Vivek's blog: 2011

Tuesday, July 5, 2011

The Fallacy of Cause & Effect, Part 1

Many folks measure short-term success by assessing outcomes. That is, they look at actions, look at results, and then judge the worth of those actions based on the results. This sounds reasonable, but it can be woefully misguided. Here's why.

Say you play chess like I do - badly.

You gleefully take an opponent's rook only to realize - too late! - that you've left your queen exposed. Your adversary takes your queen and proceeds to wipe the board with you.

Anyone observing this game would be correct in dismissing your chess skills. You've lost because you made a stupid mistake.

But let's change the game. Let's say that you're playing poker instead. Imagine that you've been dealt a pair of Aces. (You're playing Texas Hold 'Em, by the way.) The person to your left goes all in. Everyone else folds, and you gleefully call. He flips up a King and a Queen - not a bad hand, but pathetic compared with your pocket rockets.

Then up comes the flop: King, Ten, Five. Then another King on the turn. Then a Queen on the river. Your Aces have been comprehensively trounced by his full house, and you are out of the game.

That guy goes on to win the tournament. Commentators praise his 'gutsy' play and his 'uncanny' instincts. He's a millionaire, and you're out on the street.

So what did you do wrong? Absolutely nothing. You played perfectly - and you outplayed your opponent. But he lucked out, and you got screwed.

Evaluating each person's play based on outcome would obviously be silly. The fact that he won and you lost doesn't mean that he did the right thing and you did the wrong thing. The opposite is true.

But in the 'real world,' we see this mistake made all the time. A budding executive 'takes a risk' and is catapulted to stardom or consigned to the garbage heap based on the outcome. The reward or penalty is handed out regardless of the circumstances. Was the risk justified? Did the executive gamble 100% of a company's fortune based on a 45% chance of success? Was there an alternative strategy that promised a better expected return but was ignored owing to its unsexiness - or (even worse) its having failed last time through chance? (The equivalent being folding a new pair of Aces because they were beaten in the previous hand.)

Life is far more like poker than chess. Our actions do not create unambiguous consequences; they merely provide us with small impulses in a sea of random chance. Any robust frame for judgement must take into account the circumstances that inform and determine the result of our choices.

Measuring success, in the short term, based on outcomes alone may penalize the person who 'took a chance' on the pair of Aces. And that is a losing strategy in the long run.

(Thanks to Flickr user robad0b for the image. And I've finally written a tangentially related follow-up post: The Fallacy of Cause & Effect, Part 2.)