Trader's Handbook: orders, prices, stack, funds, currency - page 2

[hrenfx]

Types of traders *

Traders can be classified indefinitely. But among professional market participants (brokers, platform developers, etc.) traders are divided into two types: GUI-clickers and algo-traders. The names fully tell about their peculiarities. As a rule, there is no division into experienced and beginners (also used as "nubs").
If the task is to attract traders, clickers are the easiest to attract and algotraders are much more difficult.
Clickers are given a nice beautiful GUI, trading conditions are given secondary attention. For this reason, in particular, clickers are on average meat (a source of profit for other market participants). Competition for clickers is high.
Algotraders are very picky about trading conditions, GUI is not of determining importance. Competition for algotraders is low, although serious efforts are spent on it.
Mat. expectation of all algotraders is much higher than the mat. expectation of clickers. The number of algotraders is much smaller, but the turnover is higher.
It so happens that the opinion of algotraders is listened to much less than that of clickers. This is what platform developers suffer from, and brokers also suffer from this. The opinion of algotraders can even apply to the way, for example, the platform is sold to a broker - for a commission on turnover or a subscription fee. I.e. all nuances of possible changes in trading conditions are taken into account.
Most often market participants are set up to satisfy only one type of trader.

It is obvious from what was written earlier that market makers are algo-traders.

Toxic flow

This term is used by market makers. Toxic flow is a systemic profitable flow of trade orders. It comes from the very ones who make money on the imperfection of MM-algorithms. In practice, 99% of the toxic flow is accounted for by algo-traders.

[hrenfx]

Broker operating models *.

Следует помнить, что брокеры на рынке Форекс не стараются вас «задавить». Они желают только сохранить свой бизнес и свои деньги. Иногда крупные брокеры помогают разместить ордер и начать торговлю, но в дальнейшем каждый сам за себя.

1. Marketmaking
2. STP
3. ECN/STP

All three models are absolutely market - this is very important to understand, as there is no fair market or unfair market. Everything is unified and intertwined.
The first one differs from the other two in the fact that it earns not on the commission from the turnover of clients, but on their drain. Let's analyse everything on one real example (all data are public), we will leave out names.

A certain broker claims that its turnover is > $100bn/month, while earning $80 per mio. Obviously, this is a market maker, as even the most inadequate STPs do not earn more than $65 per mio, let alone such a high monthly turnover.
It turns out that $80 per mio is the rate of draining their clients. This means that at an average lot of 0.5 the average client makes a trade with a profit of ~ -0.5 pips EURUSD.
If you twist it, you can roughly calculate the average lot/pip trade from MT4 tickets. Yes, many tickets will be spent on cancelled orders and I/O transactions, but this is nothing compared to the number of market orders - the favourite of clickers.

Let's reason further, if the monthly turnover is > 100 yards, it means MoneyIn - MoneyOut > $8 mio. I.e. the broker makes $8 mio monthly by draining his clients.
It is asked where it is possible to find such a number of hamsters every month. And don't forget that MoneyOut is on average about two times less (I invented it myself) than MoneyIn. That is, the broker attracts about $16 mio of new live money every month.

How does he do that?!
The average hamster carries somewhere around $200 (also made it up myself, never heard from anyone). So even by the most conservative estimate (MoneyOut = 0) you need to rob 40,000 hamsters every month.
In fact, hamsters are called hamsters because they drain many many many times. And they carry their average $200 on schedule - with every paycheck, for years. It's payment for the regular pleasure of feeling like a clicker. Just like spending monthly on cinema, beer, etc.

If a toxic flow suddenly occurs among the hamsters, it is quickly and easily identified, after which it is strongly recommended to switch to STP-type, or close the account.
Such a broker actually benefits from other types of brokers (STP and ECN/STP). These brokers are not competitors to the market maker. They even do a huge favour: they make an explicit separation between "ECN for pros" and "Standard for beginners". Because it allows the client base on Standard to get rid of toxicity. It is even favourable for them that people go to others, as there is less toxicity in the end. And it is possible to earn more from less risky MM.

STP AND ECN/STP can have serious problems with a high proportion of toxicity. They need clickers at least to dilute the toxic in front of their LPs, which are almost always vaulted on large market makers. And they act in the image of small market makers (as I described above).
The fact is that on FOREX at this stage prevail MM-algorithms of large owners - banks. If you make money on FOREX, it is almost always a loss for these banks, or rather for their algorithmic departments, which spend millions of dollars of their budgets to prevent you from making money.
To profit from the drain of another simple trader is now only indirectly possible: the trader will drain the MM-algorithm, and you will snatch this profit from him for yourself. Roughly speaking, to earn on FOREX is to rob the stolen profits.

