Sustainable balancing strategies for bear markets

In the past, FMZ officially released a sustainable grid strategy, which is quite popular among users, and the TRX trading platform has gained a lot of revenue in the past year with more risk control.

1. Parameters such as initial price, grid spacing, grid value, multi-space mode, etc. are more cumbersome to set up, with a larger impact on earnings, and more difficult for beginners to set up.
2. The long-term grid strategy has a high risk of doing nothing and a relatively low risk of doing nothing, and even if the net value set is small, the impact on the price of doing nothing is not great.
3. A perpetual contract grid can choose to do only more to avoid the risk of doing nothing, but it is still possible to see for now. But it is necessary to face the current price exceeding the initial price, which leads to the problem of holding empty stocks, and the initial price needs to be reset.

I have written about the principles of balancing strategies and the comparison with grid strategies before, and I can still refer to them:https://www.fmz.com/digest-topic/5930The equilibrium strategy always holds a fixed proportion of value or value of the position, sell a little when you go, buy when you fall, and can operate with a simple setup. Even if the price of the currency rises a lot, there is no risk of trampling. The problem with the equilibrium strategy is low capital utilization, there is no simple way to leverage. While the permanent contract can solve the problem.

For beginners, it is highly recommended to use a balancing strategy, which is simple to operate, and only needs to set a holding ratio or holding value parameter, so that you can run without worrying about the price rising.

In order to facilitate the retesting of more transaction pairs, this document will show the complete retesting process, where users can adjust the different parameters and transaction pairs for contrast. The version is Python 3, requires proxy downloading, users can download Anancoda 3 themselves or run it through Google's colab.

import requests
from datetime import date,datetime
import time
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import requests, zipfile, io
%matplotlib inline

## 当前交易对
Info = requests.get('https://fapi.binance.com/fapi/v1/exchangeInfo')
symbols = [s['symbol'] for s in Info.json()['symbols']]
symbols = list(set(filter(lambda x: x[-4:] == 'USDT', [s.split('_')[0] for s in symbols]))-
                 set(['1000SHIBUSDT','1000XECUSDT','BTCDOMUSDT','DEFIUSDT','BTCSTUSDT'])) + ['SHIBUSDT','XECUSDT']
print(symbols)

['FLMUSDT', 'ICPUSDT', 'CHZUSDT', 'APEUSDT', 'DARUSDT', 'TLMUSDT', 'ETHUSDT', 'STMXUSDT', 'ENJUSDT', 'LINKUSDT', 'OGNUSDT', 'RSRUSDT', 'QTUMUSDT', 'UNIUSDT', 'BNBUSDT', 'XLMUSDT', 'ATOMUSDT', 'LPTUSDT', 'UNFIUSDT', 'DASHUSDT', 'BTCUSDT', 'NEOUSDT', 'AAVEUSDT', 'DUSKUSDT', 'XRPUSDT', 'IOTXUSDT', 'CVCUSDT', 'SANDUSDT', 'XTZUSDT', 'IOTAUSDT', 'BELUSDT', 'MANAUSDT', 'IOSTUSDT', 'IMXUSDT', 'THETAUSDT', 'SCUSDT', 'DOGEUSDT', 'CELOUSDT', 'BNXUSDT', 'SNXUSDT', 'ZRXUSDT', 'HBARUSDT', 'DOTUSDT', 'ANKRUSDT', 'CELRUSDT', 'BAKEUSDT', 'GALUSDT', 'ICXUSDT', 'LRCUSDT', 'AVAXUSDT', 'C98USDT', 'MTLUSDT', 'FTTUSDT', 'MASKUSDT', 'RLCUSDT', 'MATICUSDT', 'COMPUSDT', 'BLZUSDT', 'CRVUSDT', 'ZECUSDT', 'RUNEUSDT', 'LITUSDT', 'ONEUSDT', 'ADAUSDT', 'NKNUSDT', 'LTCUSDT', 'ATAUSDT', 'GALAUSDT', 'BALUSDT', 'ROSEUSDT', 'EOSUSDT', 'YFIUSDT', 'SKLUSDT', 'BANDUSDT', 'ALGOUSDT', 'NEARUSDT', 'AXSUSDT', 'KSMUSDT', 'AUDIOUSDT', 'SRMUSDT', 'HNTUSDT', 'MKRUSDT', 'KLAYUSDT', 'FLOWUSDT', 'STORJUSDT', 'BCHUSDT', 'DYDXUSDT', 'ARUSDT', 'GMTUSDT', 'CHRUSDT', 'API3USDT', 'VETUSDT', 'KAVAUSDT', 'WAVESUSDT', 'EGLDUSDT', 'SFPUSDT', 'RENUSDT', 'SUSHIUSDT', 'SOLUSDT', 'RVNUSDT', 'ONTUSDT', 'BTSUSDT', 'ZILUSDT', 'GTCUSDT', 'ZENUSDT', 'ALICEUSDT', 'ETCUSDT', 'TRXUSDT', 'TOMOUSDT', 'FILUSDT', 'ARPAUSDT', 'CTKUSDT', 'BATUSDT', 'SXPUSDT', '1INCHUSDT', 'HOTUSDT', 'WOOUSDT', 'LINAUSDT', 'REEFUSDT', 'GRTUSDT', 'RAYUSDT', 'COTIUSDT', 'XMRUSDT', 'PEOPLEUSDT', 'OCEANUSDT', 'JASMYUSDT', 'TRBUSDT', 'ANTUSDT', 'XEMUSDT', 'DGBUSDT', 'ENSUSDT', 'OMGUSDT', 'ALPHAUSDT', 'FTMUSDT', 'DENTUSDT', 'KNCUSDT', 'CTSIUSDT', 'SHIBUSDT', 'XECUSDT']

