> Your argument is that latency-based front-running isn't front-running because the front-runners "earned it" via whatever system they built to front-run the tr

> Your argument is that latency-based front-running isn't front-running because the front-runners "earned it" via whatever system they built to front-run the trades. It’s not front-running. > This is the key part I disagree with. You’re not just disagreeing, you’re incorrect by definition. I explained why in my last comment. It’s not front-running to gain an advantage through capital expenditure. Capital expenditure (like colocation) is not fundamentally an advantage other market participants don’t have. It doesn’t matter if you think it is, it’s not. Insider trading is an example of an advantage that is fundamentally unavailable to other market participants. If we accept your argument, we might as well say that spending more money than competitors to build a better performing fund is unfair. > I don't see how you could interpret that as "good" for anyone other than the HFT firms who do it best. It's pretty clearly this: >bad for overall market efficiency No, it’s very clearly not, for anyone who has worked in finance or who is familiar with how market making works. High frequency traders improve net liquidity in the market. Would you prefer the pit days of preferential treatment by manual market makers, or much slower (and thus less accurate) price discovery? I don’t know what else to tell you at this point. You have done exactly the thing I was talking about in my original comment. I’ve explained to you that high frequency trading does not satisfy any legal definition of front-running. You’re free to make up an arbitrary definition of front-running, but it’s not the correct definition. Do you need me to cite research showing why high frequency trading improves liquidity and explaining why it can’t step in front of other retail traders’ orders?