Did anyone else notice a rather sudden and dramatic change (relative to the prior 2 months or so) in the behavior of their algorithmic trades starting toward the end of trading on Tuesday and continuing at least through the end of trading Wednesday? I'm not actually doing trading but just developing a proprietary algorithm that seemed to all of a sudden go whacky. Maybe I've got a bug in the program but I don't think so.
Certainly the market dropped, but it has done so before without the following-day persistence of anomalous behavior in my backtesting. I can't say what asset class this is associated with since I'm taking broad statistics on all 9k symbols provided by IEX in its daily downloads. I'll look a bit deeper into what is going on to see if I have a bug or it is isolated to an asset class, etc. I just thought someone else might have noticed a systemic change in the market that persisted for at least the day after the drop.
Definitely saw no systemic change in ES or NQ on Tues or Wed they traded as much within expectation of readings that can possibly be expected.
It's called the most important election in US history, plus COVID wave fears, PLUS no Fed safety blanket. Did I mention earnings?
"range of hft" sounds like a market indicator/metric of some sort. But perhaps you mean "the range of hft strategies got bigger" and the market is adapting to that increase. Yes. I thought of that but it struck me as odd that there would be a sudden change in the mix of strategies after this market drop that didn't show up in the earlier market drops. "Overnight" may be onto something with the mounting chaos-inducing factors approaching the election.