#获取任意周期K线的函数
def GetKlines(symbol='BTCUSDT',start='2020-8-10',end='2021-8-10',period='1h',base='fapi',v = 'v1'):
    Klines = []
    start_time = int(time.mktime(datetime.strptime(start, "%Y-%m-%d").timetuple()))*1000 + 8*60*60*1000
    end_time =  min(int(time.mktime(datetime.strptime(end, "%Y-%m-%d").timetuple()))*1000 + 8*60*60*1000,time.time()*1000)
    intervel_map = {'m':60*1000,'h':60*60*1000,'d':24*60*60*1000}
    while start_time < end_time:
        mid_time = start_time+1000*int(period[:-1])*intervel_map[period[-1]]
        url = 'https://'+base+'.binance.com/'+base+'/'+v+'/klines?symbol=%s&interval=%s&startTime=%s&endTime=%s&limit=1000'%(symbol,period,start_time,mid_time)
        #print(url)
        res = requests.get(url)
        res_list = res.json()
        if type(res_list) == list and len(res_list) > 0:
            start_time = res_list[-1][0]+int(period[:-1])*intervel_map[period[-1]]
            Klines += res_list
        if type(res_list) == list and len(res_list) == 0:
            start_time = start_time+1000*int(period[:-1])*intervel_map[period[-1]]
        if mid_time >= end_time:
            break

    df = pd.DataFrame(Klines,columns=['time','open','high','low','close','amount','end_time','volume','count','buy_amount','buy_volume','null']).astype('float')
    df.index = pd.to_datetime(df.time,unit='ms')
    return df

By downloading the closing prices of all the trading pairs so far in 2021, we can observe the overall changes in the market index: from 2021 to 2022 there is no doubt that the bull market market, once 14 times higher, can be said that the index is gold, and many currencies have risen hundreds of times. However, in 2022, we opened a bear market that has been going on for half a year, with the index once falling by 80%, and dozens of currencies withdrawing by more than 90%. Such a crash reflects the enormous risk of the grid strategy.

Currently the index is at around 3, which is an improvement of 200% compared to the beginning of 2021, which should be a relative bottom given the development of the market.

The highest price of the currency, which has risen more than 10 times since the beginning of the year:

