Flash Boys Chapter 2 - Brad Katsuyama and the Phantom Stock Market Problem
Chapter 2 of Flash Boys is where we meet Brad Katsuyama. And honestly, this is where the book really starts cooking. Because Brad is not some Wall Street hotshot from Goldman Sachs. He’s a Canadian guy from Toronto who ended up in New York almost by accident.
The Nobody from a Nobody Bank
Brad worked for RBC, the Royal Bank of Canada. Ninth biggest bank in the world. Sounds impressive, right? But here’s the thing. On Wall Street, nobody cared. One trader who moved to RBC from Morgan Stanley described it perfectly: “Holy shit, welcome to the small time!”
Brad himself said RBC was “a nobody.” The bank kept trying to make a big push into Wall Street, but hardly anyone noticed. It was like showing up to a school play audition wearing a carrot costume. That’s Michael Lewis’s comparison, and it’s perfect.
Brad arrived in New York in 2002. He was 24. Never even been to New York City before. Everything about America shocked him. The excess. The debt. Back in Canada, debt was basically a dirty word. In America, a real estate broker told him he could “afford” a $2.5 million apartment.
When the Screens Started Lying
For his first few years, everything was fine. Brad was good at his job. Really good. People called him “the golden child” and said he’d end up running the whole bank someday. He traded stocks. He trusted his screens. When the screen said 10,000 shares of Intel were available at $22, he could buy them. Simple.
Then RBC bought a company called Carlin Financial for $100 million. This is where things got weird.
Carlin was an electronic trading firm run by Jeremy Frommer. This guy walked around with a baseball bat, taking swings while the shoeshine guy tried to polish his shoes. He told college students his secret to success was knowing his friends were flying coach while he was in first class. Real charming guy.
But here’s the problem. After Carlin’s technology got bolted onto RBC’s systems, Brad’s screens stopped working right. He’d see 10,000 shares of Intel at $22, hit the buy button, and the shares would vanish. Gone. Like they were never there.
The Ghost Orders
This was not a small problem. Brad’s whole job was to sit between big investors and the market. If he couldn’t trust what his screens showed him, he couldn’t price anything. He couldn’t take risks. He was basically flying blind.
He called tech support. They said “user error.” He explained that all he did was press Enter. Kind of hard to mess that up. They sent more people. Same blank stares. Finally they sent the developers, the so-called “Golden Goose” team. They blamed it on distance, network speed, other traders.
So Brad made them stand behind him and watch. He counted out loud. “One. Two. Three. Four. Five.” Nothing moved. The shares were right there on the screen. Then he hit Enter. Boom. Everything vanished. Prices jumped.
He turned to them and said: “You see, I’m the event. I am the news.”
They had no answer. They disappeared and never came back.
The Financial Crisis Changes Everything
By the end of 2007, Brad calculated the problem was costing RBC tens of millions. Then 2008 happened. The whole financial system melted down. His body started falling apart. Sinus surgery. Blood pressure through the roof. Kidney specialist. He was ready to quit. Sat down with his fiancee every night picking cities to move to. San Diego. Atlanta. Toronto.
But then RBC fired Jeremy Frommer and asked Brad to fix the electronic trading mess. As he interviewed candidates from big banks, he realized something important: nobody understood electronic trading. Not at Goldman. Not at Morgan Stanley. The people selling this stuff were just “front men” with no idea how the technology worked.
Proving the Market Was Rigged
Brad brought back a programmer named Rob Park. Together with a small team, they convinced RBC to let them spend up to $10,000 a day running experiments on the stock market. Not to make money. Just to figure out what was going on.
They tried sending orders to one exchange at a time. It worked perfectly. They got everything the screen showed them. So it wasn’t the exchanges cheating individually.
Then they sent orders to multiple exchanges at once. And the more exchanges they included, the less stock they actually got. The shares just disappeared. Except at one exchange, BATS, where they always got 100% of what was offered. Rob thought BATS was just a really great exchange.
Then one morning, Rob was in the shower. He remembered a chart showing how long it took orders to travel to each exchange. The bars were all different heights. Different distances meant different travel times. What if that was the answer?
The THOR Moment
Here’s what was happening. Brad’s orders arrived at the closest exchange first. Someone there saw the order, raced ahead to all the other exchanges, and bought up the shares before Brad’s orders arrived. Then they sold those shares back to Brad at a higher price. All in milliseconds.
And why did BATS always work? Because it was the closest exchange to Brad’s office. His orders got there first, before anyone could race ahead.
The fix was backwards from what everyone expected. Everyone said you needed to go faster. Brad’s team went slower. Allen Zhang, their programmer, wrote code that added tiny delays to orders heading to closer exchanges. The goal was to make all orders arrive at all exchanges at the exact same time. No more racing ahead.
They tested it. The screen lit up green. It worked. Rob ran to Brad’s desk shouting, “It worked! It fucking worked!” They called it Thor. It became a verb on the trading floor. Traders would yell, “Thor it!”
The Tax Nobody Could See
Brad took Thor to big investors like T. Rowe Price. He showed them what happened when you could buy stock without being front-run. On a test trade of 10 million Citigroup shares, they saved $29,000. Sounds small. But daily US trading volume was $225 billion. At that rate, the invisible tax was over $160 million per day. Your retirement fund. My retirement fund.
The scariest part? Brad had been dealing with this for almost three years. He knew he couldn’t be the first to figure it out. So what happened to everyone else? Simple. Everyone who understood the game was making too much money from it to blow the whistle.
What Brad Did Next
Brad had a choice. He could use this knowledge to make money. It would have been easy. Instead, he decided to go on an educational campaign. Tell investors they were being robbed. Try to fix the system.
That took guts. Especially from a guy at a bank nobody cared about. Especially when powerful people inside his own bank wanted to do the opposite and sell their customers out to high-frequency traders.
As Rob Park put it: “The only card left to play was honesty.”
And that’s exactly what they played.
Previous: Chapter 1 - Hidden in Plain Sight