by Bill Glasheen » Mon Aug 20, 2012 11:57 pm
In my line of work, Justin, we don't just give that lip service. We quantify the amount of variability we explain, and feel comfortable with the fact that not all variability is accounted for. But understand that if - for instance - we can explain 20 to 50 percent of the variability with a future medical cost model, that's good enough for folks like the Center for Medicare and Medicaid Services (a.k.a. CMS) to rely on those models to steer billions of dollars to health care practitioners and entities based on how sick people are and are expected to be.
The "chaotic element" - if you in fact mean mathematical chaos - comes into play only in special situations. It's there in Nature only when nonlinear systems are operating in special domains.
The "random element" is often that which we choose not to account for. Sometimes that's intentional. Other parts are things we knowingly can't quantify.
The point is that hard work matters, and being smart matters. On the flip side, being a bum matters, and being stupid matters. When you look at it on the extremes, it makes perfect sense. The "unexplained" part of variability in predicting success in this case wouldn't be so much what general place you fell on the success vs. failure spectrum, but rather in being precise about quantifying it. Horse shoes and hand grenades... Doing well - or not - is something that should pass a sniff test, and not be subject to 5-digit precision measurement. I don't expect a healthy, industrious college grad (who majored in something reasonably relevant) to be begging on the streets, and I don't expect a high school dropout stoner to win a Nobel Prize.
- Bill