‘MKRUSDT’: 10.294, ‘CRVUSDT’: 10.513, ‘STORJUSDT’: 10.674, ‘SKLUSDT’: 11.009, ‘CVCUSDT’: 11.026, ‘SRMUSDT’: 11.031, ‘QTUMUSDT’: 12.066, ‘ALPHAUSDT’: 12.103, ‘ZENUSDT’: 12.631, ‘VETUSDT’: 13.296, ‘ROSEUSDT’: 13.429, ‘FTTUSDT’: 13.705, ‘IOSTUSDT’: 13.786, ‘COTIUSDT’: 13.958, ‘NEARUSDT’: 14.855, ‘HBARUSDT’: 15.312, ‘RLCUSDT’: 15.432, ‘SCUSDT’: 15.6, ‘GALAUSDT’: 15.722, ‘RUNEUSDT’: 15.795, ‘ADAUSDT’: 16.94, ‘MTLUSDT’: 17.18, ‘BNBUSDT’: 17.899, ‘RVNUSDT’: 18.169, ‘EGLDUSDT’: 18.879, ‘LRCUSDT’: 19.499, ‘ANKRUSDT’: 21.398, ‘ETCUSDT’: 23.51, ‘DUSKUSDT’: 23.55, ‘AUDIOUSDT’: 25.306, ‘OGNUSDT’: 25.524, ‘GMTUSDT’: 28.83, ‘ENJUSDT’: 33.073, ‘STMXUSDT’: 33.18, ‘IOTXUSDT’: 35.866, ‘AVAXUSDT’: 36.946, ‘CHZUSDT’: 37.128, ‘CELRUSDT’: 37.273, ‘HNTUSDT’: 38.779, ‘CTSIUSDT’: 41.108, ‘HOTUSDT’: 46.466, ‘CHRUSDT’: 61.091, ‘MANAUSDT’: 62.143, ‘NKNUSDT’: 70.636, ‘ONEUSDT’: 84.132, ‘DENTUSDT’: 99.973, ‘DOGEUSDT’: 121.447, ‘SOLUSDT’: 140.296, ‘MATICUSDT’: 161.846, ‘FTMUSDT’: 192.507, ‘SANDUSDT’: 203.219, ‘AXSUSDT’: 270.41

Currencies that are currently retreating above 80% of their highs:

ICPUSDT’: 0.022, ‘FILUSDT’: 0.043, ‘BAKEUSDT’: 0.046, ‘TLMUSDT’: 0.05, ‘LITUSDT’: 0.053, ‘LINAUSDT’: 0.054, ‘JASMYUSDT’: 0.056, ‘ALPHAUSDT’: 0.062, ‘RAYUSDT’: 0.062, ‘GRTUSDT’: 0.067, ‘DENTUSDT’: 0.068, ‘RSRUSDT’: 0.068, ‘XEMUSDT’: 0.068, ‘UNFIUSDT’: 0.072, ‘DYDXUSDT’: 0.074, ‘SUSHIUSDT’: 0.074, ‘OGNUSDT’: 0.074, ‘COMPUSDT’: 0.074, ‘NKNUSDT’: 0.078, ‘SKLUSDT’: 0.08, ‘DGBUSDT’: 0.081, ‘RLCUSDT’: 0.085, ‘REEFUSDT’: 0.086, ‘BANDUSDT’: 