Annie Duke, World Series of Poker bracelet winner and behavioral science author (also sister to Howard Lederer, two-time World Series of Poker bracelet winner), describes the resulting fallacy in her book Thinking in Bets as having evolutionary roots.  According to Duke:

To start, our brains evolved to create certainty and order.  We are uncomfortable with the idea that luck plays a significant role in our lives.  We recognize the existence of luck, but we resist the idea that, despite our best efforts, things might not work out the way we want.  It feels better for us to imagine the world as an orderly place, where randomness does not wreak havoc and things are perfectly predictable.  We evolves to see the world that way.  Creating order out of chaos has been necessary for our survival.

When our ancestors heard rustling on the savanna and a lion jumped out, making a connection between "rustling" and "lions" could save their lives on later occasions.  Finding predictable connections is, literally, how our species survived...Incorrectly interpreting rustling from wind as an oncoming lion is called a type I error, a false positive.  The consequences of such an error were much less grave than those of a type II error, a false negative.  A false negative could have been fatal: hearing rustling and always assuming it's the wind would have gotten our ancestors eaten, and we wouldn't be here.

Seeking certainty helped keep us alive all this time, but it can wreak havoc on our decisions in an uncertain world.  When we work backward from results to figure out why those things happened, we are susceptible to a variety of cognitive traps, like assuming causation when there is only correlation, or cherry-picking data to confirm a narrative we prefer.  We will pound a lot of square pegs into round holes to maintain the illusion of a tight relationship between our outcomes and our decisions.

I take issue with much of the criticism of Jim Harbaugh I have seen on this site (such as equating a 7--1 team to the winless Detroit Lions) because I believe much of it is rooted in resulting.  The "BPONE" is your lizard brain hearing any rustling in the trees and emotionally fleeing the jungle.  It's the same urge that drives many folks to place the blame on play calls when a unit struggles, regardless of the talent and execution of the players on the field.  That's an important distinction: a play call is bad if the process leading to make the call is flawed, not because it yields bad results. 

Poker players know this phenomenon well: playing heads up, imagine you have a low pocket pair (let's say 5/5) and you put your opponent on two over cards (say King/Jack unsuited). When your opponent pushes all in, calling their raise is not a bad decision.  You have slightly better than a coin flip's odds.  All else being equal, if your opponent pairs an over card on the board and beats you, that does not change the fact that the odds were slightly in your favor, and your 5's should be good slightly more than half the time.  So as I strain this analogy to its brink, the question becomes, is Jim Harbaugh playing pocket 5s correctly? 

With that context, I aim here to evaluate a few of Harbaugh's high leverage decisions from the MSU game from a process perspective.  To do so, I think it is important to distinguish what I view as the three primary ways a head coach influences wins and losses (not breaking any ground here):

1.  Recruiting & Player Development: scouting, recruiting, retention, and developing players.  Building relationships with players, developing a culture and identity within the program.

2.  Scheme: hiring assistants, empowering them to run their schemes, ensuring scheme matches talent, evaluating assistant performance and influencing game plans accordingly, fostering cohesion among staff, etc.

3.  In-game decision making (loosely, game theory): use of timeouts, 4th down decisions, use of gadget plays, onside kicks, etc.

Anyone who has engaged with me on this site this year (or any time, probably) knows I am pretty dogmatic in my desire to have Michigan retain Jim Harbaugh for the long term.  There are many reasons for my belief in Harbaugh, but for now I will boil it down to categories 1 and 2 above. 

Michigan has shown it can recruit well, but not at an elite level, under Harbaugh. Perhaps you would argue it took too long, but the hiring of young assistants, several of whom are viewed as elite recruiters (or logically will be - such as Ron Bellamy), was a clear admission that recruiting, as well as the culture within Schembechler hall, was not what it needed to be.  Jim Harbaugh saw a flaw in the recruiting process (assistants hired for their schematic prowess more than recruiting chops, many at toward the end of their careers), and adjusted accordingly.  It is still too early to determine if this will bear fruit, but the logic of the hires and willingness to adapt assistant qualifications in a sea change fashion show a coach focused on the process, not just the results.

The same goes for scheme.  The Don Brown experiment ended disastrously, but let's not forget how strong that hire looked at the outset. Jim modified his process here too: while still leaning on those he trusts (this time his brother rather than Belichick), he plucked a young coach with high potential and zero coordinator experience to run his defense.  Harbaugh eschewed a focus on results (after famously finding Don Brown by looking up the number 1 overall defense) in favor of a process that sought to align scheme more with Michigan's talent, as well as the talent and schemes of their opponents (OSU).  Additionally, Harbaugh sought a coach who would bring scheme versatility and complexity rather than relying on a simplistic approach.  The Macdonald hire clearly shows a shift in philosophy: there is more than one way to solve your problems, and it does not always have to be aggression. 