But all this sounds strange only if you look through rose-coloured glasses at many events taking place in the world. For a long time the market (stock market or FOREX) has become a legal tool for taking money from the population. But not everything is so bad.
Let's return to our MM-broker. Does he have a graphical scheme? Not exactly. The thing is that MM-broker is able to earn even much more than $80 per mio in a very volatile and trendy market. But also to go into a decent minus on a long and withering market. These are quite predictable peculiarities of trading of an average hamster. I.e. MM model of brokerage is a risky model, but at the same time it can be very profitable under favourable market conditions.

For a MM broker to give up the MM model, he needs to believe that the STP model is able to give not much less on commissions. This means that he needs to increase turnovers several times. This is almost unrealistic to do on a huge base of clickers, because the turnover is made by algotraders. And algotraders will go only when trading conditions will be one of the best. And this, in turn, can happen with serious financial injections into STP-infrastructure.
Moreover, if there is such a huge herd of meat in the form of clickers, why should we give their drain to greedy MM-banks, receiving only a commission on their drain. If you can take everything completely for yourself. In general, such simple reasoning allows us to understand that a MM-broker with a huge meat-base becomes a hostage of itself.
It can get ridiculous here, if the base has some national mentality. For example, the number of Russian hamsters decreases by an order of magnitude if you ask them to fill in a little more fields in the questionnaire, as required by the regulators. This largely forces MM brokers to go offshore.

But then again, it's not all bad....

P.S. All information from the public - analysis of bits and pieces from forums and thematic resources.

[---]

Classification of brokers of the FOREX market *

Следует помнить, что брокеры на рынке Форекс не стараются вас «задавить». Они желают только сохранить свой бизнес и свои деньги. Иногда крупные брокеры помогают разместить ордер и начать торговлю, но в дальнейшем каждый сам за себя.

Dealing Desk broker, in the abbreviated version DD-broker, is a company that works on the currency market by placing its own orders. Such a broker has a fixed spread. The distinctive feature of DD-broker 's money making is the difference (spread) and currency trading operations against their clients. Another name for this type of activity is market maker. As it is clear from the name, a market maker is both a seller and a buyer in one person. This gives an advantage to execute a deal and place a counter trade order.

In Russia this type of activity is called a dealing centre, which is not quite correct. Traders working with this broker cannot see the real market price. This is used by so-called market makers (Forex market maker), conducting cunning manipulations with currency on the market. As you know, the spread is the difference between the ask and the bid, and this allows market makers to make a profit even in the case of losing trades. This activity involves the opposite side of the trading relationship.

The trick to getting a Dealing Desk broker 's money is one simple thing. He is essentially a counterparty and tracks all the actions of his clients. In doing so, it divides all clients into two large groups. On those who trade successfully and others who cannot afford to make successful trades and as a result lead the broker to profit.

But DD-broker is a counterparty for that, that he can earn on both. Successful DD-broker clients are more often tracked and are subject to requoting (which means delaying quotes until your order is confirmed). In case of high market volatility, the trading orders of successful clients are used, and then the orders of those who open less successful trades are used. It should not be forgotten that the dealer will not leave himself in losses and will limit his losses, which cannot be said about clients.

Another not less important indicator is the integrity of this broker. It is determined by the NDD (No Dealing Desk) regulations of the company. This company offers access for market makers without processing orders. The absence of requoting is also a plus. The advantage of this method is that trading operations take place without restrictions.

NDD broker gets profit due to commissions and increased spread. In case of the latter, the client is charged a commission. These types of brokers are divided into two subspecies. The first are STP brokers, the second are ECN+STP.

STP stands for Straight Through Processing. Such broker sends orders directly to the liquidity provider from the client. The advantage is that the supplier can consist of several persons, which has a positive effect on the clients' profit. Another plus is that when the number of banks increases, the liquidity grows, positively affecting the profit of clients.

A big advantage of this broker is that traders can get access to real quotes of the international financial market. At the same time, the dealer himself does not interfere in the action of orders.

