Confessions of a Wall Street Insider

by Michael Kimelman

  “C’mon. Is that a real question? Let’s do Capital Grille. We can always walk back to the office if we need to.”

  “Tomorrow? 4:30?”

  “See you then.”

  We were drinking at Redemption after the close that evening when I told Zvi the plan.

  “I’m coming with,” he insisted, beaming from ear to ear. “Dave loves me. Let’s go in and ask for the world. They’ve got pikers there now. Dave Plate is the only guy at RBC worth a damn.”

  “Dave Plate? The guy who sat next to you at Schottenfeld is there?”

  “Yep.”

  “You told me he was going to put $500K into Incremental and come trade with us.”

  “Yeah, well it didn’t work out. Rick fired him after he blew up.”

  No doubt on your AMD tip, I thought, recalling a conversation outside of Schottenfeld on 48th and 3rd between me and Shankar and Plate. I’d been walking to Grand Central down Third Avenue, and Shankar and Plate were on the corner of 45th and 3rd, in the midst of a seemingly serious discussion. After a quick hello, they asked in a solemn tone if I was in AMD. “Not anymore,” I admitted, after losing $100K-plus on a Zvi “sure thing.”

  Horrible company, horrible call.

  Plate admitted he had lost almost $800K plus in AMD alone last month, Shankar about half that. So I guess it shouldn’t have been a surprise that Plate had now “decided” to go elsewhere after being led blindfolded off the cliff.

  But I still had to wonder why Zvi had never bothered to update me about Plate’s status. After all, we were talking a $500K investment, which, along with Plate’s trading tickets, I had incorporated into our models when pitching Incremental’s projections.

  Given what Abramson had said to me about Zvi, it was only natural that I bring him along to the Capital Grille. We shot the shit over a few drinks, then got down to business and tried to lay it all out for Dave. You could see his barely contained excitement when we spoke. He was leaning in and tapping his right foot like a jackhammer. The pitch was simple. With my trading and operational skills, Nu’s capital and experience, and Zvi’s Galleon research, calls, and very able traders who just missed the Raj/Stevie cut, Incremental was poised to be an inevitable powerhouse.

  Abramson asked what we needed. Seeing that he was already hooked, we went straight for the dream list:

  A minimum of a $100 million to play with, going up to $250 million in twelve months. Tickets at a penny or less (since we charged ten cents, we made at least a nickel on every share after all fees). And a two-tier payment structure—tickets paid … profit and loss (P&L) paid … and no netting, which meant if we made money on tickets and lost money on P&L, he couldn’t offset one with the other.

  Abramson didn’t flinch. He said it didn’t sound unreasonable, but that he would still have to run it past RBC’s legal department. That was more than fair—so we shook on it and ordered another round of drinks.

  One week later, Zvi and I walked out of RBC’s offices at 666 Third Avenue with a term sheet giving us that deal … with just a few constraints. For starters, RBC would have a hand in approving new traders. In addition, per the P&L, RBC got a 10 percent slice, while we took 40 percent, with the remaining 50 percent going to the trader. There was no capital risk, a guaranteed $500K-plus in pure tickets profit per month into our pockets (something which would only increase as Incremental grew), and a brand name bank umbrella we could recruit under, coupled with the promise of all the capital we needed. We wondered with one another if it was really too good to be true—except it was actually both, too good and definitely true. Still, I had questions gnawing at the back of my mind. Would RBC stick with us once they realized they were taking all the risk for only a paltry slice of the upside? Sure, as long as we never lost money, it was a great deal for everyone; but what happened when we lost, say, $1 million in P&L, while also making $1 million in commissions, and then demanded RBC pay us our million (yeah, that’s what “no netting” means)? For now, none of that mattered, it was a future worry. We had our deal, and, in a pure “The world is mine” Scarface moment, we were off to Sutton Place to swim in fountains full of vodka and champagne.

