My Life as a Con Man
swyx
Posted on June 14, 2020
First, My Story
"How confident are you?"
My boss would often ask this question of me, his eyes trying to pierce through mine into the cloud of self-doubt beyond.
Before I was a developer, I was an equities analyst in a market neutral hedge fund. If that sounds like a lot of gobbledygook, just think of it as the purest form of stock market trading possible. Regardless of the direction of the market that day, my job was to come up with stock and options trading ideas to buy low, sell high, and make money. Not necessarily in that order. My boss, the portfolio manager, would assess these ideas and implement them in our portfolio according to how much he agreed with us.
I would always stall on the confidence question. Most stock pitches take the same shape - a list of 4-6 key reasons why the trade might go well, and a shorter list of 2-3 reasons why it might not. Any intellectually honest analyst should be able to argue both sides of an investment thesis, and be able to change their minds when the facts change.
But if the facts were obviously one-sided, then others in the market would realize it too, and the payoffs would adjust to account for that fact. No free lunch, right?
So you'd never rarely get a "90% sure thing", you were often pitching 60%-40% reward to risk probabilities (I am grossly oversimplifying here). This is a problem when your confidence intervals on your confidence exceed 10%.
My job title was Analyst, I viewed myself as an analyst – an analyst's job was to be as right as possible, to be a source of truth. So I gave balanced views, reasonable estimates, and admitted all my blind spots.
My boss, the portfolio manager, weighed that strongly against my confidence. You'd do the same thing! If an advisor of yours made a recommendation, but didn't seem very sure about it, you'd weight it less than someone who had a clear message on what to do. Watch Ray Dalio call it believability weighting in his TED talk.
My coworker, G, (also an analyst) was exactly that kind of person. G would make absurd extrapolations from unproven datapoints. He'd wildly exaggerate potential profits, and diminish any risks. Any data point that remotely supported the thesis got thrown into the pitch, quantified or not. The more bullet points the better. Anything more equivocal in nature got thrown out.
G did everything you're not supposed to do in quantitative trading. He'd backtest a massive battery of strategies and only pitch things that had worked enormously well in hindsight. He'd throw arbitrary weights on statistically insignificant data because the story fit better. He'd muck with chart axes to make divergences seem bigger than they really were and pick time windows that looked the most favorable. Financial models, if at all done, were just used to reinforce a preconceived idea.
And then he'd get up in front of the room and deliver the most bombastic, hyped-up pitch of a trade idea of anyone in our firm. He was a champion debater in his youth – he was quick on his feet and had a counter to every criticism, a riposte to every response. Above all, he was confident.
He knew it was intellectually dishonest.
But he also knew it didn't matter. He knew we knew. And that we'd let him do it... as long as he made money.
It shouldn't surprise you that his pitches were accepted and traded in far bigger size compared to mine, despite us having the same tenure. His blowups were huge, but his wins were massive too. In our first year he carried the team, and got probably 1-2 million for his troubles. By our second year he was dictating his own terms, and for all intents and purposes left to start trading his own portfolio in a different office. A superstar in the making. He'd wind up in circles of industry legends like Paul Tudor Jones and Peter Thiel, and we lost touch.
I'd try to imitate him, of course, since he'd been so effective. I spent less time in the details and more in the big picture. More time on confident persuasion rather than collecting information. My heart wasn't in it. Every time my boss asked: "How confident are you?" I'd break eye contact. I'd assert confidence, and then see my recommendation basically get ignored when my boss saw right through me.
Every time I put my reputation on the line I also put my job on the line. Eventually the stress got to me and I burned out.
Realizations
With the distance of a few years I now know a few things I didn't then.
The institutional investment business suffers from the classic principal-agent problem. You took a percentage of profits on the upside, but none of the losses if you lost. Furthermore, if you were a young analyst and put up stupendous numbers early on, you would be viewed as a rockstar and have an accelerated career. Nobody was interested in "average" or "just fine". War stories of huge wins and terrible losses were way more interesting than slow and steady gains in-line with peers.
It turns out that it is easier to act like you have found a great trade idea than it is to find a truly overlooked idea, especially when you have to do it on a schedule.