Another type of broker - STP, is characterised by the possibility to choose floating or fixed spreads. This broker receives quotes and spreads by organising trading operations through the markets. It receives them as an intermediary. Most of the banks act as market makers, offering fixed spreads. But STP broker has another option - to choose between leaving the spread at the same level or zeroing it and choosing the best prices for currency sales and purchases. The second case is the floating spread.

STP-broker does not trade against its clients, and they get profit at the expense of the spread premium. Another technique that STP-broker uses to make profit is adding pips to the price for buying, or subtracting pips from the price for selling. The result allows STP-broker to make profit.

There is ECN system, which means Electronic Communications Network. The essence of ECN is to carry out direct trading operations between the participants of the system. This broker provides private traders, market makers and banks with a platform on which they carry out their transactions. The advantage of this method is the execution of a trade order at the best price. Operations on currency trading are in real time and a small commission is taken for it.

There are situations when ECN-brokers call themselves STP-brokers and vice versa. To determine the effectiveness of the broker, you can conduct a small test. It will show the full picture of what is going on, namely you will see the whole "depth of the market" in the market "glass". This is, first of all, full information about active orders of the system participants and their orders, which determines the market liquidity.

Do not forget that ECN-brokers always offer floating spreads. They have no profit on spreads and they are the only ones who set the commission.

NDD broker always offers such advantages as trading on the real market, which is completely open ("clean") for the participants of the system and high speed of order execution.

<br/ translate="no">Summarise.
ECN-brokers on the currency market earn, in fact, not so much money, if they are compared with dealing centres. The profit is formed from the commission of the system users, so they benefit from successful transactions of the participants.

STP-brokers earn income from the spread. Setting the price is their business. Using markups, they get the prices they need. This type of trading allows you to get a number of advantages, which allows you to quickly execute a trade transaction, reduce the size of the deposit and get anonymity of trading operations. It is also considered a plus that they care about the success of their clients' trades.

Market makers - "make money" on the losses of their clients. In case of successful trading such dealing centres may close access to the platform where you conduct successful trades.

Also read the CRFIN standardisation *

excerpt from the document:
3.1. Forex brokers of group A are legal entities providing transactions with OTC financial instruments of the Forex market to individuals and legal entities, but not limited to it, and having equity capital sufficient to cover market risks of non-performance of obligations to clients.
3.1..2. A forex broker of Group A has the right:
3.2.1. to open sub-accounts for clients and accumulate clients' funds accepted as collateral for transactions;
3.2.2. to independently decide on the necessity of overlapping client positions on external counterparties;
3.2.3. to act as an external counterparty for forex brokers of Group B.

3..3. Group B forex brokers are legal entities providing transactions with over-the-counter financial instruments of the forex market to individuals and legal entities, but not limited to it, and do not have own capital sufficient to cover market risks of possible default of obligations to clients.
3 3.4. Forex brokers of group B:
3.4.1. have the right to open sub-accounts for clients and accept clients' funds accepted as collateral for transactions;
3.4.2. are obliged to overlap the aggregate client position with external counterparties;
3.4.3. are obliged to place clients' funds accepted as collateral for transactions on their sub-account with an external counterparty to ensure dynamic overlapping of their clients' open positions.

3..5 Group C forex brokers (introducing brokers) - legal entities and/or individual entrepreneurs attracting clients for forex brokers in order to receive remuneration from them.

[hrenfx]

Quants.
Quants are, as a rule, educated people with an analytical mind. Unfortunately, in life it happens that good education, mastery of exact sciences, etc. is not enough for success. Usually, the more effort spent on studying something, the more importance is attached to it. There are a huge number of mediocre individuals among scientists who do quite well in the career features of their profession. These include publishing in journals, attending summits, giving lectures, etc. One of the near-scientific topics is the comprehensive study of near-trade. Everything is done the way it is taught. I.e. mat. apparatus is applied head-on, associations with physics are made, various market models are created, etc. Such scientists are called quants. A huge part of them are careerists. But there are exceptions, as in any business.

Co-integration.
It is possible to introduce various quantum heresies, giving the term cointegration clear mathematical definitions. But all this has little relation to trading practice. It is possible to explain cointegration in a simple way. Imagine that you have a lot of trading symbols. If there is a possibility (quantity of each symbol) to buy/sell these symbols in such a way that the profit fluctuates near some value that does not change much, then such symbols are cointegrated. Of course, it is quite childish, but it conveys 90% of the essence. Such a combination of symbols is called a cointegrated portfolio. A classic example of such a portfolio: an asset and its futures.

