked the Aggies out of the SEC Championship Game. Instead, Georgia will face Alabama for the SEC crown as the Crimson Tide were able to defeat Auburn.

In other SEC action, Ole Miss won the Egg Bowl versus Mississippi State to essentially lock up a playoff spot. Oklahoma is also likely in good shape to secure a playoff spot after a weekend win over LSU. 

The other SEC team that made a statement was Vanderbilt, which is now 10-2 as well after an upset win over Tennessee. In theory, the SEC can make a case for up to seven teams to make the playoffs, but most likely there will only be space for five.

In the Big 12, BYU and Texas Tech both won this weekend as expected against Central Florida and West Virginia to secure spots in the conference championship game. Texas Tech is likely safely in the playoffs, but BYU will need to beat the Red Raiders in order to secure an automatic bid.

Utah defeated Kansas to finish 10-2 on the season, but the Utes are too far behind in the rankings to have a realistic chance to make the playoffs.

In the ACC, Virginia easily took care of rivals Virginia Tech to claim a spot in the ACC Championship Game. SMU would have joined them had they escaped Berkeley with a win. Unfortunately for the Mustangs, California upset SMU forcing a five-way tie for second place.

In the end, Duke (7-5) will face the Cavaliers for the ACC crown thanks to having the highest opponent win percentage among the five tied teams. Duke was able to beat Wake Forest to secure the bid. 

With this turn of events, Miami (10-2) is now a long shot to make the playoffs despite a 10-win season including a Week One win over Notre Dame. Ironically, the Fighting Irish have an identical record and seem likely to make the playoff field over the Hurricanes

In Group of Five action, North Texas defeated Temple and Tulane shut out Charlotte and as a result those two teams will face each other for the American Athletic Conference title. If Tulane wins, the Green Wave are virtually assured a spot in the playoffs. 

However, if North Texas wins, it is possible that the committee could rank Sun Belt champion James Madison ahead of the Mean Green in the final playoff pole. This assumes that James Madison beats Troy in the Sun Belt Championship Game.

I have also been closely following San Diego State as a possible dark horse playoff team. However, the Aztecs were upset by New Mexico this week and failed to qualify for the Mountain West Championship Game.

A potentially interesting and not-so farfetched scenario is the one where Duke upsets Virginia to win the ACC Title. In this case, the playoff selection committee would need to compare Duke, James Madison (if they win), and the winner between Tulane and North Texas. 

The rules state that the five highest ranked conference champions shall receive automatic bids to the 12-team college football playoff. If Duke is ranked below both the AAC and Sun Belt Champions, the ACC could get shut out of the playoffs completely.

An even stranger scenario would occur if the committee were to elevate Miami to a high enough spot to allow the Hurricanes to claim an at large bid while simultaneously leaving Duke out. Although compelling, this would require the committee to only take four SEC teams or to leave Notre Dame out, neither of which seem likely.

Against all odds we have reached the end of the regular season. Michigan State's season may be over, and it was a bit of a turkey but I will continue to provide potentially dubious financial advice and other math-based analysis on the remainder of the college football season. Stay tuned.