The War Of The Bots

Many retail traders learned about the evil market making bots through mainstream media. Yet, they are not the ones that rule the game. Their nemesis, institution bots, are the ones even market making firms are afraid of.

Institution Level Trading Is Different From Retail Trading Completely

Institutions participating in the stock markets cannot just go out and buy (or sell) a particular stock at market. Especially when the intention is to accumulate or unload a stock with market moving size, they have to do that in stealth mode. If these institutions simply put a market order out there, they will move the particular issue wildly. The end result is very bad average price for the order. It may even halt the issue due to the extreme volatility induced. No one wants to be put under the microscope of the authorities when it is totally unnecessary.

Many boutique firms are specialized in handling this type of large size orders for the institutions. Their traders (and their bots) are highly specialized in breaking the large orders into small chucks so that no one can tell if such pending supply (or demand) is there. All kinds of strategies are employed to hide the true intention. For example, if an institution needs to unload a large block of shares, the traders handling the order do not just unload the issue all day long. They would at times buy the shares back to nudge the market higher so that they can unload at a better price.

Secrecy Has Always Been A Top Priority

Think about those who some how gain the knowledge of a particular institution’s intention and the particulars of its trading instructions. Due to the sheer size of the pending order, it will imbalance the short term dynamics of the particular stock, and sometimes it would affect the related indices too if its weighting is significant (like Apple). These players who learned the intention can lean on the imbalance to gain absolute advantage for (almost) sure win.

The leak of the intention and the actions taken by the other players as noted above will hurt the institution as they are the one that will be at the receiving end of worse fills on their order. Hence, institutions like to guard their orders and intentions as secretive as possible.

Within the past 10 years, the introduction of specialized client side bots built for these institutions partially solved the problem. For average large size orders, they can be handled without human intervention, thus reduces the chance having the intention leaked to unnecessary parties.

Here Comes The Catch

The use of specialized bots by the institutions induced a technology war with the market making firms. When the institutions having orders entering the markets, the market making bots are the ones that has to absorb the liquidity imbalance. No one wants to lose money simply because some institution decided one day to throw large size orders around. Thus, a race between these 2 classes of bots had started.

Market making bots are now smarter and faster then ever in identifying potential intention of orders coming from a particular brokerage. The goal is to at least not being burn by the institution. Of course, early detection of the intention would not only save the market making firms a lot of money, they can also design their bots to take advantage of the situation to profit from it.

For example, if an institution successfully disguised its intention and unloaded so many shares to the market, the market making bots would have been lured to keep buying as they failed to recognize the unusual situation. When these market making bots are loaded up with too many shares, they have no choice but to dump whatever they have bought. In another words, the institution successfully transferred the risk to the market makers.

In the same situation, if the market making bots figured out the intention of the institution early and not taking the bait to accumulate the shares, the institution would have no choice but to dump at worse price level, unless, they can wait for another day to execute the plan.

 Members only content, login now or sign up to view the rest of this article.

Share