Arbitrage.
Any portfolio has Bid and Ask prices. Moreover, it has its Level2 (easily calculated from the Level2 of symbols included in it). If there is a possibility to make a trade on a cointegrated portfolio so that there is a positive profit after its closing, this possibility is called arbitrage. Such a cointegrated portfolio itself is called an arbitrage portfolio (I just invented it).
The lower the trading costs (execution quality, commissions), the more chances a cointegrated portfolio has to become an arbitrage portfolio.

[hrenfx]

Markup

Markup is a deterioration of the price (Level2) by a certain amount. For example, a broker can worsen prices to make money on it. It can take a marketing step - zeroing the commission (clickers do not like difficulties), but adding it to the markup. There are also positive (profitable) for the trader markups, but their understanding is possible only with a good understanding of pricing, so we will leave them out for now.

Trading arbitrage

Let's assume that there is a desire to engage in arbitrage. To do this, you need to create a cointegrated portfolio at least. The simplest cointegrated portfolio consists of two symbols of the same name: one for one broker and the second for another.
Let's take, for example, the so popular EURUSD and give the symbols appropriate names for convenience: EURUSD1 and EURUSD2. The most important thing to fully realise is that EURUSD1 and EURUSD2 are completely different symbols. They could be called differently by brokers, have very (by an order of magnitude, for example) different prices and other differences. Only one thing is important - they are cointegrated. But for simplicity we will consider an elementary case: EURUSD1 and EURUSD2.
Before comparing prices, we make an algorithmic markup on them in order to include all possible trading costs (execution quality for each broker and commissions for each broker). We will further assume that all prices are already marqued.
So, you have trading accounts with certain money in each broker. If we look at arbitrage very primitively, we need to find the moments Ask1 < Bid2 and Ask2 < Bid1. And at these moments open/close opposite positions in each of the brokers.
This is the simplest and frontal implementation. Let's make a small digression to a more generalised and universal vision of such trading.
In this case, the portfolio cointegration shows that Synth = EURUSD1 / EURSD2 fluctuates near one. This Synth has its own Synth_Bid and Synth_Ask (Synth_Level2) prices. If it is possible to construct a ZigZag with tops on Synth_Bid and bottoms on Synth_Ask, then our Synth portfolio is arbitrage. But this is a diversion.
Let's return to the more usual for most people view of trading. In fact, in some cases it is justified to create something high level for trading convenience. And for arbitrage this high-level is done in the following way:
We take Level2_1 and Level2_2 and simply combine them into Level2_All, to which the created artificial high-level symbol EURUSD_All starts to correspond. Very simple trading functions are written that are able to trade EURUSD_All. For example, if you want to sell EURUSD_ALL, then OrderSend(EURUSD_All, OP_SELL) sends a SELL order to the broker whose Bid-price is the highest, i.e. its Bid-price is on the best gang in Level2_All.
Now we should say a few words about Level2_All. In its internal representation the band now contains not only prices and volume, but also the name of the source of this data.
With this implementation you just need to wait for the situation when Ask_All < Bid_All and at this moment simultaneously open multidirectional positions on EURUSD_All. As a result, you get a high-level profit and no open positions on EURUSD_All. It is convenient, isn't it? An Expert Advisor in such a high-level language would take 10 lines: saw negative spread, traded it, wait further.
If we go down from the high-level vision of such trading, we will notice that at the moment when we have no positions on EURUSD_All, we will have an open position on EURUSD1 and its opposite on EURUSD2. This in turn will cause Equity1 and Equity2 to be naturally skewed. Yes, roughly speaking, Equity_All = Equity1 + Equity2 will grow as we trade, but we know that Equity1 and Equity2 must be at least positive. And our distortions may well make the account on one of the brokers simply zero, though the other one will grow.
How to solve this problem? The first thing that comes to mind is to transfer money from one broker to another. That is, order withdrawals from the broker where there is more money and send them to the one where there is less. But it is long, very long. Besides, it is also extremely expensive - bank and other types of transfers are not free.
But imagine that there appears someone who says that all the distortions will be levelled for a small fee. That someone is clearing.

