Why chasing latency arbitrage from a home connection is a losing contest

[Shadow Trader]

The pitch sounds seductive: price differences between brokers or feeds, a few milliseconds apart, and a trader who gets there first captures the gap. It's an old idea, and professional firms have spent fortunes on it.

The race is measured in microseconds. They sit in the same data centers as the liquidity providers, use specialized hardware, and send orders over dedicated lines. A retail trader on a standard connection is a very long way behind.

Even if an apparent gap appears on two platforms, it's usually a trap. Brokers protect themselves against this exact behavior, with last look, delayed execution, and rejection of orders that exploit stale prices. Accounts that show a pattern can be restricted or closed, and profits cancelled.

What remains is a distorted picture on a retail chart: a feed that updates less often than the true market, creating the illusion of an edge that disappears when an order is sent.

Nothing says the idea can't be tested. Doing it with small size and honest logging shows the reality quickly.

What did a small test of cross-feed price differences actually reveal about fills?