0.086, ‘HOTUSDT’: 0.092, ‘SRMUSDT’: 0.092, ‘RENUSDT’: 0.092, ‘BTSUSDT’: 0.093, ‘THETAUSDT’: 0.094, ‘FLMUSDT’: 0.094, ‘EOSUSDT’: 0.095, ‘TRBUSDT’: 0.095, ‘SXPUSDT’: 0.095, ‘ATAUSDT’: 0.096, ‘NEOUSDT’: 0.096, ‘FLOWUSDT’: 0.097, ‘YFIUSDT’: 0.101, ‘BALUSDT’: 0.106, ‘MASKUSDT’: 0.106, ‘ONTUSDT’: 0.108, ‘CELRUSDT’: 0.108, ‘AUDIOUSDT’: 0.108, ‘SCUSDT’: 0.11, ‘GALAUSDT’: 0.113, ‘GTCUSDT’: 0.117, ‘CTSIUSDT’: 0.117, ‘STMXUSDT’: 0.118, ‘DARUSDT’: 0.118, ‘ALICEUSDT’: 0.119, ‘SNXUSDT’: 0.124, ‘FTMUSDT’: 0.126, ‘BCHUSDT’: 0.127, ‘SFPUSDT’: 0.127, ‘ROSEUSDT’: 0.128, ‘DOGEUSDT’: 0.128, ‘RVNUSDT’: 0.129, ‘OCEANUSDT’: 0.129, ‘VETUSDT’: 0.13, ‘KSMUSDT’: 0.131, ‘ICXUSDT’: 0.131, ‘UNIUSDT’: 0.131, ‘ONEUSDT’: 0.131, ‘1INCHUSDT’: 0.134, ‘IOTAUSDT’: 0.139, ‘C98USDT’: 0.139, ‘WAVESUSDT’: 0.14, ‘DUSKUSDT’: 0.141, ‘LINKUSDT’: 0.143, ‘DASHUSDT’: 0.143, ‘OMGUSDT’: 0.143, ‘PEOPLEUSDT’: 0.143, ‘AXSUSDT’: 0.15, ‘ENJUSDT’: 0.15, ‘QTUMUSDT’: 0.152, ‘SHIBUSDT’: 0.154, ‘ZENUSDT’: 0.154, ‘BLZUSDT’: 0.154, ‘ANTUSDT’: 0.155, ‘XECUSDT’: 0.155, ‘CHZUSDT’: 0.158, ‘RUNEUSDT’: 0.163, ‘ENSUSDT’: 0.165, ‘LRCUSDT’: 0.167, ‘CHRUSDT’: 0.168, ‘IOTXUSDT’: 0.174, ‘TOMOUSDT’: 0.176, ‘ALGOUSDT’: 0.177, ‘EGLDUSDT’: 0.177, ‘ARUSDT’: 0.178, ‘LTCUSDT’: 0.178, ‘HNTUSDT’: 0.18, ‘LPTUSDT’: 0.181, ‘SOLUSDT’: 0.183, ‘ARPAUSDT’: 0.184, ‘BELUSDT’: 0.184, ‘ETCUSDT’: 0.186, ‘ZRXUSDT’: 0.187, ‘AAVEUSDT’: 0.187, ‘CVCUSDT’: 0.188, ‘STORJUSDT’: 0.189, ‘COTIUSDT’: 0.19, ‘CELOUSDT’: 0.191, ‘SANDUSDT’: 0.191, ‘ADAUSDT’: 0.192, ‘HBARUSDT’: 0.194, ‘DOTUSDT’: 0.195, ‘XLMUSDT’: 0.195

