BeliEVE’s trading style

A review of BeliEVE, a new cutting edge Savings Management platform, part II

Pi3rrot
9 min readJul 9, 2019

I. Introduction — few additional words about BeliEVE

BeliEVE positions itself as a Savings Management platform on crypto-markets. A first article introducing this system, and presenting its performances from April to June 2019 can be read here.

This second article is now more oriented toward people already using BeliEVE, and who wish to get a deeper insight about BeliEVE’s trading style.

Before delving into trade data, it is important to clarify that this article has been written without direct exchanges with BeliEVE development team. Hence I have tried as much as possible to present facts only, and certainly avoided to draw any firm conclusions from them.

Nonetheless, I am taking the opportunity of this introduction to recall two major challenges (among others) a Savings Management platform such as BeliEVE has to face:

  • Ensuring steady profits, even if minimal, on the opposite to large profits every once in a while with possible significant losses in-between. The first scenario (steady profits) certainly contradicts the reality of markets, and most notably crypto-markets which are known to be volatile. In addition, cycles can be short, and bear markets do be as present as bull markets. A Savings Management system has thus to embed specific features to be able to cope with these constraints.
  • Scaling with the asset it manages, i.e. maintaining its performances as the asset under management is growing. It also means ensuring that asset from all users is used in a balanced way.

These two challenges necessarily shape the way a Savings Management platform such as BeliEVE works. I thus invite the interrogative reader to keep these in mind when exploring topics tackled below:

  • Traded pairs and their contribution to monthly performances,
  • Average gain profile per trade,
  • Distributions of trades versus price increase & profits,
  • BeliEVE’s transaction volume estimation.

To support these analyses, data from the 5 portfolios (5 user accounts) presented in previous article are used in chapter 2. Data from a 6th portfolio, Aj retaining Aggressive mode, has then been added in subsequent chapters.

II. Traded pairs and their contribution to monthly performances

BeliEVE conducts transactions on Binance USDT market. Trades from April to June 2019 were achieved with following list of crypto-currencies and tokens: ADA, BCHABC, BCHSV, BNB, BTC, EOS, ETC, ETH, ICX, IOTA, LTC, NEO, ONT, TRX, VET, XLM, XRP.

It is worth mentioning that Binance announced delisting of BCHSV the 15th of April 2019, and since then is not traded by BeliEVE any longer.

Graphs below gather per crypto contributions to gross profits for our 5 portfolios over April, May & June (please read previous article to have some details about the notion of “gross profits”).

Contributions per crypto to gross profits for 5 different portfolios — April 2019
Contributions per crypto to gross profits for 5 different portfolios — May 2019
Contributions per crypto to gross profits for 5 different portfolios — June 2019

While all the data delivered in this form may be a bit hard to review, let’s outline some information:

  • We could notice in previous article that performances vary between different portfolios, even with same mode. We can see here one reason as trading opportunities from different crypto pairs are not distributed in an equal fashion between them. For instance, in April, Ak portfolio benefited significantly from ONT trades, while contributions from the different cryptos are more balanced for the other users. In May, Ap portfolio saw a major share of its profits coming from LTC trades. And finally in June, C6M3A1 got most of its earning from BNB trades. From this review, BeliEVE seems to operate a “rotation” in use of asset as new crypto opportunities appear.
  • Nonetheless, one may also see that the hierarchy of contributions per crypto is kept between the different portfolios. ONT, ETH, IOTA & NEO are for instance the most profitable cryptos in April. In May, ONT, NEO & IOTA remains in the top 4, while ETH is replaced with LTC. Finally in June, LTC and ONT stays among the best performers, joined by BNB & TRX which push aside ETH & IOTA. Over these 3 months, one may finally observe that BeliEVE appears especially efficient in trading with ONT.
  • As a last comment and a surprise, while BTC does be traded as well, its contribution remains unoticed in studied portfolios compared to those of the other cryptos.

III. Average gain profile per trade

As introduced previously, BeliEVE conducts trades on Binance exchange. It does not operate margin trading (and most notably no short orders).

With this information, it is possible to draw simply an average gain profile per trade showing net gains, BeliEVE’s fees (12,5% assuming these are paid with PCL), and exchange’s fees (knowing that at the start, these amount on Binance to 0,1% of the total transaction volume, i.e. buy & sell operations).

Monthly average gain profile per trade — all portfolios

Trade data supporting above graphs span from the 1st of April to the 23rd of June.

They show that BeliEVE does not focus on achieving necessarily large profits per trade. In June especially, net gain is actually quite low. On the other hand, BeliEVE is handling a huge amount of trades. Indeed according considered transaction data, it reached more than 21 000 transactions this last month.

While with such a trading strategy, most of the trades may certainly be closed successfully faster (hence possibly ensuring a steady rotation in used asset), one may notice that Binance retains an important share of the total trade profits through its fees. In June it reaches a max, amounting to 50% of the total trade profit.

However, thanks to this important transaction volume, and by acquiring the appropriate number of BNBs, BeliEVE activated VIP2 level the 24th of June. With this upgrade, BeliEVE benefits from lower Binance’s fees, and thus can operate more profitable trades. Curious readers may wonder which level could be reached in the short term and an estimation is proposed chapter 5.