  Flash forward to a long hot August spent getting used to each other, not unlike new lovers early on in a relationship. Both parties were putting their best feet forward and on their best behavior. A good solid August naturally emboldened us, and we started to turn up the gas, as Incremental was up more than 3 percent in September. But almost immediately, we were met with resistance from Dave: intraday limits, holding times, you name it. We were already getting the feeling that RBC’s reins on Incremental were just too damn tight. We had gone on the customary three dates, and it was time for some action. Deep down, I know that Dave understood this was the right move, but he wasn’t ready to stick his neck out for us. Then again, he was politically cautious, a survivor who understood these were dangerous waters, and it was unlikely he’d ever risk his neck for anyone.

  So this inevitably led us up the chain to Dave’s boss, Jeremy Frommer, who was in charge of RBC Cap Markets. The prop desk fell under his jurisdiction. He said we’d been getting a lot of buzz around RBC. We told him we were now asking them for some more rope and leeway. So Frommer decided it was time for us to meet, and suggested drinks or dinner. I thought it was a great opportunity to explain our strategy to someone other than Abramson.

  Zvi and I were waiting for him at Capital Grille. Frommer arrived late, with an aura of brusque arrogance, as if he were doing us a huge favor just by being there. After some inevitable small talk, I launched into our pitch—it was the same one RBC had already signed up for, but this time I offered a lot more details and specifics. Frommer kept playing with his cocktail, mixing and stirring it, then ripping off pieces of bread and slapping on lots of butter. Maybe he was just hungry, but it felt an awful lot like he wasn’t even listening to me. I started to get annoyed, and was three seconds away from asking him if he needed a fucking Ritalin to focus, when Frommer swiveled on his bar stool toward us. He raised his hand, as though he intended to say something original and important, and then proceeded to echo Abramson’s overly cautious approach. He spoke about how he’d come to the conclusion that taking very small bites at the apple—the no-risk model—was the best way to go. He didn’t use the words “piker model,” but that’s what he was talking about—a model that took very little risk, and made very little money.

  It was definitely not the model we were having so much early success with at Incremental. You sure as hell can’t ask someone who ran $75 million at SAC to have a $5000 intraday loss limit and put a deposit down. You’d be laughed at right in your face. But Frommer clearly felt that he’d made his point convincingly and asked—with seeming earnestness—“So, do you guys understand what I’m sayin’?”

  For a few seconds neither Zvi nor I so much as blinked.

  And then, Zvi loudly proclaimed, “With all due respect, you couldn’t be more fucking wrong!”

  The place went quiet. People in nearby booths stopped eating, drinking, and gabbing, and instead leaned over to listen in an E. F. Hutton moment.

  “Jeremy, that model is dead!” Zvi boomed. “All the places that still operate like that are folding or making no money. These aren’t the glory days of day trading. To succeed, you need more leeway, more capital, and more talent to extract money from the market. Right now, you have a once in a generation opportunity to land real talent. A huge restructuring is underway on Wall Street. Guys who returned double digits for ten years and made millions, tens of millions, are suddenly free agents on the market. They need jobs. It’s like Derek Jeter is suddenly available to play for our expansion team, and we can sign him with no upfront, just on a per-hit basis. I’m telling you, you won’t get another opportunity like this. Incremental can build a world-class desk that rivals any of the funds out there, for almost nothing. This dislocation will last less than ten months. We can build something in that time that would normally be impossible. Just think about i
t, Jeremy. You can either keep clinging to that last dead model until RBC takes you off life support, or you can”—Zvi paused here for effect—“grab your balls and become a presence in the space. Either way, you know I’m right.”

  Zvi stopped and took a long sip from his drink.

  He had done it—completely reversed the tenor of the meeting, and knocked Jeremy back on his heels. By the end of the meal, Frommer was ready to play ball. There are moments, few and far between, when I understood exactly how Zvi got his admirers. This was one of them. Frommer went back to the office and greenlit us, and all that October we knocked it right out of the park.

  November went well too. We were up slightly, but most firms were in a tailspin. During December we were flat or down small, yet we still got a large check for our commissions. This left Abramson fuming. Then, as we had known it would, it dawned on the RBC brain trust that they had been shouldering all the risk even as they were forced to eat the losses. Abramson started making noise. Over on the risk management end, he was also doing a lot of head-butting with Zvi and myself over trader risk.