#下载所有交易对的收盘价
start_date = '2021-1-1'
end_date = '2022-05-30'
period = '1d'
df_all = pd.DataFrame(index=pd.date_range(start=start_date, end=end_date, freq=period),columns=symbols)
for i in range(len(symbols)):
    #print(symbols[i])
    symbol = symbols[i]
    df_s = GetKlines(symbol=symbol,start=start_date,end=end_date,period=period,base='api',v='v3')
    df_all[symbol] = df_s[~df_s.index.duplicated(keep='first')].close

#指数变化
df_norm = df_all/df_all.fillna(method='bfill').iloc[0] #归一化
df_norm.mean(axis=1).plot(figsize=(15,6),grid=True);

#比年初的最高涨幅
max_up = df_all.max()/df_all.fillna(method='bfill').iloc[0]
print(max_up.map(lambda x:round(x,3)).sort_values().to_dict())

{'JASMYUSDT': 1.0, 'ICPUSDT': 1.0, 'LINAUSDT': 1.0, 'WOOUSDT': 1.0, 'GALUSDT': 1.0, 'PEOPLEUSDT': 1.0, 'XECUSDT': 1.026, 'ENSUSDT': 1.032, 'TLMUSDT': 1.039, 'IMXUSDT': 1.099, 'FLOWUSDT': 1.155, 'ATAUSDT': 1.216, 'DARUSDT': 1.261, 'ALICEUSDT': 1.312, 'BNXUSDT': 1.522, 'API3USDT': 1.732, 'GTCUSDT': 1.833, 'KLAYUSDT': 1.891, 'BAKEUSDT': 1.892, 'DYDXUSDT': 2.062, 'SHIBUSDT': 2.281, 'BTCUSDT': 2.302, 'MASKUSDT': 2.396, 'SFPUSDT': 2.74, 'LPTUSDT': 2.75, 'APEUSDT': 2.783, 'ARUSDT': 2.928, 'CELOUSDT': 2.951, 'ZILUSDT': 2.999, 'LTCUSDT': 3.072, 'SNXUSDT': 3.266, 'XEMUSDT': 3.555, 'XMRUSDT': 3.564, 'YFIUSDT': 3.794, 'BANDUSDT': 3.812, 'RAYUSDT': 3.924, 'REEFUSDT': 4.184, 'ANTUSDT': 4.205, 'XTZUSDT': 4.339, 'CTKUSDT': 4.352, 'LITUSDT': 4.38, 'RSRUSDT': 4.407, 'LINKUSDT': 4.412, 'BCHUSDT': 4.527, 'DASHUSDT': 5.037, 'BALUSDT': 5.172, 'OCEANUSDT': 5.277, 'EOSUSDT': 5.503, 'RENUSDT': 5.538, 'XLMUSDT': 5.563, 'TOMOUSDT': 5.567, 'ZECUSDT': 5.654, 'COMPUSDT': 5.87, 'DGBUSDT': 5.948, 'ALGOUSDT': 5.981, 'ONTUSDT': 5.997, 'BELUSDT': 6.101, 'TRXUSDT': 6.116, 'ZRXUSDT': 6.135, 'GRTUSDT': 6.45, '1INCHUSDT': 6.479, 'DOTUSDT': 6.502, 'ETHUSDT': 6.596, 'KAVAUSDT': 6.687, 'ICXUSDT': 6.74, 'SUSHIUSDT': 6.848, 'AAVEUSDT': 6.931, 'BTSUSDT': 6.961, 'KNCUSDT': 6.966, 'C98USDT': 7.091, 'THETAUSDT': 7.222, 'ATOMUSDT': 7.553, 'OMGUSDT': 7.556, 'SXPUSDT': 7.681, 'UNFIUSDT': 7.696, 'XRPUSDT': 7.726, 'TRBUSDT': 8.241, 'BLZUSDT': 8.434, 'NEOUSDT': 8.491, 'FLMUSDT': 8.506, 'KSMUSDT': 8.571, 'FILUSDT': 8.591, 'IOTAUSDT': 8.616, 'BATUSDT': 8.647, 'ARPAUSDT': 9.055, 'UNIUSDT': 9.104, 'WAVESUSDT': 9.106, 'MKRUSDT': 10.294, 'CRVUSDT': 10.513, 'STORJUSDT': 10.674, 'SKLUSDT': 11.009, 'CVCUSDT': 11.026, 'SRMUSDT': 11.031, 'QTUMUSDT': 12.066, 'ALPHAUSDT': 12.103, 'ZENUSDT': 12.631, 'VETUSDT': 13.296, 'ROSEUSDT': 13.429, 'FTTUSDT': 13.705, 'IOSTUSDT': 13.786, 'COTIUSDT': 13.958, 'NEARUSDT': 14.855, 'HBARUSDT': 15.312, 'RLCUSDT': 15.432, 'SCUSDT': 15.6, 'GALAUSDT': 15.722, 'RUNEUSDT': 15.795, 'ADAUSDT': 16.94, 'MTLUSDT': 17.18, 'BNBUSDT': 17.899, 'RVNUSDT': 18.169, 'EGLDUSDT': 18.879, 'LRCUSDT': 19.499, 'ANKRUSDT': 21.398, 'ETCUSDT': 23.51, 'DUSKUSDT': 23.55, 'AUDIOUSDT': 25.306, 'OGNUSDT': 25.524, 'GMTUSDT': 28.83, 'ENJUSDT': 33.073, 'STMXUSDT': 33.18, 'IOTXUSDT': 35.866, 'AVAXUSDT': 36.946, 'CHZUSDT': 37.128, 'CELRUSDT': 37.273, 'HNTUSDT': 38.779, 'CTSIUSDT': 41.108, 'HOTUSDT': 46.466, 'CHRUSDT': 61.091, 'MANAUSDT': 62.143, 'NKNUSDT': 70.636, 'ONEUSDT': 84.132, 'DENTUSDT': 99.973, 'DOGEUSDT': 121.447, 'SOLUSDT': 140.296, 'MATICUSDT': 161.846, 'FTMUSDT': 192.507, 'SANDUSDT': 203.219, 'AXSUSDT': 270.41}

#当前最大回测
draw_down = df_all.iloc[-1]/df_all.max()
print(draw_down.map(lambda x:round(x,3)).sort_values().to_dict())

