clock menu more-arrow no yes mobile

Filed under:

Selective Perception Clouds Our View Of The Game

Not creating delusions is enlightenment.
--"Wake up Sermon", Bodhi-dharma

Tom Gilbert has spent more than half of the 2010-2011 season hanging in effigy as the Edmonton media and fans blame him for nearly everything that's gone wrong with the Oilers thus far.  The Edmonton Journal's Jim Matheson and David Staples took him to task on separate occasions for his play.  Fans even have their own #BlameGilbert hash tag on Twitter, and though it started as a gag, it was pushed as a discussion point by those who did not realize it was supposed to be tongue-in-cheek.

Even though no forward on the team has a .500 or better scoring chance percentage without Gilbert, it's Gilbert that has taken the blame, not the other defenders.  Even though Gilbert's only bad stretch of the season came when paired with Jason Strudwick, the narrative says that Gilbert's miscues have been behind the terrible penalty kill and weak team defense.  Message boards, blogs, twitter and even our game threads have notes about each one of Gilbert's supposed mishaps, and each one of those supposed mishaps is more evidence of Gilbert's woeful play.  The selective perception displayed by fans is not a new phenomenon, Dustin Penner and Shawn Horcoff have been and continue to be in the crosshairs of angry fans.

Selective perception is the manner in which people experience, analyze, categorize and interpret information in a way that favors one explanation over another.   Selective perception is a psychological bias a person uses to interpret bits of information in such a way as to support their existing views or beliefs, often using confirmation bias to build evidence to prove their pre-existing view correct.

The landmark study in selective perception was conducted by Albert H. Hastorf & Hadley Cantril in 1951.  After a particularly chippy Princeton - Dartmouth college football game, Hastorf and Cantril set about surveying reaction to the game from both sets of fans:

Two steps were involved in gathering data. The first consisted of answers to a questionnaire designed to get reactions to the game and to learn something of the climate of opinion in each institution. This questionnaire was administered a week after the game to both Dartmouth and Princeton undergraduates who were taking introductory and intermediate psychology courses.

The second step consisted of showing the same motion picture of the game to a sample of undergraduates in each school and having them check on another questionnaire, as they watched the film, any infraction of the rules they saw and whether these infractions were "mild" or "flagrant."


When Princeton students looked at the movie of the game, they saw the Dartmouth team make over twice as many infractions as their own team made. And they saw the Dartmouth team make over twice as many infractions as were seen by Dartmouth students.


When Dartmouth students looked at the movie of the game they saw both teams make about the same number of infractions. And they saw their own team make only half the number of infractions the Princeton students saw them make.

In other words, each set of fans saw the game in a manner that best fit their pre-existing biases, much like those "saw him good" fans I've written about previously.

"Remember it however you want to and I’ll remember it how it actually happened."
--Dave, "Couples Retreat"

At a recent game, I sat in front of a group of hockey parents, parents with kids in a major Division I program.  As they talked over the course of the two-game series, I learned the history of their hockey viewing and knowledge - each of the six parents swore they'd been to every one of their son's games since he was a small child.  I also learned of the real issue behind the team's recent swoon -- the coach was a terrible coach and never sent anyone to the front of the net.  As if prompted by fate, just as the conversation about coaching strategy drew to a close, the son (drafted by an NHL team) of one of the parents stole the puck in the neutral zone, burst down the left wing, drew three opponents into the corner with him and flung the puck into an undefended crease, where it slid harmlessly to the opposite point and was cleared.  "SEE?!!?!?  There was no one there, again!  How is this even possible?," roared hockey dad.  Indeed, it looked like the theory held some weight.  Except his team was on a penalty kill and in the middle of a line change.  Rather than examine the context behind the situation, hockey dad immediately interpreted the event to favor his unfavorable view of the coach.

This same sequence of events happens during each Edmonton game.  Dustin Penner will fail to win a puck battle and game threads will immediately erupt with yelps of "lazy" and "disinterested".  Shawn Horcoff will miss a shot wide and the words "overpaid" are immediately splashed on the screen.  Tom Gilbert will use leverage rather than make a hit to win a puck battle and he's "soft".  Brandon Dubinsky had developed such a fierce opposition (his detractors called him soft and unskilled) in New York that George Ays went to the stats to show Dubinsky's true value.  Fans extract a single play from a game, or something resembling a single play from each of a series of games to apply a narrative and develop a view of a player.  Each successive time they use selective perception to reinforce their viewpoint, the viewpoint becomes stronger and the fan becomes more sure of themselves.  These fans are caught in perceptual vigilance, they are aware of evidence because it relates to their current viewpoint.  They are searching for it, even if they are not consciously aware of it.

Penner is disinterested even though he's leading the team in scoring chances.  Gilbert is soft and useless even though every forward loses the chances battle without him.  I'll go back to what I said in the confirmation bias article:

It's impossible for individual fans to record, catalog, process, analyze and interpret the results of hundreds of independent events occurring constantly throughout a game, but it's much easier to pick out those events and sequences of events that support their conclusions

However, to take it one step further, it's even easier for an individual to interpret the results of events to fit the previous conclusion.  This is yet another reason why statistics are so important in analyzing a game, a team, or a player.  Remember, statistics, at the lowest level, are simply a vast collection of events.  Those events are processed and cataloged for future reference to compare to other players and teams.  Cataloging that vast amount of events is something even the greatest human mind is incapable of, and statistical analysis of those reams of data is something no human being can accomplish in real time while watching a game. 

Research has shown how people tend to mangle the details of events they've personally witnessed.  Consider the number of individual events occurring simultaneously throughout a hockey game for sixty minutes and how flawed our memory must be when attempting to recall those events.  Add to that a layer of selective perception and an additional layer of confirmation bias, and "saw him good" breaks down under scrutiny.  It might feel really good to see something on the ice which confirms our view about a player, but if the foundation of that view is fatally flawed, finding additional evidence does nothing but move us away from reality.  Running into counterpoints, especially those stats-based counterpoints, isn't a reason to bristle at the person offering up the counterpoint.  Explore the argument, understand the methodology and consider the flaws inherent to a "saw him good" viewpoint.  Visual evidence isn't all it's cracked up to be.