  Ultimately, RBC also wanted veto power over who we hired as traders. A great example of this was Franz “I’m Best Buddies with Fitty” Tudor. Dave Abramson wanted no part of Franz, but Zvi was ready to go to the mat for him. Loved the guy, and wanted us to hire him. Abramson grudgingly agreed to allow Franz to plead his case at RBC’s office … to a young trader on Abramson’s team named Martin Shkreli.

  Shkreli was a freakishly skinny, on-the-spectrum biotech trader who had worked under both Jim Cramer at Cramer, Berkowitz & Co. and Wayne Holman at SAC Capital. While the general public knows Cramer from CNBC, Holman is a legitimate hedge fund legend and, after Stevie Cohen, the second biggest SAC Capital earner of all time. Holman made a fortune taking huge positions in early stage biotech companies like OSI Pharmaceuticals (OSIP), and profiting enormously when OSIP’s acquisition by Astellas was announced. Word on the Street was that Holman had an unholy “edge” with OSIP, speculation which remains unproven to this day.

  Shkreli strolled in wearing dark, poor fitting jeans, a sweater a size too small, and an off-white Hanes T-shirt that had seen better days. He was five foot nine and junkie-thin at 130 pounds. His uncombed, wild hair fell over a disturbed child’s face. The visual presentation had me mentally running through the DSM-IV table of contents before we’d even said hello.

  Deep down, at heart, nearly every person harbors the inner fear that he may not be as good as everyone thinks he is. That he’s one market move or event away from being exposed to the world as a total fraud. This biotech showdown was like Tyson-Spinks. Everyone thought Spinks was one of the best boxers on earth, maybe in history. The undisputed light heavyweight champion and a master of the sweet science. He would assuredly be able to stick and move and keep Tyson off him. But Tyson exposed him as an overrated, historically insignificant heavyweight champ and brutally ripped away the whole façade with one uppercut to the gut a few seconds into the first round. Spinks crumpled.

  That’s what Shkreli did to Franz, peppering him with rapid fire questions about stock values, Net Present Value (NPV) methodology, and ongoing Phase II results. Franz couldn’t hold a candle to Shkreli’s psychopathic level of statistical recall and brainpower … but trading is only about half brainpower. Equally important are discipline, psychology, and perspicacity; figuring out what everybody else knows (and doesn’t know), where everybody else is on a trade, where the market is, what’s priced in, and where the pressure points are. You have to know at what level do the longs give up, do the shorts get squeezed, and does the risk manager come tap-tapping. Shkreli was an NPV absolutist, finance’s equivalent of the data geeks that think fashion, religion, art, or any other matter can be broken down into a precise quantifiable algorithm. The NPV absolutists think investment is a neat little science, and that they can precisely model what future earnings and revenues will be, then discount that back to the present to get the exact number a stock should currently be trading at based on all available information. If a stock is trading for less than NPV, buy it. If it’s trading for more, sell it (while giving yourself some cushion on both sides, of course, because nobody’s perfect all the time). But like most modelers, and certainly all of the guys in the risk departments leading up to the crisis, they all suffer from the same fatal flaw: a model is only as good as its inputs. If the inputs are off (which they almost always are when you’re trying to handicap the likelihood of future events), then the model is much more of a directional compass than a set of exact GPS coordinates. It’s the correct starting point for stock valuation, sure, but it’s only one part of the equation.

  That should have been Franz’s short rebuttal. But instead, he was drowning, drinking from a firehose as Shkreli fired off question after question, cutting off his answers, or dismissing them as ridiculous or wrong.

  I wasn’t sure if Franz was holding back because they thought this was a friendly discussion, or if he wanted to appear civil because a possible capital backer was in the room. Either he didn’t have the tools to fight back or didn’t realize there was an actual fight going on. Whatever the case, the ref was about to step in and call it based on the three knockdown rule. I realized it was time to let Franz know this was a “gloves off” encounter. If the interview/interrogation ended like this, there was zero chance he would get an offer from Dave, and Incremental would look foolish for having offered this stumblebum lamb up for slaughter.

