Michigan football and the Anna Karenina principle
In 1997, the UCLA geographer and anthropologist Jared Diamond wrote about the domestication of animals in his Pulitzer Prize-winning book Guns, Germs and Steel. The book discussed why Eurasian and North African groups have done most of the historical global conquering. In it, he popularized a concept called the Anna Karenina principle, named after the famous 19th-century Tolstoy novel. The first line of that novel states: “Happy marriages are all alike. Unhappy marriages are each unhappy in their own way.”
This is taken to mean that with respect to the factors necessary to produce a happy marriage — compatibility in the areas of communication, emotional needs, financial tendencies, sexual and intimacy requirements, religiosity and child-rearing, among others — they must all meet a basic standard. A large enough deficit in even a single factor is likely to produce an unhappy marriage, even if the other factors are present in abundance.
Professor Diamond used this principle to show that it takes a large number of factors to make an animal a successful candidate for domestication, and that if even one of the factors was unsatisfactory, domestication wasn’t possible. It didn’t matter which factor was unsatisfactory, and it didn’t matter how excellent a fit for domestication the other factors were if one of them failed to meet the threshold.
The animals must be:
-
big enough to be useful
-
small enough to be manageable
-
easy to feed
-
able to grow quickly enough to make economic sense
-
able to remain calm
-
able to breed in captivity
-
even-tempered enough to not pose a major risk to handlers
Success, in this sense, was not an overabundance of a few qualities. Success was the ability of all of these factors to meet a basic standard. For example, zebras meet all of the qualities for domestication, except one: they are ill-tempered and fierce. They bite all the time. Useful for the savanna; not so useful for the farm.
This pattern easily translates to football. Whether we are talking about a player, a play, a position group, a game or a team, the Anna Karenina principle must be satisfied if they are to have successful outcomes. They must meet a basic threshold across several domains, with failure in one or two domains severely limiting the success of the whole enterprise.
For example, a running back with great size, speed and balance but poor vision or inability to pass block has a ceiling to his utility. The success of any offensive play usually depends on the execution of several blocks, correct reads by the QB or RB, an accurate throw and sure hands for passes, and an ability to find the gap and hold onto the ball for runs; failure to execute even one of these things usually dooms the play. A defense with a deficit in one of its position groups will be exploited over and over, no matter how great its other components.
The players and coaches who get the brunt of fan frustration are often those who have some superb skills, but a fatal flaw or two:
Erick All looks like one of the best TE blockers in the last 10 years, has good size, runs solid routes and knows how to get open. But he’s had 5 drops in four games, limiting him, at least for now (by the way, sometimes he lets the ball into his body too much, rather than trying to catch it out in front of him, and I think that’s part of the issue). He did this in the Wisconsin game on the second drive, on a corner route throw from Milton that was otherwise perfect, and another downfield throw in the third quarter hit him directly in the palm). The Anna Karenina principle has the last word on his viability.
Joe Milton has been well dissected, of course, but his skill set is primarily limited by his inability to see and feel zone defenders; no matter how good the rest of his skills become, he will be of fairly little use to the offense if this ability doesn’t improve. The Wisconsin game was a dud, but the other skills are improving — he’ll throw into double coverage less over time, and his deep ball, while still occasionally overthrown, has already improved some. The Anna Karenina principle will have the last word on his career.
It applies to plays, too: on the first interception Saturday, Milton threw a ball that would’ve hit Eubanks between the eyes. Everything went right on that play — pass pro? Check. Good read, or at least reasonable enough? Check. Accurate throw? Check. Only problem was that Eubanks didn’t catch the ball in front of him. Which meant that the play didn’t go right. The Anna Karenina principle had the last word.
You can apply the Anna Karenina principle to position groups (e.g., the 2017 OL had most of what it needed, except a functional center, which torpedoed its abilities), to a team (the 2018 team was excellent at several positions, but had basically one composite functional DT, the end) or to a program, where poor recruiting in one or two key areas has created the current state of the roster, despite very good to excellent recruiting elsewhere on the roster.
The almosts, coulda-beens and if-onlys that have marked Michigan football since 2008 have often had the Anna Karenina principle at work. That principle can’t explain all of the failures, but it certainly can help frame our current discontents. When Michigan returns to success, it won’t be because of brilliance in one area. It will be because all potential deficiencies will have been identified and avoided.
