We present your attention to the translation of trader and investor Sylvain Ribes (Sylvain Ribes) article.
In this post I will show you why I think that over $ 3 billion. All trade cryptocurrency fabricated and how OKex, in terms of volume №1 market was the main culprit, because its business does not really exist until 93%. I will try to prove it by analyzing public data.
When I started collecting information for this post, there was no idea that I would write about the fabulous volume.
First, I plan to collect information about the liquidity of crypto-currency assets, which can measure the amount of trade. I thought I would have an interesting index for the value analysis of the property.
I decided to collect a glass quote for the main transactions for this purpose and to check how much a specific crypto currency is being checked if it is sold for $ 50 000. In this article, I use this image as "slipse" (the term is included in the definition appendix). Then I will adjust the slipjas indicator, at the price of each exchange, and sell more less, and calculate the crypto currency market capitalization and change the sales volume.
When I expected a decreasing function slippage of volume, depending on the currency, there might be some difference. Finally, if you have a lot of business in a particular pair, but there should be a big competition among market makers to meet the active buyers and vendor requests. This competition should quote the quote and reduce the spread.
is not it?
It turned out that there was a clear tendency, yes. However, where I anticipated small differences between specific crypto coins, I discovered huge discrimination between the exchange. At your fingertips ("because their users behave differently"), they can only be hand-dropping, but they can only be interpreted by a few more than 95% of the height.
The leader of the group, with OKeng, №1 and CoinMarketCap and LiveCoinWatch's rating, total trading volume of $ 1,7 billion.
OKex is a Scrub city
OKex, Cracen, Bitfinex and GDAX: The average amount of graph slip and trading volume between all couples represents cryptocurrency calculation with daily volume trading volume of $ 48 at $ 24, 000 for the first time in 100 kriptobirzhah.
For example, you can notice that the blue dot below the chart is a pair of GDAX exchange, about $ 200 million trading volume, slippage less than 0.1%
It shows the schedule is simply amazing, although the first three exchanges behave similarly, as all pairs of tattoo exchanges are marked by red, they have seen much larger slip than the total trading volume. As I explained before, this can only say one thing: the majority of OKex is completely fabricated.
In addition, for the benefit of reading, I decided to remove all the information of more than 4% of the slip from the array. If you enter this data, the OKex chart will be more weird and you will need a logarithmic scale:
If you decide to sell only $ 50 000 property, the slip will be up to 10% even though this pair of one pair and analysis (5/06/03) considers the following pairs to make trading volume up to $ 18 million. -
Although these statistics clearly show by themselves that the vast majority of the trading volume of the Oxex Exchange is fabricated, I personally did not see how they did it. So, I went to their platform and looked at the history of trading for some couples. They are, in reality, viciously distorting the volume in a clear and artificial way:
These exchanges will compare this absolutely stabilizing schedule in a continuous sinusoid form with what happens:
The effects of frost on acute growth, sharp storms, high volatility, it does not look like the schedule of oscilloscope at school.
"But in China, there is no case in the case! I do not think the scheduler OKex did not indicate the high quality of thinking about engineers who really think about how to make a less visible volume of a constant flow of fabulous trade transactions, but whatever they did, it was a perfect sign wave.
How bad is it?
Although it is obvious that most of the Oaxax volume is fraud, how can it be calculated, 90, 95 or 99%? I suggest using the following methods:
Make a list of trusted exchanges that work closely with such violations;
Setting up a regression on their data sets, which will be able to predict the amount of trades based on the extinction slippage;
Comparing the declared volume of OKex with the calculation volume according to our model.
I used the following exchange data for its calculation: bitfinex, gdax, poloniex, bistamp, mithun and cracken. Instead of $ 50 000 sales sales analysis, I imitated $ 20 000 amounts: given data volatility with low trading volume, I also decided to change the indicator usage.
The information on reliable exchanges shows that, at this time, the volume as a function of the slip shows:
Note, due to very volatile data, if SLIP exceeds 0,7%, then any model becomes weird. The above recommended model is suitable for spacing under 0.7%. After this threshold quality, the only reasonable estimate is that the expected volume is less than $ 1 million.
As you can see, when OKex data is entered in the above mentioned array, the image may be wrong:
Selected crypto-coins from 28, with more than $ 100 000 trading volumes, only 11% less than 0.7% of slippages have been shown:
According to the impressive 93,6% according to the total ratio of the volume of fraud for this pair, according to the model. If you add the remote pair (slip> 0.7%), the situation will not change in any direction.
Perhaps the regression I used does not work very well with very large volumes, since there is no reliable data set. In this case, only the honest solution will be removed BTC / USD. Still, this image has become illusory: 92.1% of all trading volume of Wexx may possibly be fraud.
Exchange Huobi, there is not much difference
Just like Oaksx, Hubabis stopped due to strict restrictions in China, but Hubi was reopened after the license. Using the procedure mentioned above, we receive the following results:
81.8% of the volumes are fraudulent, not as bad as the closest competitor, but still much more.
If you are sure that Huobi is trading once, it is much more than OKex to look at the history of biological, see the history of continuous, not particularly obvious, fake transactions, will still provide significant support to the general volume:
Scamers Chinese Armed
Perhaps you have noticed, or maybe not, but more Chinese coins have added a group of Chinese exchanges, all of which show higher trading volume, but no one has heard anything about them. Obviously, most of them use the same user interface and trading engine.
Although the list can continue, firstly, I used the following exchanges: Lbank, Exx, RightBTC, CoinEgg, Zb, BitZ, Bibox, CoinEx, BTC-Alpha ...
