The flash crash of May 6 explained - one human, thousands of algorithms involved | Hacker News Reader

The flash crash of May 6 explained - one human, thousands of algorithms involved

ft.com

6 pointserikstarck10 comments

Here is the original report. Far more useful, in my opinion.

http://media.ft.com/cms/5b341edc-cd80-11df-9c82-00144feab49a...

You have to love the way these electronic futures (e-Mini S&P's - 50x leveraged) are geared. One trader's 4.1bln stone in the pond triggers a chain of ripples that causes $900+bln losses. Its gotten so extreme that "The current average daily implied volume for the E-mini is over $140 billion, far exceeding the combined traded dollar volume of the underlying 500 stocks." - wikipedia

That said I don't think the problem lies in leverage. I'm sure if this Waddell & Reed trader had tried to pull a fast one in the trading pits where they trade the 'big boy' S&P futures (500x leveraged), the effect would have been much more muted because fellow traders would have pinpointed the seller and reacted accordingly. When the trades are coming in electronically, the 'enemy is faceless'...

For some related fun listen to the squawk in the pits:

http://catastrophist.wordpress.com/2010/05/09/ben-lichtenste...

(The excitement/fear is palpable - bear in mind Ben is a professional watching billions wasting away)

I'm interested in the headline, but really don't want to have to register to read the content.

Try this: http://www.ft.com/cms/s/0/8ee1a816-cd81-11df-9c82-00144feab4...

EDIT OK, try Googling "flash crash single order" and clicking the link from there.

Sorry, I didn't realize it was a log-in article. I didn't have to log in, got there from Google.

EDIT: why are they (FT) doing that? Coming from Google doesn't require registration. Coming from HN does. Weird.

Because if they didn't, Google wouldn't crawl the article, and then users wouldn't find it through google.

IMO, Google needs to stop that.

Well, if they want the traffic from Google they have to put out the content. Google is very clear on that and I agree with them. It just doesn't make sense to close out traffic from other sites. A visitor is a visitor is a visitor, coming from Google or not.

According to "suit logic", it absolutely makes sense. hey're trying to upsell the visitors. Whether that's actually profitable, I don't know.

They want to have it both ways: have google crawl and report there's a page there without registering, and then try to upsell the humans that come by. It's sleazy.

Alternative read? http://www.msnbc.msn.com/id/39460135/ns/business-stocks_and_...

disabling javascript allows you to read the article without the notification :)

Reply on news.ycombinator.com