My mistake was viewing my job as trying to be as right as possible. G played a different game. He understood asymmetric tradeoffs not just on a per-trade level, but also at a career level. I played play the game by its written rules, G played it by its unwritten ones. From a nihilist, cynical perspective, if you viewed this whole game as a series of coin tosses, I was the shmuck trying to perfect the theoretical expected value of coin tossing. G went and found the biggest heaviest coin he liked, tossed it a couple times, and used the momentum from his wins to start a bank. Don't get me wrong – he was right more often than not. But so was I, to a lesser extent. There was skill mixed with luck. The difference was he really knew how to leverage the shit out of his situation and make the most out of his existing skills.
My other posthoc realization is that I had a terrible boss, and by extension was in a terrible firm. It was irresponsible to let a confident employee effectively take over the team portfolio on intellectually dishonest grounds. He should not have delegated his judgment to meta-judgment of his own analysts rather than objectively assess the facts as presented. It might be evident to you, with the benefit of the narrative I laid out, but trust me that you wouldn't have any idea that this is not ideal as a first year buyside analyst who was just happy to be there.
Trafficking in Confidence
A hedge fund is paid a LOT to take risk. Otherwise why would you invest in a hedge fund at all? Just stick everything in an index ETF and call it a day. Our real job was to manufacture confidence for taking risk.
We weren't stock analysts. We were Confidence Men.
The truth is, I suspect that how our fund worked is how most of the world works.
Now that I am out of finance I still see this pattern everywhere in life. The people who shout the loudest get the most attention on social media. Entire mobs converge on the absolute correct policy prescription with anecdotal information and no controlled tests (and of course, no responsibility to follow through on implementation). Politicians and businesspeople respond in kind.
NOTE: THIS IS NOT A CRITIQUE OF CURRENT AFFAIRS. This is just an observation of human behavior that has been in my mind for >5 years.
Finance is rife with Con Men. Finance Con Men know the price of everything and the value of nothing. You are being fed empty confidence whenever a stock ticker rolls across your TV screen with percentage changes quoted down to two decimal places. In the leadup to the 2008 crisis, Con Men conned other Con Men with AAA-rated CDO's, once a convenient lie in the Li Copula was found that gave dumber Con Men confidence. Your own financial advisers have to give you confidence they do not have, because you pay them to.
I think trafficking in confidence happens at much smaller scales too. VC's who extolled the virtues of remote work insisted on meeting entrepreneurs in person before funding them, presumably to gain confidence based on a firm handshake and a healthy dose of implicit bias pattern matching. Overloaded managers ask their direct reports to make estimates and keep up metrics without really caring whether those estimates and metrics are the right ones. Employee OKRs become mechanisms for translating Principal-Agent problems into Goodhart's Law. Even we in our personal capacities instinctively prejudice and change our assessment of products, companies, even ahem blogposts based on social proof: likes, retweets, endorsements, testimonials.
All of these are tools of the trade of the Con Man.
When you really think about it, we are addicted to confidence. We view the less confident as weak and the more confident as leaders. We see confident people as more attractive and employ all sorts of tricks to make ourselves look more confident. We encourage people to "fake it til you make it".
I view this as a modern malaise because we've made it so much easier to engender confidence without substance. Yet our reliance on confidence itself has always been a necessary evil of our existence.
Humans aren't anything special in this regard. The concept of mimicry in nature describes this process. At the beginning, poisonous snakes might evolve some distinctive markings. So predators learned to avoid snakes with those markings. Then, non-poisonous snakes also evolved those markings, because it turned out that those features also helped with survival. Every marking, every behavior, every indicator that has signal eventually becomes noise. I can no longer find it, but there was a excellent essay going around in the aftermath of the 2008 crisis applying Rene Girard's ideas of human mimesis to how the ratings agencies were tricked into approving high risk investments for AAA ratings by financially engineering according to their publicly disclosed criteria. This goes back as far back as the 1500's, when this was known as Gresham's law.
It's all a Confidence shell game, and you're both playing it and being played by it.
I wish I had a prescriptive four step plan to fix it for you, but I'm afraid I don't. The best I can do is to make you aware of it. I'm not confident anything can be done more than that. If that makes you like this piece less, perhaps that is lesson one :)
If the devil is in the details, the greatest trick the devil ever pulled was convincing the world that details don't matter.
Besides what you see, I have confidence in me! - Julie Andrews, in the Sound of Music – not really related, it's just the only "confidence" related song I can think of
Posted on June 14, 2020
Join Our Newsletter. No Spam, Only the good stuff.
Sign up to receive the latest update from our blog.
Related
November 21, 2024