These platforms are so unceremoniously describing their trading volume lies, even it does not need to analyze them using a model, but you find it yourself ... It's a shame that CoinMarketCap and LiveCoinWatch will say this on a par with legal exchange, she is deceiving Been, sometimes not experiencing the best of time.
Hit BTC and Beyonce
For a number of reasons, I got suspicions about two leaders in the business of AltaCone, Hitbius and Beyonce. Here they show results compared to the different "decent" exchanges:
It is very easy to understand that in certain quantities, both exchanges, especially bills, marked oranges, very little liquids, and therefore offer suspicions.
Run with Okex and Huba before the same analysis, we can get the following results. First HitBTC:
Although the above figures do not seem very important, they prove that HitBTC models are somewhat less liquid than the exchange. This small difference between the declared and re-calculated trading volumes may be displayed for a number of reasons, due to general deviation due to general deviation.
But the results of the business are more interesting.
70% are concerned about the differences with our mathematical predictions. But do not forget that our model's primary information is a specific pairing slip, which is not the full volume of trading.
Of course, I know from my own experience that the Briens strictly prohibits the control of trade with the use of robots. For some time I argued with them how stupid these restrictions are trying to make, because they only interfere with growth and liquidity.
It is possible that due to the limitations of the data, the use of a lot of different exchange market meykingovye strategies, do not use them in Binance, because it must be blocked permanently, even without knowing what must have exceeded the limits.
Without a few professional market makers, the quotes of the quotes become slim and it is easy to see that my model will stop working. Nevertheless, it can be effective for closely monitoring the declared trading volumes of Bin Nies, although the history of trading volumes does not analyze and publish obvious activities.
Caution
Despite the fact that I have almost no doubt about my statement, the statistics should not be without looking behind. And here's why:
As I mentioned, restrictions on the business's use of robots play an important role in the case of insurance. Improving the liquidity in specific exchanges It is easy to use good quality robots.
The effect can also have a commission. More stars, less market makers will exceed their quote and reduce spreads.
I do 24 hours only to collect an average sample of data and deviate control. I'm not a scientist, not a fan of statistics, but as a result, it's all very reliable. I'm ready to hear other arguments.
Icebergs and hidden orders may be considered by some exchanges to give their users the chance to hide their limit orders, however, that Bitfinex offers such opportunities and thus the same as many other "exchange models", I believe that liquidity does not work to change this volume if we icebergs , To ignore the effects of hidden orders.
Different groups of users can behave differently on different exchanges, however, based on the personal experience of my algorithmic trading, I can say that the effects of these differences on the overall situation are not at least systemic.
Why bothered?
Someone might say that "the market is not controlled, such steps are not even illegal, why exchange should not be exchanged, what do they want?" And he'll be wrong. Correctly because the market is not controlled, responsible behavior is beneficial for the market players. Information dissemination and deviation that uses arrogant practice is the lowest that we can do.
Someone can say that "they do not hurt anybody", and that's also wrong. First, by increasing their trading volume, they get the opportunity to cheat the trust of ordinary investors. Moreover, if you are not even a credible private investor and are not an enterprise investor, these actions and some crypto currency may have an impact on the valuation. In particular, up to 75% of the trading volume
Showing most artificial volumes, these coins seem to be more attractive for traders, since they actually attract more attention than they (the best way to volume the volumes). Also, if fraudulent rules are always present on spreads, people are slightly esteemed compared to such a crypto currency.
Dishonest exchanges can attract more real-time customers, with more real volume and liquidity, at the same time allow users to trade faster and / or at a lower rate due to slippers.
Finally, if the trading volume is distorted, even if, strictly speaking, illegal operations, it is better to be a precursor, more serious violation, so all users may take special care, such as taking decisions related to exchange trading.
Information too
According to my calculations, over $ 3 billion. Trading volume is unreal. Probably more somehow, if this practice is not encouraged, if at least, the majority of the popular connectors, and their users are neglected when all you have to do is just look at the numbers to realize that something is wrong with them.
At the moment, after the bullish excitement of 2017, the crypto resource bear is under serious pressure from the market. I firmly believe that in a fair trading environment the growth will not continue. The market awareness of the entire ecosystem and the crypto currency has increased so much that it is impossible to approve such Frank manipulation.
We all repeat: "Crypto-currencies do not require regulation!" Now it is time to prove it. Otherwise, in the current situation, market conditions are crypto-currency, it seems, that the free market does not work.
Apply
My definition of slippage is a bit wrong. I set the percentage change in the quoted value of the quote and the average value of the price, which it will agree to sell the asset.
For those who want to study my collected information, here is the original information link, 1 sheet
For data collection, I used the CCXT Python library at https://xiuthub.com/ccxt/ccxt
Finally, all the posts in this post, which, in my opinion, have a false description list of their trading volume. If you look only on the chart of trading history of Tidex, Liqui and Wex, although I do not believe it (although they do not believe they are probably falsified strongly). Most likely, such exchanges are very big. If we analyze the use of other methods on OKex, the futures contract is recommended, because OKex futures will be the surprise to be sure about the claims of huge amounts.
Popular bitratex is absolutely clear. As far as I know, there is no fear of cryptopia or Kuoquin (although there is some doubt later, because I know for sure that it has started its work in a large amount of fraud trade).
I did not see the Korea Exchange analysis due to lack of data. With Bithumb, it seems to be all right, but it still costs to check. With respect to Coinnest or Upbit, I can not comment on anything.