With that out of the way, let's dissect a few decisions from Saturday:

1.  The two-minute drill at the end of the first half.  There are a few decision points here that I think bear review.  First, the decision to take multiple shots at the end zone.  In the moment, I argued with friends you would be better off shooting for a play at the sticks accompanied by a quick spike rather than multiple shots in the end zone.  But the counterargument is clear: any play short of the end zone inbounds risks a turnover or the clock running out.  Kicking a field goal to go up 2 scores is significant: I think you would have hopefully seen more risk taking if Michigan was only up by 3 at that point.  The other decision here is what to do after Mel Tucker uses all of his timeouts to ice your kicker.  To me, that created a potentially huge opportunity: send out Moody following the last timeout, then either sub him out or sneak a QB on the field as the holder and go for 6.  MSU's inability to call timeout and panicked ensuing substitution should increase the odds of a TD.  You're the chip leader by just a bit, you can choose to lean your opponent hand by hand, or you can bluff a big pot and potentially take a commanding lead.  But again, taking the points to go up 9 makes a lot of sense.

2.  The "QB controversy" in the four minute offense up 3.  I do not care whether Mcnamara was medically unavailable for this play.  As Ian Boyd pointed out, McCarthy makes sense in that situation as you are clearly going to run the ball, and a plus one with a QB run/option could be significant.  But more importantly for me, your process ALL season has been to prepare #9 for a moment like that, giving him snaps at key times in the Wisconsin and Nebraska games, for example.  Also, plenty of coaches believe in the "go right back to the guy who fumbled" philosophy, and I tend to agree.  It is not as if JJ was playing poorly in this game, either.  If we add the wrinkle that we can assume #12 could not play, your only other alternatives are to burn a timeout or play a 3rd stringer, neither of which are particularly palatable.  I really struggle to see how anyone could have issue with this call, especially those who consistently claim that Mcnamara is the one who handicaps this offense, and the run game in particular.

3.  Substitutions on defense.  This one is the most clear mistake, and Harbaugh has admitted as such post-game.  The frustrating aspect of this is that it has occurred all year - going back to the Washington game, Michigan subbed into tempo and got punished for it a bit.  Scott Frost also gamed the back judge effectively enough to get a score in the red zone with a late substitution.  It's a learning curve for an NFL coordinator used to middle of the field hashes, long play clocks, and infrequent tempo.  But while it seems to be a glaring error to sub the way Michigan did in this game, it is important to recognize that the defensive line platoon has proved largely successful, and is arguably one of the reasons Michigan has improved so much defensively.  In part because Macdonald believes sincerely in matching personnel, but also because of the talent make up of this team.  While we have seen flashes and big strides from Smith, Hinton, Ojabo, Jenkins, and Morris, Michigan still lacks a true impact down-to-down defensive tackle, which we all "knew" before the season started.  Consider whether Michigan is subbing the defensive line so much if Maurice Hurst or Jacob Slade is on the roster.  My point is, some of this is from lingering Don Brown recruiting issues (although you may be able to attribute some blame to Nua as well).  So shutting down the D line platoons in this game was probably not realistic. The glaring error in my view was not calling a timeout when half your defense  onto the field as the ball is about to be snapped (IMO, the risk reward of calling timeouts earlier in the half weighs heavily in favor of using them on defense rather than offense, and calling one while staring squarely into a potential opponent touchdown is a worthwhile timeout).

There are many more decisions you can question, but I believe each one deserves a truthful examination of the decision making process.  Other criticisms essentially equate to "call the touchdown play." 

For all the merit to Harbaugh's track record and processes, I do think that for this team to take the next step, he should ratchet up his risk tolerance in games against evenly matched teams.  He has shown he is capable of doing so; you only have to go back to this season's Wisconsin game to see a guy going for it on fourth down from the minus 35.  The hire of Matt Weiss also suggests Harbaugh understands he had/has some blind spots on this front.

If I have not massacred the rhetorical device enough, my point is basically this.  Sometimes when your opponent pairs his over card on the river to beat your pocket pair, those are just the breaks.  But it is also true that sometimes you should try to push your opponent to fold before the flop and not even let them see the river.  

As always, Go Blue.

P.S. The title of this post is from Flight of the Conchords, and the music video is worth a watch.