A simple aggregator

It wasn't hard to see how convenient high-level trading is. It is convenient to trade not only arbitrage, but any strategy at all, because EURUSD_All prices cannot be worse than EURUSD1 and EURUSD2 based on its construction. I.e. the artificial symbol is more favourable and the profit on it is obviously higher. Here comes the idea of creating similar artificial high-level symbols for any type of strategies. At the same time, we can use the same principle to take not two sources (brokers in the example) of cointegrated symbols, but any number of them. Obviously, the more sources, the more favourable (prices are better) high-level symbols are.
Such high-level formations are called liquidity aggregators. Obviously, it is an algorithm - some kind of software that allows trading in this way. Among the companies participating in the life of the market, there are software companies that deal with just such aggregation algorithms. Providing their clients with an opportunity not to bother with low-level trading, but to trade at a high level, spending their intellectual forces on other things.
Currenex and Integral are the most famous among these software companies. They sell their software for a commission on turnover. For this reason in particular, some write their own aggregators to be more competitive.

Creating a trading platform

You want to create a trading platform without much effort. Very roughly it looks like this. You come to a well-established software aggregator company and declare your desire. They reply that no problem, but you need to fulfil some conditions. First, you must have agreements with the companies that are the sources of the prices you want to aggregate. Then you need one single account with a prime broker, which already has agreements with your source companies and is able to organise clearing so that there are no distortions. Clearing is also part of the commission.

Next, the aggregator software company sets up its aggregator software with your sources and prime broker, with which it has agreements for the possibility of such activity, and gives you a turnkey trading platform.
Note that sources can be the same aggregators, and they can also have some aggregators as sources. All of this can end up intertwined and repetitive, which causes sometimes inflated (duplicate sources) volumes on some of your Level2_All gangs. And, of course, at the very ends of this network are banks with their MM algorithms.

[hrenfx]

Indicators of a single MM algorithm. *

As it was already mentioned, the owners of MM-algorithms are not necessarily banks. A prime example of such an owner is Lucid Markets. One of the largest market makers on FOREX. Unlike the corresponding algorithmic departments of banks with multi-million dollar budgets, the brain of Lucid Markets looks more than modest: < 10 mathematicians. What can't be said about their indicators: daily turnover > $50bn, average profit ~ $11s mln - roughly speaking, EURUSD expectation ~ 1.5 pips (5 digits).

It is indicative that after the mandatory publication of Lucid Markets trading activity data during the purchase of the controlling stake, the heads of algorithmic departments of the largest MM-banks were summoned "on the carpet".

We will certainly consider the simplest HFT MM algorithm and the necessary conditions when it is profitable. But before that we will have to cover in detail the peculiarities of STP and ECN/STP trading platforms.

[---]

The cisco website has a number of informative documents (may be of interest to some)

Trading Floor Architecture

Competitive environment of global exchanges functioning

Risk management system

Network Architecture for Financial Market Data Processing

[hrenfx]

PriceTaker and PriceGiver.
Market participants are always divided into two categories: those who create (PriceGiver) market offers (prices and quantities) and those who accept (PriceTaker) them.
Market offers themselves are bids: ready to buy/sell this much at this price or better. The aggregate of these bids (Level2) conventionally shows the current market liquidity.

Liquidity Provider.
Liquidity Provider (LP) - this is what the aggregator calls its price sources, which it combines into one single virtual (high-level) Level2. Obviously, an LP is a PriceGiver. A significant proportion of PriceGivers are MM algorithms.

STP.
STP is an aggregator where the client is only allowed to be a PriceTaker. This means that clients are not able to form their own bids that anyone else would be able to bid on. They can only accept bids from PriceGivers.
As stated earlier, the aggregator is a kind of high-level virtual trading service. Therefore, all limit orders of STP-aggregator clients are virtual and stored on the aggregator's server.
As a consequence, if someone takes an STP-aggregator as an LP, he will not be able to trade with its clients. Almost all MM algorithms running on an STP-aggregator will not produce positive profits.

Simple STP aggregator.
In a simple STP, all virtual client trade orders come to his PriceGivers as market orders. For example, a client has a SellLimit. As soon as Bid >= SellLimit occurs, a SELL market order is sent to the LP to which the Bid corresponds. The LP response to such a trade order is the result of execution of the client's SellLimit.
Obviously, in such an implementation slippages occur, the average of which is less than zero. I.e. clients incur losses.

