The Grand Illusion
Last week, the American gymnast Ali Raisman tied for third place at the Olympics in the all-around competition… and lost the tie-breaker because she was a more consistent performer than the Russian gymnast with whom she was tied. Yes… that’s correct. The more consistent performer lost in a competition designed to reward the most consistent I doubt that was what the gymnastics federation had in mind when they drew up the tie-breaker rule, but that sort of result was absolutely and mathematically inevitable because of the rule, which provided that, in the event of a tie, the lowest score each of the two gymnasts had, out of the four events, would be thrown out, and the one with the highest remaining score would be declared the winner. The result mathematically is that when two gymnasts are tied, if one has a particularly bad single event, the winner will always be that one.
This is an excellent example of how what seems, on the surface, to be a perfectly logical “solution” created a result totally at odds with the goal of the competition. Unhappily, this doesn’t just happen in Olympic gymnastics, but in all too many areas of society, business, and government. It occurs because too many decision-makers, from politicians to business CEOs, don’t think through the implications and ramifications of their decisions. Sometimes, that occurs because they don’t think events will ever require contingency plans – as in the case of safety requirements at Japanese nuclear facilities. After all, who could have predicted the freakish combination of earthquake and tsunami? And in gymnastics, what was the probability of a tie with that many judges and four events with scores measured in thousandths of a point?
Results at variance with what one might call common sense also occur when situations change and the rules or procedures don’t. Or they occur because everyone is so concerned about the moment that something totally predictable that occurs periodically, but at long intervals, is totally overlooked, as in the case of Delta Airlines forgetting to renew their online security certification at a time when they had cut commissions to travel agents and increased the fees required for telephone booking, thus increasing the percentage of reservations and payments made online.
All of these situations are the result of failure, in some way, to consider the implications of either certain actions or of failing to act… and all are preventable… but, given human nature, few will be.
side note:
I have a fundamental problem with sports where the scores are reliant upon judging. Attempts at making things fair always fall short and things like the example above occur.
(That being said, I watch them. Yes, take away my man card – but I’ll tell you that these young men and women make far better rolemodels for my daughters than most other potential rolemodels put forward by pop culture, so I watch with my daughters who can’t get enough of them and I cherish the time.)
Back on topic:
“They didn’t think things through” is often code for “the law of unintended consequences just bit them in the ass.” It also proves that while intentions might be good, there a reason Hell’s Road is paved with them. Few people (myself included) want to take the time to sit down and actually think about something.
Checking the math (in whatever form it takes) is tedious, time consuming, and useful only up to a point because there is literally no way to think through all the permutations/combinations that occur when a plan hits real life.
It doesn’t mean it shouldn’t be done. It absolutely should because then most simple, foolish mistakes would be caught. But sometimes big whoppers occur and no one can see them coming because of timeframe, other existing conditions, or issues that come up afterwards.
Anyone who makes major decisions should be able to think through the first two or three branches of problems that an idea creates… but further than that is requires a skill, training, and life experience that most people do not have… and it all varies according to the topic. As an example, a NYC congressmen making an agricultural bill (or even commenting on one) would generally be farcical… but they do it anyway.
I too raise a Spockian eyebrow at any sport that requires judging to decide the winner or loser. The biggest mistake would be to let the rule remain as it is for the next international competition. Changing the rule now won’t help Ali Raisman, but this situation must not be allowed to happen again.
I recall something (forming new discussion lists in something akin to Usenet) that had _very_ conservative requirements to approve a new list: some minimum number of votes pro or con (say 100) to establish a level of participation and interest, _and_ a supermajority (say 2/3) in favor. The odd effect was that if the supermajority was over-satisfied but turnout wasn’t quite high enough, a vote against could actually result in approval…which would be very frustrating to those opposed.
When that was recognized, the rule was rewritten to something like a supermajority of at least 2/3 and at least 67 “yes” votes. The tradeoff was that in the event of an even larger supermajority, less total participation would be required to obtain approval.
Any time humans make attempts to decide among other humans, or allocate or tax among them, making the procedure more complex might be done to provide a broad acceptance that it’s fair, by taking into account a variety of concerns. But very quickly, additional complexity leads to enough oddities and loopholes to undermine the very confidence it was intended to create.
There are certain indications that it might be very difficult to come up with a set of rules that perfectly embody nontrivial and somewhat abstract (let along subjective) objectives. Gödel’s theorem, the uncertainty principle, the halting problem – those and more seem to me to hint at that.
Aside from taking care to avoid unintended consequences, maybe even running simulations or attempting formal proofs, there’s another problem: rule-makers always want to seem pro-active, so they constantly tinker, leading to an incomprehensible total of rules. Aside from a few laws that have no need to change (against murder, assault, theft, fraud, etc), all that should be needed is to update them to make it clear they apply to new scenarios (e-fraud is still fraud); and many laws intended to prevent some much more specific abuse that is a matter of mindset rather than inherent to human nature should have expiration dates or conditions (sufficient voluntary compliance to get rid of the law).
The illusion of pursuit of an unachievable ideal can lead to an accumulation of overhead and lost liberty sufficient to give societies expiration dates, if they allow it.
I guess it would be too much to ask for plain speech from a politician. Pro-active tinkering reminds me of the ‘stupid and energetic’ corner of the box.
If someone just said, “I’m going to raise taxes a little bit on everyone and cut spending for everyone so that we can fix our economy” I’d vote for them just for being clear and concise.