November 16th, 2020 at 3:41 AM ^
I like this but am most interested in applying it to the program, rather than the 2020 team. Which factors best determine the ability of an elite coach (still hidden or already uncovered), for example, to both (a) be hired here and (b) succeed at the highest level?
I could guess about some of those Anna-Karenina/Jared-Diamond factors:
- brand
- dollars
- fanbase
- history/prestige
- recruiting footprint
- athletic director relationship/mentorship/commitment
It's not easy to guarantee these are distinct enough from one another to merit individual mention ... or, even when distinct, it's tough to promise whether each is important enough to "success probability" to deserve its listing. But looking at those, I anticipate that factors remain unmentioned because these alone wouldn't seem to explain Michigan's place in the CFB universe.
Perhaps we'd have to add more toouchy-feely or vague items to really wrap our arms around the full truth:
- ineffable toxicity/systemic self-defeating attitude(s) in the program
- unwillingness to cheat as blatantly as the Joneses
Feedback welcome!
November 16th, 2020 at 7:24 AM ^
Joby, FFS don't go high-culture (a welcome departure from the COVID-19 denial that has polluted this board since March) on us and mess up the opener, which actually is:
“Happy families are all alike; every unhappy family is unhappy in its own way.”
November 16th, 2020 at 8:09 AM ^
Doh! My mistake, blueheron. Thanks for the correction.
November 16th, 2020 at 8:58 AM ^
The Anna Karenina principle might be a promising analytical tool, looking at what's happened and spotting core elements to explain why it happened.
The challenge of the A. K. principle is using it as a planning tool, anticipating what elements are core and identifying them even before they're fully developed.
We can spot flaws after we've seen them. But can we anticipate a base of fundamentals that will blossom (developing a 3-star)? Or distinguish—before the fact—between flaws that are fixable (keeping a 5-star on track) and flaws that are embedded (a 5-star who doesn't live up to his promise)?
November 16th, 2020 at 9:09 AM ^
I should have started with my first thought: what a cool idea to apply the Anna Karenina principle to football!
November 16th, 2020 at 10:19 AM ^
Interesting: one area I'd take exception in is that football skills are traits....Can they be taught? Like better footwork for lineman or better pass catching skills? Will repetitive drills improve performance? I'd think some things, yes; others, you just might have to have the talent...
November 16th, 2020 at 11:23 AM ^
Great analogy and I agree
November 16th, 2020 at 11:46 AM ^
New England constantly takes guys that have flaws and just simply asks them to do things that do not expose their flaws. This principle does not apply to players, it does to programs.
Cordalle Patterson has made a career in the NFL of being a guy who cannot make good decisions at RB or WR but can make plays with the ball. So teams just ask him to do that.
Taysom Hill cannot throw the ball with accuracy consistently, so the Saints just ask him to run and hit the occasional deep shot.
Players can be successful with flaws, they are all flawed in college. Coaches just have to make it work.
November 17th, 2020 at 8:31 AM ^
I was thinking something similar as I was reading this. I think this idea can also be applied to team and systems, not just players. We've heard before that some think Harbaugh makes things too complicated, tries to treat college kids like NFL players, etc. The idea that one small flaw can blow up the whole thing becomes much more pronounced the more complicated the systems gets. A simplified system that lets good athletes execute simple reads leaves less room to get ruined by one mistake.
November 16th, 2020 at 2:59 PM ^
I really like this analogy - thanks. I would add something in the criteria about school administration. I was disturbed when I read that the U-M administration was happy with Harbaugh because the football program was once again profitable after years of not-so-much. I apologize because I can't remember if I read that last year, or the year before, but it made me wonder about some potential systemic issues at U-M.
November 17th, 2020 at 3:10 PM ^
Give it another year or two, then, and they will be unhappy with Harbaugh, because people are not going to devote a day or two to going to Ann Arbor to watch this dog’s breakfast of a football team he is putting on the field.
November 16th, 2020 at 3:48 PM ^
Interesting concept but I'm not sure it completely applies. It's an over simplification. In the same book Diamond asserts agriculture was the first factor in advancing humanity, specifically grains. Cultivating more that you needed facilitated trade. So it is geography and environment that influenced success more than anything else.
