And by harder, I'm also including the costs for infrastructure. With futures, you can pretty much focus on just a few exchanges, but with stocks it seems like you need infrastructure for multiple exchanges, multiple fees, and have more routing decisions to make, more communications infrastructure, etc. This is more of an opinion question on the economics and difficulty of the problem. I'm not sure there's a right or wrong answer. I am just seeking opinions about the degree of difficulty, barriers to entry, etc.
The barrier to entry in futures is higher IMO. Just the simple cost to purchase one futures contract is higher than most every (maybe with the exception of Birk-A) equity. That plus the fact that futures is pretty much a 24/7 game means you have less hardware down-time, bodies need to be awake during the hours that you run and they take the speed game seriously because you can't sit back and be slower to fight over rebates.
My understanding was that the rebate game itself was rather competitive. I'm not making the connection between slow and rebates; if you could elaborate that would be helpful.
stocks are not for rebates only. there are 1000s of stocks and only a few futures. So making markets in liquid stocks/etfs is the easy money.
Hard to explain but most tend to confuse latency and duration of positions when they think about HFT. < 10ms used to be fast. Today that os extremely slow (latency). That said, there are plenty out there doing OK in the sub-5ms space... As long as you keep up with the times and keep your speed the same relative to others. A profitable strategy at 3ms round trip today may need to be half of that in 5 months time. The really fast guys are the ones looking at order flow and the guys with matching engines. The rebate game is competitive but only on the surface. If your strategy requires you to be first in line I say you are lucky while you were first but it is only a matter of time until someone else beats you there.
Not only am I quickless but I am clueless. What is the objective of HFT? To take the bid ask spread? What difference does it make if you are doing 5 min round trip and others are doing 3 min round trip? Detail would be appreciated.
spoofing and frontrunning is about 99% of it, as far as i can tell. when both is stopped, HFT will stop
Out of say a 100 trades per minute, when the market isn't moving much, let's say around 50/50 of the market orders will be at the bid/ask. If you're an HFT and you are in queue for the bid/ask, the market orders are to you. So you get 50 buys and 50 sells, and you earn the spread. So yeah that's about it. The key is to be first in queue, and that's where the speed competition is. It's almost impossible now to determine how the edge in speed is achieved. Perhaps superior cooling of the fibers connecting the box to the exchange, lol.