Market Maker/Broker algorithm

[egro1egro]

I am terribly sorry if it seems silly to discuss, but I don't have a clear idea how market makers create their quotes.
I have only experience in Forex. And I think that this topic goes beyond this market.

In some markets broker and market maker is the same company. When broker sees all orders and stop/losses and is allowed to move the price then we can get all sorts of stop-running, spiking etc. But even in a situation when market maker is separated from a broker the broker is still able to run stops. That is an interesting question, how? But it is not what I am interested just now.

Assuming that we are market makers and have to quote the proce and have to perform transactions on the quoted price up to a certain size of an order. If someone decides to buy, we will have to sell. At at this precise moment nobody sells. Where should the price go? In my understanding the price should go down, because if the price is sent up and the same client decides to sell straight away then we will have to suffer loss. But if we send it down then even the same client might decide to sell it making us profitable. Which is quite contrary to supply/demand principle. What exactly do I get wrong?

And in general, is it a simple algorithm that market makers employ? Or is it something really tricky?

Forgive me if it all seems stupid.

 

[TheBramble]

Not stupid at all.

First, bona fide brokers have no material basis for running stops. They make their commissions on the size and numbers of trades they're doing for their clients. Running stops would not fit comfortably within their core business model nor would it benefit them unless they were the other side of their clients' trades themselves. (Somebody is bound to mention it if I don't so: SBs DO take the other side of some of their clients' bets and they most definitely WILL run a stop if there is significant financial advantage in them so doing). SBs have quite a different business model to brokers. The SB model requires maximising their client base and removing as much wealth from them as they can on a one-off basis over a short period of time. The rationale being that there will be another mug punter along in a minute. As opposed to brokers who generally will attempt to cultivate a relationship with their clients in hopes of a long lasting and mutually beneficial commercial relationship. (Thats SBs out the way)

Second, most brokers are unlikely to have the wherewithal to even consider manipulating prices to that degree.

Third, the percentage of profits derived from direct trading activities in brokerages is a very small percent of their overall income, the vast majority being commissions derived.

Professionals will attempt to run stops. They know where the majority of weak hands are likely to be persuaded to part with their positions.

To answer your hypothetical question on Market Makers: They are at all times attempting to keep their position flat. To offset. Unless they have a distinct informational advantage where they will trade off that information to their own commercial advantage, they will attempt to match orders or bias the bid/ask to attract flow to do so.

So in your example (and it wouldn't exactly happen like this in real life as there is aggregation to consider), if they were compelled to sell (and therefore end up net short) they would in theory wish to attract sellers and they could do this by any combination of increasing the Bid and keeping the Ask where it is or increasing the Ask less than the increase on the Bid. Obviously keeping the spread within the mandatory criteria of their MM obligations. So your intuitive take on resultant price action was incorrect. The price would increase in those circumstance. But this is an extremely hypothetical treatment as transactions rarely happen in isolation in liquid stocks and in illiquid stocks, the Bid/Ask is another story altogether.

It's worth considering the Supply/Demand model from 'our' perspective' (Traders) as an infinite pool controlled by the MM. The MM can decide to control just how much 'supply' there is, and 'demand' there is, by appropriate use of the Bid/Ask spread and relationship. Bearing in mind my earlier comment that these guys like to keep flat under normal circumstances, you might wonder then, well, where exactly IS all ther stock if they flat...? Of course, you have it - but they control it. Or rather how much the weak hands want it/don't want it.

But the scenario would rarely develop as you describe. And in practise, they would not wait until they were in this position anyway. Unless there were no get outs they would have seen it coming (in the general case) and covered their anticipated position either directly in the market or with other professionals. They don't often wait to react to events but are to a large extent, ready for them, and oh so very very occasionally ,the actual precipitators of events themselves.

 

[mauzj]

I used to make markets on a few illiquid european equity index futures.

I was paid EUR xx,xxx per month to quote a spread of n points for atleast 80% of market hours. All I did was use a computer to quote a price and hedge off the risk by trading similar products. These hedges were imperfect and it invariably lost money. It was very easy for the losses to exceed the EUR xx,xxx that I was paid. I therefore gave up and just punted when I felt like it. Shorter hours and much easier to make money!

