Can anyone explain what this guy was doing?

Monitoring other exchanges for large orders then using his latency to "front run" the orders to the exchange in which he had the latency advantage.
 
He displayed large sell orders to drive the price down, so he could buy. then he lifted the sell orders hoping that the price would rise without that selling pressure. He would then exit for a tick profit.

And I seriously doubt he fooled any HFTs. More likely it fooled dumber/slower algorithms that monitored the order book.
 
Mind explaining how he did it then?
He devised an algorithm that would place huge Emini futures orders 3-4 ticks on either side of the NBBO or market price. As the orders got closer to the market (1-2 ticks), he would pull them (cancel)....and then add new orders back at the 3-4 tick level. He would do this until the market started to move in the direction he wanted, then he would pull the "fake" orders and place real orders in the opposite direction using a stop order a tick or two above the market if buying or a opposite for the sell. If those orders got too far away from the market, he would cancel them and place orders closer to the market.
With his fake orders, he would show huge size (10k+ contracts). On his "real" orders, he would be trading 100-200 contracts. Since he had a seat on the CME, his commission costs were really low.
 
....and then add new orders back at the 3-4 tick level. He would do this until the market started to move in the direction he wanted, then he would pull the "fake" orders and place real orders in the opposite direction using a stop order a tick or two above the market if buying or a opposite for the sell.

So the final direction of his executed orders is basically predetermined or decided on-the-fly?
 
Back
Top