Another STP-aggregator.
Here all virtual client trade orders come to its PriceGivers as limit orders (a market order is a limit order at a price that is decently (determined by the STP-aggregator) worse than the current one).
Let's analyse the execution algorithm on the example above:
At the moment Bid >= SellLimit, the virtual SellLimit is frozen (taken out of account by the execution system, the client can do nothing with it) and the same SellLimit is sent to the corresponding LP. All these actions have no effect on pricing. I.e. a little later than the freezing (for example, by 1 ms) a better Bid can be formed (from other LPs), but it almost (there are nuances) will not participate in the execution of our client's SellLimit.
LP, where SellLimit was sent, responds that it has executed one part of SellLimit, and the rest of the SellLimit for various reasons did not execute it - reject. After that the client receives a high-level open SELL-position for the executed volume and unfrozen SellLimit for the remaining volume.
The speed and quality of LP response depends on many factors. The duration of the response can be up to several seconds. It is not excluded situations when LP does not reply.

Note that (with such STP implementation) the aggregator's clients' limiters do not slide into the negative zone. Moreover, there are often positive slippages, which cover a significant part of fixed trading costs - commission.

Here you can also see how much the execution and pricing differs between exchanges and FOREX.

Nuances of execution.
It is very difficult to describe the many subtleties that affect the quality of execution. The simplest thing that aggregators do is to improve communication channels between LPs. However, the determinant of execution quality (FillRate) is the STP-aggregator's execution algorithms. This is a whole class of constantly improving (not all) interesting algorithms, some of which even affect STP-aggregator pricing. They are largely unaffiliated competitive solutions and very specific to the trader's guide, so I probably won't tell you about them.
The fact is that two STP-aggregators, which are in equal conditions, can differ significantly in FillRate, due to the use of different execution algorithms. This can be especially noticeable in toxics. I.e. the trading strategy's performance can be very dependent on the used aggregator, especially when trading volumes are growing.

If the trading indicators of two identical TSs launched in equal conditions on the same aggregator coincide, it is an indirect sign of a very high level of quality of execution algorithms of this aggregator.

[hrenfx]

ECN/STP.
ECN/STP - as an aggregator, algorithmically identical to STP. The only difference is that a new virtual LP is added to the aggregation - LP0: orders of the aggregator's clients.
With this simple technique ECN/STP aggregator's clients become PriceGivers as well. This means that the clients can make trades with each other directly.
The greater the turnover of LP0, the more profitable it is for the ECN/STP-aggregator, because unlike other LPs, LP0 does not require the aggregator to pay a commission for its use. Moreover, the flow of trade orders falling on LP0 can in no way be characterised as toxic. Since it is already trading of clients among themselves, even if there will be MM algorithms among them.
It turns out to be such an unproblematic LP.
But it should be clearly understood that to be able to be a PriceTaker for LP0-PriceGiver of an ECN/STP aggregator, you need to be its client. I.e. clients of other aggregators cannot take advantage of the sometimes great (best) price offers of LP0.
For example, you trade in two aggregators. In ECN/STP you put a limit inside the spread, improving Bid or Ask. But in the other aggregator you cannot buy/sell at that price from yourself, as they are not visible. That is, you will buy/sell, most likely, at a less favourable price than the one offered by your limiter.

ECN/STP Metamorphosis.
The strength of ECN/STP lies in the ability to exchange liquidity.
Imagine that some STP-aggregator has taken an ECN/STP-aggregator as its LP. This automatically means that clients of the STP-aggregator become PriceTakers of the ECN/STP-aggregator in particular (and other LPs from the STP-aggregator's list). It turns out that in the example above with two aggregators, you would see your order in the STP-aggregator placed in the other ECN/STP-aggregator. And if it turned out to be the best in the final STP-Level2, then you could make a trade with yourself. I.e. your order would be available to a much larger number of PriceTakers, which means that the probability of your order execution would increase.
It should be understood that each aggregator has its own PriceTakers base, and such a simple technique allows exchanging these bases, increasing the probability of execution of PriceGivers' orders. And this also benefits PriceTakers, as the prices are better.
Of course, if two ECN/STP-aggregators make each other their LPs, there will be a serious mismatch in Level2 of each - recursively the liquidity will be increased many times. That is why ECN/STP-aggregators exchange (or rather, can exchange) internal (LP0) liquidity among themselves through certain technical arrangements - tags.

