When did I ever say price discovery is bad? Creating latency opportunities by stuffing the pipes is what I have a problem with.
depending on who you trade with you can have access to HFTs via ALMMs like ATD,and others. if you had these routes you would see what type of edge they give you over other traders who dont have them. I started a company to address all of these issues for discretionary traders. If you dont scalp, for pennies, why the beef though? what is your strategy what do you do? did you know, depending on your broker you may be able to colocate for free or even a couple of hundred bucks? did you know that you could get into this space for very low cost via private cloud? so dont tell me now you cant afford a server or you cant afford space at NY4. you do have options. the same if you were a race car driver and you have a shitty car, and I was willing to lend you my high performance car, for a few bucks a mile, you would take that risk wouldnt you? the race is worth 1M say... as a businessman, you would right? if you wouldnt, I would tell you to pack your bags and go home. you are equivalent to a telephone trader failing to see why you should spend a few hundred on a direct access platform and further why you should have to pay for full market depth. it is what it is you know? if you are a trader with a model, then grab your balls, upgrade and compete.
Access to HFT's? What do you mean? Dark routes? What does your company do? People here have reading comprehension issues. I've been a manual and partial auto trader for many years. I make a good living at this. You defending quote stuffing is absurd to me. I have no problems with HFT/algo's making markets, or trying to beat each other at normal arbitrage games. Those aren't abusive. Quote stuffing is. During the flash crash, and on a few occasions since, there have been street wide data problems due to excessive quotes in strained markets. This abusive practice directly affects you and I.
The problem, according to the Nanex article, is system stability. Massive quote-stuffing ties up so many resources that the system engines bog down for all stocks processed by that engine, and potentially could cause it to collapse (not my opinion - stated in the article). If this is not a direct threat now, it's obvious that the clock is ticking. Just expand the capacity of the quote engines, you say? Well, it doesn't take a genius to see that the ceiling will just have to keep going higher & higher. The cost of expansion at some point will become greater than the edge gained by the competitors. So what you have, in effect, will be a transfer of wealth from the exchanges to the firms via expansion-cost. The exchange is not obligated to make sure you are profitable. So if you need to be able to quote-stuff more than the exchange can handle in order to maintain profitability, the onus is on you to find a better strategy - not on the exchange to expand. A different solution has to be found, because there's no way around it: capacity-expansion won't work.
The solution is the exchange can block your port from sending messages if you are coming close to overloading their system. They already do it. 8 pages in and no one can explain how this devious strategy Nanex is talking about 1) could possibly help any trader out there get an advantage or 2) how it hurt any other traders. The explanation for what happened here is a few pages back. But sadly it is benign and doesn't include the bogey man so it is ignored.
I suspect that if it were so simple as port-blocking, sophisticated information technology companies like Nanex would not be writing articles of concern about it.
Sarcasm much?? OK I'll bite. The "fake orders" assumption is total BS. The misconception is that most think that HFT just spams the markets with quotes/orders. Reality is that most HFT/automated traders measure their exposure as positions + orders and an order (or a quote) is on the books until the confirmation of cancel is received back. Think of a simple market making strategy in the financial space. You focus on four names: GS, JPM, BAC and C (picking random names). Assume for this that each stock is trading at ~$100 (or a spread of $100.01 x $99.99 with last trade at $100 even. If you are making a market in these four names you'll start out by quoting both sides of all four names (so you'll offer $100.01 in all four giving you $400.04 of potential short exposure and you'll bid $399.96 - which is an "almost" market-neutral strategy. If the market gaps up you'll get filled on your offers and be net short your basket of financials however the bids are still out there. If you cancel a single bid now you have $400.04 of short exposure and only $299.97 of long exposure - you aren't neutral and your portfolio isn't balanced. The burger king analogy is wrong in that once a quote hits the tape it's a real/live order. Once you tell the cashier you want 367 #5 combo meals she'll ask you to pay for it - if you don't pay it never hits the kitchen. Same in trading - once that quote hits the tape you put your money where your mouth is and had the buying power to assume that order would be filled. Most often the chatter people refer to as "fake orders" is actually a trader trying to hedge or balance their portfolio - if you are generally short homebuilders you'll be bidding that sector (or a proxy) to hedge. The only time those bids change is to re-balance or if the position is closed - then there's no reason to hedge. If that didn't make sense I'm happy to explain more - but just like the burger king order doesn't hit the kitchen until it's paid at the register... your broker checks and puts a hold on your buying power before you can send an order.
simply posting orders on both sides will get your ass kicked. try it some time. thats why they have to play games.