IV. Distributions of trades versus price increase & profits

IV.a. Distribution of trades versus price increase

After having reviewed the average gain profile per trade, the acute reader will deduce that BeliEVE affords itself a low increase between sell and buy prices. This can be confirmed by drawing corresponding distributions of trades versus achieved price increase.

Trade distribution versus price increase (sell price increase in percent vs buy price)

A majority of trades are closed with a price increase between 0,25% and 0,4% . The profile of these distributions is surprisingly constant through the 3 months of study. Other results (average gain profile, distribution of trades versus achieved profits discussed below) show more change over this same period, possibly because of a higher dependency to crypto-market conditions.

IV.b. Distribution of trades versus achieved profits

Distribution of trades versus achieved profits are drawn in following graphs using a logarithmic scale to better distinguish low gain levels.

Trade profit distributions (blue curve, left vertical axis / green curve, right vertical axis)

Several comments can be made from these graphs:

  • They show first that a majority of trades brings back less than 0,1 USDT each. This status has already been introduced through past chapters 3 and 4.a.
  • Additionally, the discerning reader will realize that profit distribution in April is clearly not centered around the average gross profit given in chapter 3 (1,15 USDT). After a closer look on data, one actually realizes that only 8,7% of the trades bring 89% of the gross profits. A similar status can be identified in subsequent months, albeit less noticeable.
  • An increase in the number of trades month per month can be noticed. This is partly due to an increase in the total asset of our 6 portfolios. Nonetheless, at near constant asset, an increase of trades is also noticed over these 3 months. A second reason could be thus a change in the crypto-market itself over this period, from a ranging market to a bull market.
  • Finally, drawn quantities reveal the number of trades which profit is re-invested in Portfolio (green curves). In April and June, this amount has been low relatively to the number of trades delivering directly profits into the Sweet pot. It rose however to a much higher fraction in May (nearing to one third of this number).

V. BeliEVE’s transaction volume estimation

V.a. Motivation, quantities of interest & estimation methodology

Average gain profile for our 6 portfolios has been presented in chapter 3. The noticable share of Binance’s fees over the total trade profit has most notably been emphasized (from about 26% in April to 50% in June).

By switching to VIP2 level the 24th of June, BeliEVE now benefits from a reduction of 20% on transaction fees on limit orders. Considering a gain profile like that of June, it means an increase in net profit of 17,5%. Hence, curious reader may thus wonders if BeliEVE could pretend to a higher VIP level to benefit from larger reductions.

VIP levels on Binance are depending on both the number of BNB tokens held on the account, and the transaction volume over the past 30 days in BTC. Assuming PECULIUM team is able to acquire the required number of BNB tokens, we will make an estimation of current transaction volume. To assess this quantity, unknown to me, following data are used:

  • the total asset managed by BeliEVE, communicated by the team to be over 500 kUSDT in April, over 600 kUSDT in May and more recently above 700 kUSDT begining of July,
  • the total asset managed for our 6 portfolios,
  • and finally the transaction volume over the past 30 days in BTC for our 6 portfolios.

I used the two first quantities as a scaling factor applied to the last one. This is certainly a simple and approximate approach, which gives nonetheless some hints about current status.

V.b. Estimation

Below graph gathers several information. Estimated BeliEVE’s transaction volume over the past 30 days in BTC is of course drawn (in green) along with minimal levels for VIP1, 2, 3 & 4 status (horizontal dashed lines). Finally, the BTC/USDT close rate is also added to the chart (in blue).

Estimation of BeliEVE’s transaction volume over the past 30 days in BTC (green) & BTC close rate in USDT on Binance (blue)

According to this estimation, BeliEVE cannot currently pretend to VIP3 level as transaction volume over the past 30 days is about 3,3 kBTC (4,5 kBTC required for VIP3 level). This volume appears to have been decreasing since end of May, and even more sharply starting the 10th of June, as BTC price has steadily increased. The fact that levels are stated in BTC is certainly not helping when BTC price is rising.

VI. Conclusions

Here ends the second part of this analysis focused on BeliEVE’s trading style. Similarly to the first part, I invite the reader to take some care when reviewing this study based on data issued over 3 months only.

Several information, not so easily accessible at first, have been underlined, such as the list of traded cryptos and their contribution to BeliEVE’s profits, what is the average gain profile like and the noticable share of Binance’s fees, an estimation of BeliEVE’s transaction volume, etc.

Among other aspects, BeliEVE presents a trading strategy that appears favoring a high number of successful trades with small profits over a few ones with larger gains. This approach may present some advantages for a Savings Management platform such as BeliEVE, striving to handle large asset size and to make returns steady.

If you feel this work presents some interesting information, and would you like to give BeliEVE a try by signing up an account, please, consider using my referral link. ;)

If you do invest, please, do it wisely by respecting the one golden rule in investment: never invest money that you can’t afford to lose.

To know more about BeliEVE:

English Telegram-Group
French Telegram-Group
German Telegram-Group
Telegram Announcement Channel
Twitter
Facebook
Reddit
LinkedIn
Github
Youtube

--

--