{'ICPUSDT': 0.022, 'FILUSDT': 0.043, 'BAKEUSDT': 0.046, 'TLMUSDT': 0.05, 'LITUSDT': 0.053, 'LINAUSDT': 0.054, 'JASMYUSDT': 0.056, 'ALPHAUSDT': 0.062, 'RAYUSDT': 0.062, 'GRTUSDT': 0.067, 'DENTUSDT': 0.068, 'RSRUSDT': 0.068, 'XEMUSDT': 0.068, 'UNFIUSDT': 0.072, 'DYDXUSDT': 0.074, 'SUSHIUSDT': 0.074, 'OGNUSDT': 0.074, 'COMPUSDT': 0.074, 'NKNUSDT': 0.078, 'SKLUSDT': 0.08, 'DGBUSDT': 0.081, 'RLCUSDT': 0.085, 'REEFUSDT': 0.086, 'BANDUSDT': 0.086, 'HOTUSDT': 0.092, 'SRMUSDT': 0.092, 'RENUSDT': 0.092, 'BTSUSDT': 0.093, 'THETAUSDT': 0.094, 'FLMUSDT': 0.094, 'EOSUSDT': 0.095, 'TRBUSDT': 0.095, 'SXPUSDT': 0.095, 'ATAUSDT': 0.096, 'NEOUSDT': 0.096, 'FLOWUSDT': 0.097, 'YFIUSDT': 0.101, 'BALUSDT': 0.106, 'MASKUSDT': 0.106, 'ONTUSDT': 0.108, 'CELRUSDT': 0.108, 'AUDIOUSDT': 0.108, 'SCUSDT': 0.11, 'GALAUSDT': 0.113, 'GTCUSDT': 0.117, 'CTSIUSDT': 0.117, 'STMXUSDT': 0.118, 'DARUSDT': 0.118, 'ALICEUSDT': 0.119, 'SNXUSDT': 0.124, 'FTMUSDT': 0.126, 'BCHUSDT': 0.127, 'SFPUSDT': 0.127, 'ROSEUSDT': 0.128, 'DOGEUSDT': 0.128, 'RVNUSDT': 0.129, 'OCEANUSDT': 0.129, 'VETUSDT': 0.13, 'KSMUSDT': 0.131, 'ICXUSDT': 0.131, 'UNIUSDT': 0.131, 'ONEUSDT': 0.131, '1INCHUSDT': 0.134, 'IOTAUSDT': 0.139, 'C98USDT': 0.139, 'WAVESUSDT': 0.14, 'DUSKUSDT': 0.141, 'LINKUSDT': 0.143, 'DASHUSDT': 0.143, 'OMGUSDT': 0.143, 'PEOPLEUSDT': 0.143, 'AXSUSDT': 0.15, 'ENJUSDT': 0.15, 'QTUMUSDT': 0.152, 'SHIBUSDT': 0.154, 'ZENUSDT': 0.154, 'BLZUSDT': 0.154, 'ANTUSDT': 0.155, 'XECUSDT': 0.155, 'CHZUSDT': 0.158, 'RUNEUSDT': 0.163, 'ENSUSDT': 0.165, 'LRCUSDT': 0.167, 'CHRUSDT': 0.168, 'IOTXUSDT': 0.174, 'TOMOUSDT': 0.176, 'ALGOUSDT': 0.177, 'EGLDUSDT': 0.177, 'ARUSDT': 0.178, 'LTCUSDT': 0.178, 'HNTUSDT': 0.18, 'LPTUSDT': 0.181, 'SOLUSDT': 0.183, 'ARPAUSDT': 0.184, 'BELUSDT': 0.184, 'ETCUSDT': 0.186, 'ZRXUSDT': 0.187, 'AAVEUSDT': 0.187, 'CVCUSDT': 0.188, 'STORJUSDT': 0.189, 'COTIUSDT': 0.19, 'CELOUSDT': 0.191, 'SANDUSDT': 0.191, 'ADAUSDT': 0.192, 'HBARUSDT': 0.194, 'DOTUSDT': 0.195, 'XLMUSDT': 0.195, 'AVAXUSDT': 0.206, 'ANKRUSDT': 0.207, 'MTLUSDT': 0.208, 'MANAUSDT': 0.209, 'CRVUSDT': 0.213, 'API3USDT': 0.221, 'IOSTUSDT': 0.227, 'XRPUSDT': 0.228, 'BATUSDT': 0.228, 'MKRUSDT': 0.229, 'MATICUSDT': 0.229, 'CTKUSDT': 0.233, 'ZILUSDT': 0.233, 'WOOUSDT': 0.234, 'ATOMUSDT': 0.237, 'KLAYUSDT': 0.239, 'XTZUSDT': 0.245, 'IMXUSDT': 0.278, 'NEARUSDT': 0.285, 'GALUSDT': 0.299, 'APEUSDT': 0.305, 'ZECUSDT': 0.309, 'KAVAUSDT': 0.31, 'GMTUSDT': 0.327, 'FTTUSDT': 0.366, 'KNCUSDT': 0.401, 'ETHUSDT': 0.416, 'XMRUSDT': 0.422, 'BTCUSDT': 0.47, 'BNBUSDT': 0.476, 'TRXUSDT': 0.507, 'BNXUSDT': 0.64}