I have a buddy who makes markets on government bonds and STIRS. For the bonds, all he seems to do is quote a price based the Bund +/- a user defined parameter. He'd try to keep his book flat overall, and this invariably lost cash, but he had to do it to maintain his employer's prestge.. For the OTC STIRs, he'd have a spreadsheet to help him judge where the fair price was, quote a price about the fair price, and try to hedge off the position afterwards. This is supposed to be a good way to make money, as the clients are not very financially astute and are not trading for profit..

 

[egro1egro]

Thank you for the response.

Does it mean that it is not always easy for market maker to keep flat at all times?
In what way do they hedge the losses?

What i am interested really is how market makers move the price. You have to move it all the time. You can anticipate future client orders coming in but you cannot guarantee them at any precise moment. You don't have two opposite orders of the same size at exactly the same moment. Right? Then you have to move the price (or just a half of it - bid/ask) and you have to move it by how much?

It looks like Forex is not very civilised place to trade if forex broker is a market maker at the same time. And scalping, therefore, should be considered by them as a sin.

 

[wasp]

I have long intended to write an article for the knowledge lab entitiled 'a day in the life of a forex market maker' in order to shed some light on how it works in interbank circles (which in turn would hopefully give T2W members an understanding of how the retail market makers, who have a basically similar job, work).

One of these days I'll actually get the time to do this, but sadly it ain't today.

Sorry

GJ

Look forward to when you get the time GJ, it will be much appreciated.

 

[egro1egro]

it would be interesting for me too.

It looks to me that Forex is mostly in "running stops/hunting for orders" mode like 80% of time and that is what makes all the patterns on the chart - ranges, consolidations, false breakouts all the time. Between these formations that have nothing to do with supply/demand we have some occasional rapid jumps from level to level follwed by uncertainty again. Many of the chart patterns can be explained in terms of running stops.

I guess that forex brokers/market makers are very keen on "squeezing small hands". The only way to make profit in forex is to cash in on anticipation of such "squeezes" or by scalping when the broker cannot simply see you orders and your intentions. Although, in the last case one needs to avoid having mental stops and exits in common places.
Also, profiting from long-term moves is an option too.

 

[theEdge]

Are there any market makers/brokers who admit to "squeezing small hands"?

I would have thought that most would be more concerned with balancing their positions and earning money from the spread.

If their prices jumped around too much they'd be vulnerable to arbitrageurs.

 

[egro1egro]

I disagree with the last post.
They may never admit it, but what could possibly prevent them from spiking me? Spiking is not prone to arbitrage if it is done properly. By "properly" I mean that no other stop/limit orders are standing in between my entry point and stop/loss and that it is done quickly.
Fear of arbitrage would prevent broker/MM from moving price in strange ways only if there is a constant flow of "at market price" orders. To be more precise, moving price up and down withing half-a-second doesn't give anybody a chance to issue and execute "at market" order while the price is on top or at bottom. But if there are many "at market" orders flowing in/out constantly then the price cannot be moved to and back quickly without substantial loss for the MM and therefore arbitrage opportunity could be exploited if the price is moving slowly in a wrong direction.
Another way to squeeze little hands would be to move price slighlty up and down slowly, constantly violating new highs/lows and sometime not reaching them. If the amount of resting orders is significant we can withstand a couple of occasional arbitraguers for the sake of running a big bunch of stop-orders. If the market suddenly wakes up and starts sending more "at market" orders the price can quickly be returned to make broker position flat. In the same way as it does in usuall circumstances.
Looking at FX charts of popular pairs (EUR/USD, USD/JPY) I can conclude that flow of "at market" orders is significant only sometime (about 10-20% or even less). Rest of the time, broker/MM can create wonderful fractals of price movements with a purpose of squeezing little hands. Even when the price moves as a trend it can be caused by broker/MM seeing that there are more pending orders to fill in a certain direction rather than just all over around the current price. But this situation is in agreement with supply/demand principle. Whereas, choppy market that happens most of the time is not.