Bloodless

26

C​lifford Masolino, forensic accountant for the Federal Bureau of Investigation, returned from lunch to find a new assignment waiting for him, delivered from Georgia by special courier, with an attached note written by one Special Agent A. X. L. Pendergast.

As Masolino took his seat in the windowless basement office, he wiped the grease off his hands from the gyro he’d just eaten. He used a paper towel, a roll of which he kept handy, because he was large and soft and had a tendency to sweat. Pendergast, Pendergast... the name was familiar, and he vaguely associated it with something unpleasant. The reason suddenly came to him as he opened the note: that crazy episode years ago at the New York Museum of Natural History, where a bunch of people got killed and, if memory served, there was a cover-up. A special agent named Pendergast was involved with that, and... and now Masolino recalled a spectral image of the man. Masolino had been a rookie forensic accountant in New York at the time, just getting started with the FBI, and he’d assisted in analyzing the museum’s accounts after the bloodbath, where they found significant fraud involving donated monies. It was Masolino’s first case and he had done well. Very well. Those were heady times indeed.

That was years ago. Pendergast’s note was written with indigo-colored ink in an elegant script:

The handwritten note was rather an eccentric touch — didn’t the fellow have a computer? — but the project itself was straightforward enough, nothing different from what Masolino had done a thousand times over the past decade.

He plugged in the first hard drive; examined it for viruses, malware, and rootkits; found it clean; and then copied the contents to his powerful, air-gapped Mac Pro, a version ordered to his own specifications, which included a 2.5GHz 28-core Intel processor, 1.5 terabytes of RAM, two Radeon Pro Vega II Duo graphics cards, 4 terabytes of SSD storage, and an Afterburner card. The monster was brand-new and had cost the FBI more than fifty grand — a sign of his value to the organization.

As he scrolled through the files, Masolino noticed that this guy, Ellerby, hadn’t encrypted them. This was a pretty good hint that what he had been doing was probably legal. Of course, that wasn’t saying much: a lot of the dishonest, manipulative, sleazy shit that traders did was legal, which was the main reason why Masolino didn’t personally invest in the market. The last thing he wanted to be was another chump. If small-time investors knew how they were being reamed out by the big boys every day, they’d never invest again.

He opened one trading account and looked it over, just to get a feel for it. In this account, Ellerby was trading Big Board stocks, Dow Jones Industrials. It all looked aboveboard and he was openly trading in his own name, not through some offshore entity or LLC. The first thing Masolino noticed was how small many of the trades were, and how short the duration. Almost all were in and out in less than an hour. But in the end, the guy had made a shitload of money.

Masolino went through the trades, one by one. No giant killings here on any single trade, but significant amounts that nevertheless added up. With the extra leverage in the puts and calls, he was making some nifty profits. Now that Masolino was looking more closely, he could see Ellerby had made a profit not just on most trades, but on every single one.

This was crazy. That simply didn’t happen. So it wasn’t legal, after all: there must be a scam hidden in here somewhere.

Delving deeper into the specifics, noting dates and times, he was struck again. Going back farther, he found hundreds of even quicker trades, in and out in a minute or less. But this was not computerized high-speed trading: Ellerby was doing it online by hand. Masolino called up matching historical data for the stock price movements bracketing each trade and was further astonished. Ellerby’s trades came right before a stock moved sharply upward, in such a way that the trade produced a sweet little profit. The data had all the earmarks of an algorithmic trading program... but if this was such a program, it was worth billions, because it was never wrong.

No — that obviously couldn’t be it. This was something illegal, for sure. This was trading tied to some information source, an inside flow of data, probably from the trading desk of a large investment bank, which would know what time the bank would be buying or selling large blocks.

Masolino let out a slow, relieved breath. So it was typical insider stuff after all — and, hotel manager or not, Ellerby was playing it like some back-office IT drone... except he was too stupid to hide his dirty trades amid a lot of financial noise. Masolino had been worried for a moment there, but this was going to be easy. All he had to do was identify which investment bank happened to be trading large blocks at the time these trades were made, and go from there.

The activity he was currently examining came from a few years back, so Masolino moved his focus closer to the present. He quickly noticed that the trading pattern had changed a few weeks ago. At that point, this Ellerby started making bigger trades, for bigger profits, and held on to the trades for longer — sometimes as long as an hour.

Masolino had a number of programs that he’d coded himself, and now he ran one that compared block trading of equities by large investment banks to Ellerby’s trading, looking for the match.

There was no match.

This was odd. Another thing that was odd was that the stocks were always Dow Jones Industrials listed stocks.

Now he began comparing the time stamp of each trade with stock price movements in general. Ellerby’s trades usually came during periods of high volatility and took advantage of small fluctuations that occurred sometimes within seconds after the trade. How was this possible? The trades always came right before an uptick in the stock. Not huge upticks, but decent enough to make money — although the money made in the last three weeks had increased dramatically.

He couldn’t get around it: this was the classic pattern of someone getting tips from an insider with access to private information, probably an influential stock picker or newsletter. Masolino had a database that aggregated the stock picks of thousands of such sources, and now he ran it against Ellerby’s trades.

Nothing. No match.

Now, that was damned strange.

Perhaps the trades were based on acquiring advance news about companies, like earnings reports or drug approvals, that hadn’t yet been released but that some insider was privy to. He had a program that handled that, too, comparing the trades to news reports involving the same stock.

That came up empty as well.

Masolino ran Ellerby’s trades against every program in his digital toolbox that matched trades with outside events: merger announcements, lawsuit filings, earnings reports, commodity movements, political news, and a host of other things that move stocks abruptly — and still couldn’t find a pattern.

Next he looked at who was buying and selling these stocks just before or after Ellerby’s trades. If Ellerby knew someone who bought and sold large blocks of stock, and he acted before that person did, he could make a profit from the movement of the later, larger trades.

Again, nothing came up. No big blocks of stock were dumped or purchased, no insider selling or buying from company executives. Ellerby simply seemed to have anticipated ahead of time an uptick in the stock price and bought into it, then sold at a profit. He could have made much larger profits by buying and holding some of these stocks. But he never did. The trades were quick, simple, and unremarkable — and every single damn one made money.

As Masolino went forward in time, he again saw the break that took place three weeks before the end. He saw it in every trading account. In recent weeks, the trades got bigger, more profitable, and longer.

Four hours later, Masolino, bathed in sweat, a pile of damp paper towels on the floor behind him, hands trembling, finally shut down his system. It was only two o’clock in the afternoon, but he was going home early to have a stiff martini.

Ellerby’s trading went back years and years, via every imaginable financial instrument, all over the map: those same quick little trades making modest profits. Every trade was legal, or so it seemed. Masolino could think of only one answer: Ellerby was a stock-trading genius the likes of which the world had never seen. Given the short time frames on so many of the trades, he must have developed some incredibly powerful mathematical quantitative trading algorithms that monitored markets and traded accordingly. An algorithm like that — that never made a mistake and always made a profit — would be the holy grail of Wall Street. But such programs, no matter how powerful or ingenious, could never be 100 percent accurate. It was impossible, given the random fluctuations of the market, to ever be perfectly accurate. But the hard drive held only records of transactions — no indication of how the trades were identified and executed and no algorithmic trading program.

And by the end of it all, Ellerby had amassed a paper fortune of close to $300 million. A hotel manager. Three hundred million. And $200 million of that had been made in just the last three weeks.

Christ, Masolino needed that martini.