First, we simulate the next downward trend using the simplest code and look at the price of the different holdings. Since the strategy always holds multiple positions, there is no risk of a rise. The initial capital is 1000, the coin price is 1, the adjustment ratio is 0.01. The result is as follows. It can be seen that the risk of doing multiple positions is not low, with 1.5 times leverage, it can resist a 50% drop.

for Hold_value in [300,500,800,1000,1500,2000,3000,5000,10000]:
    amount = Hold_value/1
    hold_price = 1
    margin = 1000
    Pct = 0.01
    i = 0
    while margin > 0:
        i += 1
        if i>500:
            break
        buy_price = (1-Pct)*Hold_value/amount
        buy_amount = Hold_value*Pct/buy_price
        hold_price = (amount * hold_price + buy_amount * buy_price) / (buy_amount + amount)
        amount += buy_amount
        margin = 1000 + amount * (buy_price - hold_price)
    print(Hold_value, round(buy_price,3))

300 0.035
500 0.133
800 0.285
1000 0.362
1500 0.51
2000 0.599
3000 0.711
5000 0.81
10000 0.904

#还是用原来的回测引擎
class Exchange:
    
    def __init__(self, trade_symbols, fee=0.0004, initial_balance=10000):
        self.initial_balance = initial_balance #初始的资产
        self.fee = fee
        self.trade_symbols = trade_symbols
        self.account = {'USDT':{'realised_profit':0, 'unrealised_profit':0, 'total':initial_balance, 'fee':0}}
        for symbol in trade_symbols:
            self.account[symbol] = {'amount':0, 'hold_price':0, 'value':0, 'price':0, 'realised_profit':0,'unrealised_profit':0,'fee':0}
            
    def Trade(self, symbol, direction, price, amount):
        
        cover_amount = 0 if direction*self.account[symbol]['amount'] >=0 else min(abs(self.account[symbol]['amount']), amount)
        open_amount = amount - cover_amount
        self.account['USDT']['realised_profit'] -= price*amount*self.fee #扣除手续费
        self.account['USDT']['fee'] += price*amount*self.fee
        self.account[symbol]['fee'] += price*amount*self.fee

        if cover_amount > 0: #先平仓
            self.account['USDT']['realised_profit'] += -direction*(price - self.account[symbol]['hold_price'])*cover_amount  #利润
            self.account[symbol]['realised_profit'] += -direction*(price - self.account[symbol]['hold_price'])*cover_amount
            
            self.account[symbol]['amount'] -= -direction*cover_amount
            self.account[symbol]['hold_price'] = 0 if self.account[symbol]['amount'] == 0 else self.account[symbol]['hold_price']
            
        if open_amount > 0:
            total_cost = self.account[symbol]['hold_price']*direction*self.account[symbol]['amount'] + price*open_amount
            total_amount = direction*self.account[symbol]['amount']+open_amount
            
            self.account[symbol]['hold_price'] = total_cost/total_amount
            self.account[symbol]['amount'] += direction*open_amount
                    
    
    def Buy(self, symbol, price, amount):
        self.Trade(symbol, 1, price, amount)
        
    def Sell(self, symbol, price, amount):
        self.Trade(symbol, -1, price, amount)
        
    def Update(self, close_price): #对资产进行更新
        self.account['USDT']['unrealised_profit'] = 0
        for symbol in self.trade_symbols:
            self.account[symbol]['unrealised_profit'] = (close_price[symbol] - self.account[symbol]['hold_price'])*self.account[symbol]['amount']
            self.account[symbol]['price'] = close_price[symbol]
            self.account[symbol]['value'] = abs(self.account[symbol]['amount'])*close_price[symbol]
            self.account['USDT']['unrealised_profit'] += self.account[symbol]['unrealised_profit']
        self.account['USDT']['total'] = round(self.account['USDT']['realised_profit'] + self.initial_balance + self.account['USDT']['unrealised_profit'],6)

First, we look back at the performance of the TRX balance strategy, the TRX's maximum pullback in the bear market in this round is relatively small, so there is a certain peculiarity. The data selected the 5minK line so far in 2021, with an initial capital of 1000, adjusted ratio of 0.01, holding value of 2000, processing fee of 0.0002.