By that analysis we should focus on local recruitment. Something we haven't done well the last couple of years, especially considering some of our best current player; Hutch, Mayfield and Hayes are Michigan products
It's nice that we should recruit big animals that check a lot of boxes. Erick All did not have a drop against UW. The pass you are talking about and with Eubanks were both heavily contested and both could have been called.
The staff I'm sure has evaluated everyone on all facets. Harbaugh has praised All for his talent and suggested he might break records. At this point it would be premature to say 'Erick all you can't catch so you'll only block from now on.'
I do think, as others have advanced, there is a reluctance to modify the system to fit the skill set of the players. Ben Mason was key in game one and has not seen much playing time since. In 4th and short against Minnesota Milton followed Mason from under center. Since then its been Haskins or Milton from the gun.
You look at john Harbaugh, he retooled his offense to fit Lamar Jackson's skill-set.
Don Brown has not adjusted to his players strength and weaknesses and is 1-5 with 4 blowouts.
UM generally recruits in the top 10 annually with 4 and 5 stars. So, don't you feel it's more reasonable to assume that lack development and antiquated or inflexible systems are more likely causes of team failure than recruiting egregiously fatally players?
November 17th, 2020 at 12:53 AM ^
Anna Karenina was an adulteress who threw herself in front of a train.
Michigan football refuses to whore itself out but yet still manages to get hit by the same train the last Saturday of every season.
November 17th, 2020 at 8:58 AM ^
I took Russian literature (somewhere around 2003) and the teacher asked us what killed Anna Karenina? I raised my hand and after a brief pause I said, "the train hit her." Everyone chuckled. Apparently I was mistaken...
November 17th, 2020 at 4:06 AM ^
Yeah, right. This is stupid. Over intellectualizing the game of football.
How about the "Mad Magazine Theory of Why Michigan Sucks at the Footballs"?
Or
Rub dirt on it
Or
If it floats its a witch
My favorite is the Detroit Lions theory of failure. I.e. Michigan has turned into the Lions.
November 17th, 2020 at 10:06 AM ^
Yes, but there are a great many things wrong. And Jared Diamond is a popular read, but a decidedly mixed bag when it comes to real rigor.
One could as easily constitute a "good enough mother" principle wherein you adduced the five or six attributes common to all successful programs and showed how a lot of other variables could not be met while they remained successful.
At any rate, we suck at most of them right now. And a happy marriage is not a constant.
November 17th, 2020 at 12:13 PM ^
since we have no guns on offense, no steel on defense, I am hoping the germs end this season before 12/12
November 17th, 2020 at 2:07 PM ^
Very interesting take. I do disagree with your assessment of Milton's first interception. It was hard to tell from the field camera angle, but there was either a design flaw in the play, or an execution flaw. UM faked a run left, and got half the defense to flow that way, they sent 3 receivers right. The problem was they all went to nearly the same place. There was not enough spacing (through design or execution). I didn't get a chance to count, but I'm sure there were 4 or 5 Wisconsin defenders in the area. Those are never good odds by which to complete a pass, much less avoid an interception. I can't remember when (either later in the game, or in a game on Sunday), another team (it might have been Wisconsin) ran an identical play to the other side of the field. Their receivers, however, were spaced at about 5/10/15 yards downfield, so that the QB actually had multiple options, and receivers only had to beat their defender. While the interception could hardly be blamed on Milton. In fact, I'm pretty sure that particular play was called to instill some confident in him, from the get-go. The question is, was it a poorly designed play, that was at risk of failure from the start or did poor execution on the players' part force Milton to have to make a perfect throw?
November 18th, 2020 at 7:42 AM ^
I've noticed that Milton has thrown balls that were spot on from a targeting perspective. However, they have had little arc on them in addition to having had flames shooting off from them.
I just don't see the average college player being able to catch many of them.
And, believe me, UM has many (sub?) average receivers out there
November 18th, 2020 at 3:34 PM ^
A.K. principle is used in management practices. Just replace family with company. I teach it to my clients. Related to it is pre-mortem analysis. Before a project begins, all stakeholders gather in a room and the leader says: "it is one year from now and the project has failed. Why?" It asks people to think of all the things that could wrong. It works best, if the leader comes up with the first reason. Then others will feel free to come forward with their own reasons. I have been part of countless projects and it is amazing how some neglected weak point is mentioned during these meetings.
We apply the AK principle to see what the critical factors are and then the pre-mortem analysis pokes as many hole as it can to find the points of failure.
Comments