[BookmarkSaver]

I've had a couple people who work in high frequency trading say that Flash Boys is extremely inaccurate, but they still recommend The Big Short. I've not really heard anything significantly negative about TBS though I'm sure there are probably disputed opinions about it out there. And while the character narratives in TBS could be entirely fabricated (I mean, I'm sure they aren't totally false), the underlying issues and events line up with my previous understanding of the crisis.

[yummyfajitas]

I'd probably be such a person (former HFT). But here's the thing - I know fuck all about mortgages.

I know a bit more than the average person, e.g. I know the actual mechanics of a CMO. Certainly I can see past the "evil vampire squids pooping toxic waste onto innocent homeowners" story the media pushes. But if Michael Lewis wanted to fool HFTs giving the Big Short a casual read, he could.

In much the same way, I'm sure mortgage people could potentially see through BS in the Big Short (if such exists) but couldn't spot it in Flash Boys. I also suspect that HFTs (read: computer geeks, the sort of folks who read HN) are far more vocal on the internet than mortgage guys.

[at5]

Not much in the way of bs in the big short tbh. A little dramatisation of the characters probably but hey, gotta sell a book right? Facts were pretty spot on. The book atleast; no clue about the movie.

[remarkEon]

My father worked at one of the Hedge Funds referenced in The Big Short for about 15 years, and was there during the collapse. I've talked to him ad nauseam about these books and the only one he didn't like was Flash Boys, but calls The Big Short "sort of scary" in how accurate it was.

In reference to Flash Boys he said that the technical aspects of how HFT works were so poorly illustrated that he couldn't really take the book serious and felt that it would limit his ability to further consider future Michael Lewis books. Wish I understood more of that, but his point was, I think, that it's not as nefarious as it's often made out to be.

[Nrsolis]

Look, HFT as a concept is already black magic to people who trade.

It's a fundamentally simple idea but when you actually try to apply it to a real living breathing market, it becomes indistinguishable from electronic warfare because that's what it is!

A lot of the shenanigans that people point to are just different techniques to try and suss out the market microstructure and are in effect no different from trading techniques that evolved in the 1920s.

The best HFT book I ever read was "Reminiscences of a Stock Operator" because if you look at what he's saying and then replace the people in the book with electronic agents, you end up right where we are with HFT.

[remarkEon]

Thanks for the book recommendation. I'll definitely take a look and pass it on.

[kevstev]

Interestingly, I did a brief detour from algorithmic trading/HFT to work on the exotic derivatives desk (which traded CDS, CDOs, etc) at GS in 2007 for a year. I didn't like GS very much, I felt like I gave them 130%, and they gave me shit- I was suckered into a bad role that was clearly considered a cost center. I was also almost autistically focused on the housing market, wondering how I was making good money but couldn't afford a house by traditional standards and reading tons of data and blogs to figure out who was mad (me or everyone else). But anyway...

The Big Short was a lot more accurate. Moreso because Lewis's thing is to tell a story from one side, usually with some kind of reluctant or unlikely hero, and the Big Short fit that structure a lot better. Everyone, and I mean EVERYONE, as in 99+% of the US population, was all in on the housing market. I mean everyone was chugging the kool-aid. Insurance companies (IE AIG), pension funds, cities, all looked at these securities as a way to get safe returns with an attractive yield. Where things really started to get off the rails with them is that they weren't looking at the underlying loans themselves- they trusted the rubber stamped AAA rating that the ratings companies were putting on them. What's worse, is that the first line of defense against this sort of thing is supposed to be the loan officers- but they looked the other way as they pushed NINJA (No Income, Job, Assets) loans through the system. And while "Wall St" gets yelled at for being greedy, there were maybe a few thousand people on Wall St who got rich. The number of mortgage brokers and loan officers who were making deep in the 6 figures, often with just a year out of school was in the hundreds of thousands, if not millions- I know because many of my classmates made me extremely jealous. Anyway, my point being is that the Big short was quite accurate.

So why was Flash Boys less so? Well, the number of players involved is a lot higher, and its not so easy when you really look at it to paint a clear picture of good vs bad when you really understand the stuff. First off, he only seemed to talk to people at IEX- Katsyuma and his crew, who clearly had a vested interest in making themselves look like heroes. IEX actually launching was a more or less non-event, they were merely one of 37 venues to trade at that point, and in Oct' 13, they barely registered as a blip until Flash Boys came out. Also, HFT is an ever-changing game, the Thor strategy was widely known by mid 2009 and every major sell-side algo had it implemented by then or shortly after. My firm happened to call it "sync-take" (IE synchronized "taking" of liquidity).

My problem with Flash Boys is that it just told so little of the story, and clearly wanted to make Katsyuna a hero figure protecting the little guy from the big bad HFT's. But there are two huge parts of the story that I think he should have at least mentioned. The first part is that the desire for speed and great lengths to get it are nothing new on wall st. CGuys using phones to get ahead of bucket shops at the turn of the 19th century, the whole reason guys paid big bucks to become floor traders, SOES bandits on Nasdaq in the 80s, are all just a part of the very long story of people trying to get a speed advantage to trade on the stock market. It just became a lot scarier because its on computers (many HFT strategies are actually fairly simple- the whole point is to be simple as they just want to use their speed to ensure as close to a risk-free profit as they can.

The second major part of the story that they don't talk about it is how corrupt the markets used to be. Floor traders were shady and would literally front-run orders. That is more or less impossible now, there is a clear audit trail. But even aside from that, which did happen, but was relatively small potatoes, is the fact that the "spread" you would pay when buying and selling stocks used to be huge. Transaction costs used to be a huge part of trading, from your broker charging $50 just to make a trade, to the huge bid/ask spread that went in the pockets of market makers. There used to be armies of "traders" in the 80s and 90s that brought home 6 figure salaries when that really meant something that were just automated out of a job in the 00's. Your typical spread on IBM is now maybe a few cents, where it used to be over a dollar in the 90s. Your typical brokerage now charges about $7 a trade, with some as low as $1 (Interactive Brokers- which I use).

This is a topic I tend to ramble about, which I have done here, but hopefully that paints more of a picture. If you have questions or want me to expand on any particular area, let me know.