strongvigilance

What kind of depreciation do you get on crypto mining hardware though? Will that setup still be profitable in x months, when better performing hardware exists?

Interesting - I have looked into the API, though TT tech support believe they can fix the issues I'm seeing. Any reason for choosing F#/WPF over C#/Winforms?

And agree 100% on Xtrader - very strange business decision to discontinue a 15 year old piece of software than thousands of users are willing to pay $1200 per month for!

Interesting - so do you think it would be possible to have a JS/HTML5 app refresh at 60FPS?

Not really sure, as it's not my field, but the graphics requirements are not about "drawing" - more about rapidly updating multiple grids of numbers without any lag.

There have always been middlemen, as, especially with less liquid instruments, there won't necessarily be equal supply and demand from "natural" buyers and sellers. As markets have gone from floor trading, to point+click on computer screens, to automation of different levels of sophistication, efficiency has improved - people complain about HFT stealing pennies, but in the past, someone was stealing quarters. The difference is that the skills required have changed tremendously - a few years ago, it was all about being well connected or going to the right school - now it's all about how good your code is.

It's not particularly surprising that the loudest anti-HFT voices are guys whose profit margins have been killed by it.

Disclosure: my profit margins have been killed by HFT...

Seems like quite a strong possibility, given his background:

http://en.wikipedia.org/wiki/David_Nutt#Dismissal

I don't think the actual order type was a secret, just that some HFT firms figured out a way to use it which gave them an advantage over other HFT firms. The firms who didn't understand it lost out.

Like many vocal HFT critics, he was trying to play the same game, and was outcompeted.

I believe the profits have dropped a lot recently, as competition has driven down margins for everyone involved.

http://www.businessweek.com/articles/2013-06-06/how-the-robo...

Generally it's possible to colocate servers in the same datacentre as the exchange servers. This isn't really a competitive advantage though, as anyone who needs latency that low will be doing the same, and the costs will be relatively insignificant compared to salaries of people who can build the required systems.

Competition is already high, and colocating is not that expensive for anyone with the ability to compete.

A lot of the loudest criticism of HFT is that it's too competitive - a lot of people who used to make a comfortable living from the bid/ask spread are no longer able to due to computers driving down profit margin.

Why have a limit on how close people can be to the exchange? For one thing, everyone would just colo at exactly the minimum distance, achieving nothing. For another, ability to colo is not really an issue. If you wanted to create an HFT startup, colo is not going to be a major cost compared to hiring developers.

As far as I'm aware, the exact time of release of the FOMC statement can vary by a few seconds. Nanex seem to have assumed that the release comes from Washington, but I would guess that one of the agencies that receives the embargoed data has permission to distribute it at multiple locations simultaneously. That seems like the simplest explanation for Chicago and New York reacting at the same time.

Yes - most journalists/analysts prefer to be wrong together than wrong alone, so there is usually a lot of groupthink. The previous fed minutes suggested that most members wanted to wait to taper, unless data improved, and data didn't improve. It seemed obvious to me that the taper was nowhere near a done deal.

Indeed - it's worth noting that on the non-farm payroll figure (which is closely watched by markets) this month, most markets jumped in the wrong direction a split second early, before violently reversing.

Indeed. In fact, when economists talk about full employment, they don't even mean 100% - it's thought that at least some unemployment is needed to control inflation.

http://en.wikipedia.org/wiki/Full_employment

Perhaps this is partly due to the multitude of parties, which in turn is due to the nature of the electoral system?

In a two party system, a lot of people feel like they are voting for the lesser of two evils, so negative campaigning is presumably far more helpful to your cause.

If you have to compete with multiple parties for votes, you need to offer more than simply criticism of the opposition.

Indeed - the current proposal suggests that the persecution was only wrong because he was a war hero, and not because of the injustice of it.

I'd imagine there must be some victims who are still alive today.

London has a big financial sector, so a lot of the big pay packets probably come from there - hedge funds, banks, HFT etc.

See here:

http://jobs.efinancialcareers.co.uk/Information_Technology.h...

To be fair, this kind of performance would probably be seen over any sample of the brokerage house clients.

Even professional mutual fund managers systematically underperform the market.

To an extent - I'm a trader first, programmer second - learnt the latter to help with the former.

I'd second the Harris book - very good intro to trading and markets.

A couple of years old, but this is (IMO) a much better list: http://quantivity.wordpress.com/2010/01/10/how-to-learn-algo...

The 30+ generic C++ books in the posted list says a lot, and many of the more finance-oriented books on the list are very low on signal-to-noise.

Like the numerous 9/11 conspiracies, it fails the Occam's Razor test. Even if there was a conspiracy, how many people would have to be in on it, and how would you keep it secret? And if there were such nefarious people in government/industry, wouldn't it be easier for them just to pay someone to set off a real bomb, rather than trying to stage an explosion?

Structuring a trade to reduce downside risk while profiting from unanticipated random events is easy - just buy options. The problem is that everyone else knows this, so they're priced accordingly.

It's slightly sad when science journalism relies on "an expert says", without feeling the need to reference any data.

On the about page, they suggest making a pull request if you want to add to it, so I guess it's a work in progress.

Repo is here: https://github.com/overthecs/overthecs.github.com

I imagine "specs like this" is referring to the Core i5 rather than the display.

Generally speaking, yes - the spread is the cost of buying immediately.

There is no single price for a stock at any given time. There's a bid price, which is made up of unfilled buy orders, and an ask price, which is made up of unfilled sell orders. The difference between the two prices is the spread.

Imagine a stock is bid at $1.00, and offered at $1.02. If you want to buy it, you can have it for $1.02 immediately, since the ask price represents sell orders. If you want to avoid paying the spread, you could place a limit order at $1.00, hoping that someone less patient will sell to you. Alternatively, you could place a buy order at $1.01, which would make this the new market bid price. Again, your order will only be filled if someone else wishes to sell the same price.

Not really - a lot of HFT earns money from liquidity provision, where a higher tick value would increase profitability, since they're entering and exiting trades passively attempting to earn the bid/ask spread.