Another Day , Another Fat Finger Trade
- Thread starter stock777
- Start date
We're way ahead of you:
http://www.elitetrader.com/vb/showthread.php?threadid=19929
http://www.elitetrader.com/vb/showthread.php?threadid=19931
nitro
http://www.elitetrader.com/vb/showthread.php?threadid=19929
http://www.elitetrader.com/vb/showthread.php?threadid=19931
nitro
Wouldn't this be dangerous because you would end up short if they bust the trade? Much better strategy would be to put a position on in a different exchange, eg DAX, FTSE, or even short bonds if they spike in sympathy.
I wouldn't be surprised if this is what is going on right now. The guys doing the "erroneous" trades have their limit orders already lined up in a whole host of markets. If the trade gets busted, they make a killing from the sympathy move followed by snapback in related markets. If they don't bust the trade, they probably still make enough money to offset the loss in the manipulated market.
An idea might be to have an auto-trader program that leaves limit orders continuously in the eurex products x% below the last price, moved every 30 seconds.
I wouldn't be surprised if this is what is going on right now. The guys doing the "erroneous" trades have their limit orders already lined up in a whole host of markets. If the trade gets busted, they make a killing from the sympathy move followed by snapback in related markets. If they don't bust the trade, they probably still make enough money to offset the loss in the manipulated market.
An idea might be to have an auto-trader program that leaves limit orders continuously in the eurex products x% below the last price, moved every 30 seconds.
The reason for these spikes in the electronic indexes has, IMHO, to do with a flaw in the way the e-contracts are traded that is going to have to be addressed in some manner.
There has to be some type of circuit breaker that controls how far out of wack the futures can get from the cash (premium) in the very short run. It takes time to arbitrage things so a stop of one minute when the futures and cash diverge to an extreme would probably allow enough time to arb and to prevent the quick running of stops that contribute to the avalanche. This would be preferable to busting trades after the fact.
As I mentioned before, the pit has a natural delay built in to slow down these spikes - the floor traders just put down their arms and stop trading for a minute until they figure out whats going on. The electronic needs to develop something that mimics this unintended shock absorber.
There has to be some type of circuit breaker that controls how far out of wack the futures can get from the cash (premium) in the very short run. It takes time to arbitrage things so a stop of one minute when the futures and cash diverge to an extreme would probably allow enough time to arb and to prevent the quick running of stops that contribute to the avalanche. This would be preferable to busting trades after the fact.
As I mentioned before, the pit has a natural delay built in to slow down these spikes - the floor traders just put down their arms and stop trading for a minute until they figure out whats going on. The electronic needs to develop something that mimics this unintended shock absorber.
1.) If there is a spike in say the ES, fade the spike but don't trade the source of the spike (in this case ES). Instead trade YM or NQ etc. Otherwise you could end up exposed with a busted trade.
2.) Is it news or is it a mistake/spike? If its a news related move, it would probably happen in all indexes equally to a degree. If its a "fat fingered" spike, the move would happen in one index and then spread to the others. So by monitoring divergences between the indexes, you should be able to determine if the move was news related or a mistake.
3.) Don't leave distant stops in the market. This is a tough one, but if people are allowed to game the indexes with these "mistakes" - you may be better off not having stops in which provide the fuel for these wildfires.
2.) Is it news or is it a mistake/spike? If its a news related move, it would probably happen in all indexes equally to a degree. If its a "fat fingered" spike, the move would happen in one index and then spread to the others. So by monitoring divergences between the indexes, you should be able to determine if the move was news related or a mistake.
3.) Don't leave distant stops in the market. This is a tough one, but if people are allowed to game the indexes with these "mistakes" - you may be better off not having stops in which provide the fuel for these wildfires.