It's simply how our minds so often work. We cannot fathom failure in light of success. Well, we can fathom it. We just don't really think it's going to happen - not yet, not now, not to us. Our optimism in the future runs ever-strong.
So strong, in fact, that we often create the illusion of successful outcomes ourselves, even without the sorts of cash incentives a Miller or a Law or a Madoff offers. We see a con going well, and read our desires into ambiguous signs, to convince ourselves that we've invested wisely, be it money, time, reputation, or any other precious resources. When we want to, we see signs of good fortune everywhere. It's one of the reasons for the famous hot-hand fallacy that Cornell University psychologist Thomas Gilovich identified in 1985. Gilovich had first observed the phenomenon among basketball aficionados when they pronounced players on a "hot streak" or playing with a "hot hand." The players and coaches, too, seemed to believe it - even going so far as to select certain draft picks because they were perceived to be playing hot at the time.
To Gilovich, the whole thing seemed highly unlikely. He was a cognitive psychologist, studying rationality and its departures, and there was simply no reason to assume that people's talent and skills could show such tremendous, lasting deviations. He'd also been working with Amos Tversky, who, along with Daniel Kahneman, had identified the "belief in the law of small numbers" some ten years prior: that we believe that chance rates seen over the long term should also be reflected in the short term, and if they are not, something else must be going on. For instance, since a coin is supposed to land on heads half the time, we expect it to do so if we toss it, say, ten times. We don't take into consideration the fact that averages are derived over a broader timescale. And so, if we see tails coming up time and time again, we tend to think that we're particularly lucky.
Simon Lovell, con artist turned legitimate magician, wrote a book about exploiting such tendencies when you're out on the grift. One of the simplest short cons revolves around getting people to place bets on outcomes they think either highly unlikely or highly likely because of their recent experience (an experience that, in the convincer, is unerringly positive) - and then to upend those perfectly reasonable-seeming expectations. In what's called a proposition, or prop, bet, you first make a claim, and then ask for any takers to bet against you. One example: betting that you can tie a cigarette into a knot without tearing the paper. Impossible, right? Anyone would bet against you after trying it a few times on their own. But it's not actually impossible if you first wrap the cigarette tightly in the cellophane from the box and then tie the cellophane. Prop bets take advantage of what we expect and then do something completely different. They take the psychology behind the convincer and exploit it to its logical conclusion.
Gilovich and his colleagues decided to test our hot-hand-like perceptions by analyzing the shooting records of the Philadelphia 76ers and the Boston Celtics. They failed to find any relationship suggestive of actual hot hands: a player who made one shot was no more likely to make another soon after. Any deviations from their average level of play were the result not of streaks, but of chance: they were to be expected from the probability distributions and were not the result of some magical power that a player suddenly had.
But even though the hot-hand fallacy has been largely disproven - in 2006, a review of twenty years of data found an overwhelming lack of evidence for the presence of sudden bouts of talent - it continues to govern our thoughts of the future. If a player is on a hot streak, we should give him the proverbial ball because he will make the shot. In hedge funds, Kahneman points out, we see the same thing. When a fund has been tremendously successful for a few years, investors pour in: success now, even in something as volatile and chance-like as markets, means success always. Often, though, those phenomenal returns evaporate or reverse. It is, after all, a game of chance. Sure, a manager could be quite good, but ultimately he also has to be quite lucky - and luck can often masquerade as talent when the latter is absent.
Not only do we get fooled into thinking that just because something is working now, the future will be even better, but we often project our desires onto our estimations of likely success. In other words, we tend to think that what will happen is what we want to happen, especially if the outcome is an important one.
Konnikova, Maria. The Confidence Game: Why We Fall for It . . . Every Time (pp. 211-212). Penguin Publishing Group. Kindle Edition.