Hello, friends!

In today's article, I would like to bring your attention to a new analytical tool — Power trades. The main purpose of the tool is to search for zones with a large volume, traded in a very short period of time.

Such behavior of participants may lead to a short-term change in the ratio of supply and demand and, as a result, the breaking of an existing trend or the formation of a new one. Let’s dive in this situation more deeply.

## The supply/demand in the realities of the modern market

Before we go ahead, it is necessary to make a small digression into the theory to understand how pricing occurs in the market and what conclusions can be drawn from this.

Since the appearance of first exchange markets to the present day, a huge amount of research has been carried out on the nature of the market, many of these theories formed the basis for portfolio management models, as well as the options pricing models that are widely used now.

For quite a long time it was believed that the market movement has a random nature and occurs in accordance with the normal and, subsequently, lognormal distribution models.

The refutation of this thesis was found quite quickly and consisted in the fact that according to the normal distribution the price deviation in 3 standard deviations from its mathematical expectation should have a probability in about 0.1%, but in practice, it turned out that the probability of such movement is many times higher.

Such a feature of the process was called the “Fat tail”. It turned out that with insignificant price deviations from their mathematical expectation, the market distribution behaved almost identical to the normal distribution, but the probability of a strong deviation significantly exceeded the probability observed under a normal distribution model.

Let's take a look at the illustration of these processes.

At the illustration, a plot marked as the “Fat-Tailed Distribution” has a probability (value along the Y-axis) that is 2-5 times higher than the probability with a normal distribution.

So, what conclusions can we draw from this:

- The market is not distributed by normal or lognormal distribution models.
- The distribution of the market is dynamic and constantly changes over time, in other words, the market “breathes”.
- The market distribution is very close to normal with minor price deviations from its mathematical expectation value.

The main conclusion that is important for us is that with minor price changes, the market will behave in accordance with the normal distribution, which in turn means that **the probability of movement up and down in the short term will tend to 50/50.**

Let's take a look at the depth of market and the movement inside it, keeping in mind the above thesis. Suppose we have the following DOM:

The green lines on the graph represent Bids and their volumes, the red lines respectively Asks and their volumes, the black one denotes trades. For the sake of simplicity, let's assume that the trade volume is 3.

It turns out that, with a 50/50 probability, the same number of trades at Asks and Bids levels will take place, however, due to the fact that the volume of Asks is 10 times less than the volume of Bids the fourth trade will break through the first Asks level and jump to the next one.

**Conclusion number 2. Having a probability of 50/50, the price always moves towards the least resistance, because the DOM is thinner there.**

Of course, it should be understood that the example described above is significantly simplified and does not take into account the fact that the DOM will dynamically change, but this in no way changes the basic meaning:

- In the short term, the probability tends to 50/50
- The price always goes in the direction of least resistance

## Supply/demand imbalance

Let's ask ourselves, what will happen if, in a very short period of time, a set of same directional trades with a very large volume occurs?

The answer is quite obvious - such an event will significantly weaken one of the sides of the DOM, and as we know the market always moves towards the least resistance.

This gives us the understanding that in the short term we have an imbalance and therefore we can draw the conclusion that the market can change its existing trend or, for example, go through a strong resistance/support level.

## Power Trades scanner practical usage

As a logical sequel of Volume analysis tools, "Power trades scanner" tool also relays on the symbol's volume value. It searches for zones, using the predefined settings and these settings are unique per each symbol. So, first of all, let's analyze the screen settings

**Min trade volume**- allows you to set the minimum amount of trade included in the zone**Max trade volume**- allows you to set the maximum amount of trade included in the zone**Total volume**- the minimum required volume to build the zone**Time interval**- the maximum time in seconds that can be spent on building a zone**Basis volume interval**- seconds based time, that will be used to calculate the Basis volume**Zone height**- zone height in ticks**Level 2 level count**- the number of level2 levels used**Filter by Delta**- the percentage of “clean” delta inside the zone

Special attention should be paid to **Total volume** and **Time interval** settings because they are key and allow you to determine what the total volume of the zone should be and how long it should be formed.

I also want to draw your attention to the **Basis volume interval** setting. It allows you to determine the time, for which, the volume will be taken as the basis. We need a basis in order to determine how strong the zone is.

For example, suppose we got a zone with a volume of 1000 lots, which was formed in 2 seconds in order to understand how potentially influential such a zone is, we need to correlate the volume of the zone with a certain basis.

As a basis, let's take the volume formed in 5 minutes. Let the 5 minutes volume be 10,000 lots, then the Basis ratio of our zone turned out to be 1000/10000 * 100 = 10%, thus within 2 seconds, 10% of the five-minute volume was traded, which is not very much. The following conclusion can be made - yes, this volume surge is strong but not critical, because at the moment the trading is very intense and the DOM is dense.

But what if the 5 minutes volume was not 10,000 lots, but for example 1,100? Then the base ratio of our zone turned out to be 1000/1100 * 100 = 91%, which means that the trade is not intensive now and the influence of the zone can be very significant.

Thus, the same volume of 1000 lots in different situations can be both significant and insignificant and affect the DOM differently.

Another interesting metric is the **Level2 ratio%**. This metric allows you to correlate the volume of the zone and the volume of the corresponding side of Level2, which remained after the formation of the zone.

In other words, this parameter allows you to determine how strongly at the moment the zone affected Level2. The level2 side is determined based on the Delta parameter of the corresponding zone. If Delta> 0, then the Ask side is used, otherwise Bid.

## Zones catching

At the moment, Power Trades should be configured only manually, but we are already working on semi-automatic parameters detection.

To efficiently search for zones, you should set Total volume and Time interval parameters, while the remaining parameters can be left with defaults. They will be needed if you want to adjust to an asset behavior more accurately or, for example, do your own experiments looking for some anomalies.

Let's configure Power Trades for ETHBTC. To work with parameters, we will use the quick setting screen, because for us there is no need to work with all settings list.

Let us assume that we do not know about the volumes characteristic for this symbol, therefore we will set the Time interval to 5 seconds, and Total volume - to some large value, for example, 100,000 and turn on Power Trades.

Based on these settings we didn’t find anything, which means that the volume of 100,000 in 5 seconds is very large and not appropriate for this asset.

Let's change the Total volume and Time interval parameters to 1000 in 5 seconds and see what will be found.

Excellent results are found this time:

After we have found the range of characteristic volumes for the symbol, we can already analyze the nature of the reactions of such zones and set remaining parameters more accurately.

Be sure, Power Trades scanner is very flexible and allows you to adjust to the traded asset exploring its nature with high precision.

## Zones examples and their influence

Now let's go through several examples of "Power trades" zones and try to analyze their behavior. The first zone was formed on ESM9 and led to a reversal of the trend:

After that, another zone was formed, which once again reversed the tendency:

Few other examples on ESM9

Examples on ETHBTC:

Examples on BTCUSD:

## Usage tips

- In my opinion, it is most useful to combine Power Trades with the Order flow surface, because the result of Power Trades is the zone where the imbalance of supply and demand occurred, and the Order flow allows you to observe in real time the changes of the Level 2 structure and therefore make a conclusion as to which way the market could move.
- It is also very informative to use Power Trades with the techniques of volumetric analysis since often, the found zones can serve as strong resistance/support levels or, on the contrary, a format already existing levels, giving a signal of a tendency reversal or breaking down the level.