For those who automate, I'm just curious what risk metric you consider as sufficient to get around catastrophes like Flash Crash. Do you have a pulse on sudden volume spike to prevent you from entering? How would you manage to get out if you were already in a trade (although there's not much you can really do at this point)?
Algo traders: What risk management are you using to manage crisis like Flash Crash
- Thread starter schizo
- Start date
Hello schizo,
Very good question. I think about it with my algos.
My answer so far. Nothing. If my stop gets skipped, I just lose money. It is what it is.
I would love to hear other answers.
Trade option spreads (2+ legs), then your risk is predefined/predetermined, ie. you know it in advance... for example 5%.
More info here, incl. P/L diagrams.
Of course a panic-closing by the auto-trading-program can defeat that guarantee...
More info here, incl. P/L diagrams.
Of course a panic-closing by the auto-trading-program can defeat that guarantee...
Last edited:
Well, I dunno if you witnessed the Flash Crash in person, as I did, there's NO WAY you would be able to turn on the "kill switch". The S&P tumbled over 100 points (can't remember the exact number, but huge amount for that time) in a matter of minutes--viz. blink of an eye.
If you refer to the 2010 Flash Crash, while my memory is kind of hazy from that day, I suspect lots of folks tried dumping at a percentage loss (or had breakers in-place), but couldn't get a fill in that fast moving market. Market sells got filled who-knows-where on the way down. Speaking for myself, I rarely trade on the "first move" on anything downwards b/c I've learned first moves are usually wrong and I don't like getting whipsawed. Look at a chart of 5/6/2010 and how it reversed going into the close ... how angry would you feel knowing you'd ended up selling at the bottom?
That event caught everyone off-guard, as memory serves...and Ben's audio from the pit is a trip down memory lane. IMO bottom line, sometimes the best thing to do is hold pat and ride it out until you know what's happening ... like turbulence in a plane, yes it's scary, but the only way out is through, and I didn't do a darn thing that day in any of my trading or investing accounts. And if you lose some $$ in the process, that's part of the game, as the other poster said. (and a tax harvesting tool!)
That event caught everyone off-guard, as memory serves...and Ben's audio from the pit is a trip down memory lane. IMO bottom line, sometimes the best thing to do is hold pat and ride it out until you know what's happening ... like turbulence in a plane, yes it's scary, but the only way out is through, and I didn't do a darn thing that day in any of my trading or investing accounts. And if you lose some $$ in the process, that's part of the game, as the other poster said. (and a tax harvesting tool!)
Last edited:
While I can't call myself algo trader: Size properly. If you don't hedge, you need smaller size to eat the tails.
It's of course nice if you do manage to get out in time, but you can't count on it.
It's of course nice if you do manage to get out in time, but you can't count on it.
Market close all positions turn off the computer.
Guys, this is about automated trading, most likely carried out on a remote VPS somewhere, which means you will probably not be actively following the market. You will only hear about it after the fact. Hence, manually getting out of the trade is out of the question.
So the question then boils down to what safe mechanism are you utilizing to mitigate any huge losses from unavoidable systemic risk like Flash Crash? It's also worth remembering that next Flash Crash might not be so generous as the previous one, where it recovered most of its loss by the end of the day.
So the question then boils down to what safe mechanism are you utilizing to mitigate any huge losses from unavoidable systemic risk like Flash Crash? It's also worth remembering that next Flash Crash might not be so generous as the previous one, where it recovered most of its loss by the end of the day.
Algo traders should know their stats on drawdown and should probably stop trading if hit 25th percentile drawdown (or 75th depending on how you're interpreting it).