  Franz’s performance might have been pathetic, but I still had a soft spot for him and wanted to give him an opportunity to work for us. And this Shkreli guy was insufferable. I enjoyed his act for about six minutes … then it started to piss me off. He was a bully, and an obnoxious one at that.

  “You can’t accurately model most early or mid-stage biotech,” I interrupted. “Everyone knows that. Look at the moves in that sector. Volatility is massive. Biotechs get cut in half all the time. And it’s always because someone modeled the likelihood of approval or clinical results wrong. You can’t model FDA approval from Phase I with any real degree of confidence”

  Shkreli looked at me and hesitated. He had orders to go after Franz. What was Dave’s partner doing stepping into the fray?

  Shkreli glanced over at Dave, as if asking permission to engage. With a subtle nod back at Shkreli, Dave took the leash off his pet.

  ”Of course you can,” Shkreli sneered back at me, gloves off. “But the very fact that some modeled it wrong means someone could have modeled it right. They just didn’t do their homework!”

  “Except 90 percent of that ‘homework’ involves talking to people who shouldn’t be talking,” I countered. “The biotech mafia is awfully good at modeling a double blind placebo. How does researching that work? How do you read data if there’s no data to read?”

  “It’s not as difficult as you make it sound,” Shkreli countered. “We can model likelihood of success to a statistically significant point and the resulting share value. We do it all the time.”

  “Oh, oooookay,” I said sarcastically—my gloves coming off as well. “So you’re the one on the other side of all those trades that go wrong. Now I understand. That explains why you’re running a $5 million prop account for RBC instead of running three billion for Stevie or Oracle. I gotcha.”

  All the stereotypes I harbored about quants and geeks came bubbling to the surface. Those guys could debate and massage data, peer review and tinker with models until the sun rose—sure—but ad hominem personal attacks, sarcasm, and humor—things that were a big part of big boy life on the Street—completely threw off their game and scrambled their antenna.

  “It is a science, and that’s why Wayne and the rest of the true stars in biotech know me on a first name basis!” Shkreli shrieked loudly. “The fact that you don’t even know that makes me question why Dave would be in business with you in the first place!”

  I had to give Shkreli that much. For a quant, he was holding his own.

  �
��Dave, you can call off your attack dog,” I said with a smile. “He’s made his case.”

  “Attack dog? I’ve merely proven this trader is incompetent!” Shkreli said, motioning to Franz. “But I can show you an attack dog if you’d like.”

  He gave me his best “tough guy” stare. It was like watching a starved poodle with rickets trying to impersonate a Great Dane.

  “You’ve proven nothing other than an overreliance on mechanical valuation formula that only works with suspiciously perfect info,” I told him. “And even then it doesn’t account for shifting outcomes, like Phase II results, or changing reimbursement guidelines.”

  There was something more I wanted to say. I knew something I shouldn’t. But then I did. Shkreli had a way of making you set aside your better judgment.

  “And if you speak to me like that again, I’ll walk around this table and choke you out right here in this conference room,” I added quietly.

  The physical threat snapped him out of his catbird seat. He stuttered for a few seconds and began to turn red. Then Dave stepped in.

  “Thank you, Martin. That’s all we needed you for. Please let me finish with Mike and Franz alone. I’ll talk to you in a few minutes.”

  Shkreli got up and walked out, turning back to glare at me once more from the doorway. I gave him a nice big smile in return.

  “Okay, that was … interesting,” Dave said. “Mike, I’ll give you a call later today and we can discuss. Thank you guys for coming down. I apologize for Martin, he’s really smart—just a little rough around the edges and immature. Either way, we’ll talk soon.”

  “High IQ or not, I’d be careful of anyone who has absolute faith in anything,” I told Dave. “We both know how much of this comes from real experience.”

  “Well, I’ve got ‘absolute faith’ my wife is going to kill me if I don’t head home and give her a break from the kids. So let’s chat tonight or tomorrow, yeah?”

  We left.