Reality.
The LPs of most aggregators are banks with MM algorithms. This means that the toxic problem in most aggregators is extremely pressing. If one of these LPs notices toxicity from an aggregator, it notifies the aggregator. And the aggregator, in turn, is faced with a choice: disconnect the LP (because it does not want to incur losses), or disconnect its client - the source of toxicity. Most often, the aggregator refuses a profitable client because he is less profitable. This vicious scheme for traders works in most cases.
However, there are exceptions, and this is a serious redistribution of so long ago well-proven rules on institutional FOREX. ECN/STP aggregators have appeared, potentially able to turn the huge liquidity of PriceTekers into the liquidity of PriceGivers. I.e. to close traders on themselves, significantly reducing the income of classical banking MM-algorithms. It is only worth mentioning that according to modest calculations the scattered retail-FOREX has a monthly turnover of over one trillion USD. And almost all of this turnover is meat from the bones of clickers, which are currently nibbled by DCs (market maker scheme) and banks through classic STP and ECN/STP aggregators.

[hrenfx]

Simple HFT MM algorithm.
It is obvious that ECN/STP-aggregator has some insider information: who trades through it and how. A competent holder of such information creates a classifier of clients' profitability. For example, as a result of the analysis he finds out that the aggregator's clients consistently (and with the required speed) drain GBPJPY from 12:00 to 17:00 every Wednesday. What it means - it means that their money flows to LPs, which at that moment had the most competitive prices.
It begs the question, why give the drain to someone when you can keep it for yourself? Some will follow the grey scheme - they will switch off ECN/STP for this time and actually turn into a DC according to the market maker scheme. But there is an absolutely mutually beneficial white way - HFT MM-algorithm. Let it be ours.
In our example, every Wednesday from 12 to 17 o'clock we place our BuyLimit and SellLimit inside the GBPJPY spread at a minimum distance from the outer (not formed by us) spread and follow it all the time (this is where HFT - fast order dragging - is needed). This improves prices - narrows the spread. And merging clients will make trades at more favourable prices. However (see the mention about the analysis of the drain rate), their drain will flow to us, because they will make deals with us.
The mutual benefit is that we get profit, and the clients who are draining lose money at a lower rate than if we did not use HFT MM-algorithm. And if the owner of such MM-algorithm is ECN/STP-aggregator itself, the commission for using its LP0 is zero. And the client will pay the aggregator the commission in full for the transaction. That is, in fact, for such an owner the commission is even negative.

Adaptive HFT MM algorithm.
It is not necessary to have the above-mentioned insider information. It can be obtained by a simple way - probing.
To do this, of course, you will need to spend: you create your own spread according to the scheme above, but only with a minimum lot. ECN/STP clients will always make transactions with you, thus you will get insider information. Further everything is the same as in the scheme above.
The adaptability is that the insider information is taken in a sliding window, i.e. it is adjusted (controversially, of course) to the clients' moods and the market.
Note that any HFT MM algorithm will almost never be toxic, as it only takes money from PriceTakers, among which there are almost no HFT MM algorithms.
It should probably be mentioned that these elementary HFT MM algorithms are invented by me. And the exploited HFT MM algorithms are much more intelligent. However, their essence is roughly conveyed.

One of the mutually beneficial variants of attracting HFT MM-algorithms.

ECN/STP-aggregator always benefits when there are HFT MM-algorithms among its clients, because they provide high turnover. However, for independent HFT MM-algorithms trading costs in the form of commissions are a significant obstacle for them to open their opportunities to the full. For this reason, a marketing scheme is created, which can have quite clear mathematical calculations, when HFT MM-algorithm is given a negative commission. I.e. the aggregator for all orders of HFT MM-algorithm, which were executed on LP0, does not take commission from it, moreover, pays it a part of commission, taken in full from the other side of the deal - PriceTaker. Such a scheme was mentioned in passing in the description of the exchange algorithm.

HFT MM-algorithms: Exchange vs FOREX.

Not all ECN/STP-aggregators provide T&S-data on FOREX. There can be several reasons for this. For example, to hide turnovers and not to explicitly analyse the toxic coming from ECN/STP. However, still some ECN/STP aggregators provide such information.
Exchanges have T&S, but the data on the classification of transactions (PriceGivers or PriceTakers) is far from always of official (exchange) origin. Most often it is an independent algorithmic classification.
Both in the case of an exchange and an ECN/STP aggregator, correctly classified T&S data shows when PriceTakers/PriceGivers lose/win. In such a case, even probing and any analysis of Level2 dynamics is not required (with proper tech infrastructure - speed) to create an independent simple HFT MM algorithm. Since appropriately transformed and analysed T&S data is the aforementioned insider.