Meldown - maximum % changes in S&P 500

A

abattia

I want to guestimate a potential worst case max loss on a position.

If my autotrading system issues a stop order while the market is making an extreme move, and if it takes 10 (or perhaps 100) times longer for the order to be filled (because all the rest of you happen to be trying to do the same thing at the same time!), then what is my potential worst case max loss?

Do any of you know (or can you point me towards a [free ]source) what the largest ever historical % change in S&P500 has been over a period of 10 seconds, and also over a period of 1 or 2 minutes?

Many thanks, and happy trading ...
 
Quote from abattia:

I want to guestimate a potential worst case max loss on a position.

If my autotrading system issues a stop order while the market is making an extreme move, and if it takes 10 (or perhaps 100) times longer for the order to be filled (because all the rest of you happen to be trying to do the same thing at the same time!), then what is my potential worst case max loss?

Do any of you know (or can you point me towards a [free ]source) what the largest ever historical % change in S&P500 has been over a period of 10 seconds, and also over a period of 1 or 2 minutes?

Many thanks, and happy trading ...
More...
your max loss is impossible to calculate under that senario.
 
Quote from abattia:

Thanks vhehn

Can you explain?
More...
it is not possible to predict where you might be filled since it is not possible to predict what the next black swan event might be. it will always be a guess. look at what happened after 9/11.
 
Thanks wave - So 9/11 resulted in a c. 4% drop in S&P futures “in seconds” on Globex. ..

Thanks vhehn - I understand what you are saying (and BTW I love your LLOYD's prayer ... genius!).

I am not trying to predict when the next Black Swan event might be. I am just speculating that if the next one is as bad as whatever caused the biggest historical move over 10 seconds in the past (or over 1 or 2 minutes), and if I had to wait for that whole 10 second (or 1 or 2 minute) move to finish before I was filled, then what % change would have happened in that time?

It is just an estimate, but it will give me a rough idea of what happened in the past ... won't it?

For example, if S&P futures tanked 4% “in seconds” at 9/11 (and if this was the biggest ever change in a short time), then I think some useful guestimations can be made. If say you are intraday trading an ETF such as SPY (at 4x intraday leverage), you might have to sit through a 4 x 4% “crash” before you could exit, and you might also have to sit through a limit down/up (I hadn’t thought of that possibility!), which would have resulted in this loss doubling again over the next 6 days (so a total 32% loss on an ETF). If, on the other hand, you were trading SSO (ProShares Ultra S&P, 200% of intraday change in S&P), you might have to sit through 4 x 2 x 4% “crash” (32% “in seconds”, and then 64% if you got caught by the limit down). Obviously, the situation would be even worse with futures (see the references provided by “waves” above). Sobering ....

Have I got this correct?

Sure the future might be worse than the past... (Marvin would say, "it's certain to be worse ...")
 
1. Past performance is not an indicator of future returns.

2. Maximum possible loss on long position is 100%

3. Maximum possible loss on short position is unlimited.
 
I think 5% is a pretty reasonable max during liquid market hours for any event that doesn't in and of itself prevent trade execution. That risk goes up massively if you're near limit down. I wouldn't hold a position either direction if that were the case since I think they halt trading entirely rather than allowing it to just sit at bid limit.

Obviously if a nuke goes off in the wrong spot, there is no GLOBEX to execute your stop (if you happen to be long) and/or no clearing house to guarantee the other side of the trade (if you happen to be lucky enough to be short).

I'm OK with going bankrupt or not getting paid if something that shitty happens.
 
You must log in or register to reply here.
Back
Top