The TRX was initially priced at 0.02676U, then peaked at 0.18U, and is currently at around 0.08U.

The final return of 4524U, which is already very close to the TRX's gain at 0.18, the leverage is less than 2 times from the beginning to the end below 0.4, and the probability of a breakout is decreasing, during which there is an opportunity to increase the value of the holding. However, the income below 2000U is always static. This is also one of the disadvantages of the balancing strategy.

symbol = 'TRXUSDT'
df_trx = GetKlines(symbol=symbol,start='2021-1-1',end='2022-5-30',period='5m')

df_trx.close.plot(figsize=(15,6),grid=True);

#TRX平衡策略回测
hold_value = 2000
pct = 0.01
e = Exchange([symbol], fee=0.0002, initial_balance=1000)
init_price =  df_trx.iloc[0].open
res_list = [] #用于储存中间结果
e.Buy(symbol,init_price,hold_value/init_price)
e.Update({symbol:init_price})
for row in df_trx.itertuples():
    buy_price = (1-pct)*hold_value/e.account[symbol]['amount']
    sell_price = (1+pct)*hold_value/e.account[symbol]['amount']
    
    while row.low < buy_price:
        e.Buy(symbol,buy_price,pct*hold_value/buy_price)
        e.Update({symbol:row.close})
        buy_price = (1-pct)*hold_value/e.account[symbol]['amount']
        sell_price = (1+pct)*hold_value/e.account[symbol]['amount']
    while row.high > sell_price:
        e.Sell(symbol,sell_price,pct*hold_value/sell_price)
        e.Update({symbol:row.close})
        buy_price = (1-pct)*hold_value/e.account[symbol]['amount']
        sell_price = (1+pct)*hold_value/e.account[symbol]['amount']
    if int(row.time)%(60*60*1000) == 0:
        e.Update({symbol:row.close})
        res_list.append([row.time, row.close, e.account[symbol]['amount'],e.account[symbol]['amount']*row.close, e.account['USDT']['total']-e.initial_balance])
res_trx = pd.DataFrame(data=res_list, columns=['time','price','amount','value','profit'])
res_trx.index = pd.to_datetime(res_trx.time,unit='ms')
print(pct,e.account['USDT']['realised_profit']+e.account['USDT']['unrealised_profit'] ,round(e.account['USDT']['fee'],0))

0.01 4524.226998288555 91.0

#收益
res_trx.profit.plot(figsize=(15,6),grid=True);

#实际占用杠杆
(res_trx.value/(res_trx.profit+1000)).plot(figsize=(15,6),grid=True);

Let's take a look back at WAVES again, this coin is quite special, it rose from a high of 6U to 60U at the beginning, and finally fell back to the current 8U. The final gain was 4945, far more than the gain of holding the coin.

symbol = 'WAVESUSDT'
df_waves = GetKlines(symbol=symbol,start='2021-1-1',end='2022-5-30',period='5m')

df_waves.close.plot(figsize=(15,6),grid=True);

#TWAVES平衡策略回测
hold_value = 2000
pct = 0.01
e = Exchange([symbol], fee=0.0002, initial_balance=1000)
init_price =  df_waves.iloc[0].open
res_list = [] #用于储存中间结果
e.Buy(symbol,init_price,hold_value/init_price)
e.Update({symbol:init_price})
for row in df_waves.itertuples():
    buy_price = (1-pct)*hold_value/e.account[symbol]['amount']
    sell_price = (1+pct)*hold_value/e.account[symbol]['amount']
    
    while row.low < buy_price:
        e.Buy(symbol,buy_price,pct*hold_value/buy_price)
        e.Update({symbol:row.close})
        buy_price = (1-pct)*hold_value/e.account[symbol]['amount']
        sell_price = (1+pct)*hold_value/e.account[symbol]['amount']
    while row.high > sell_price:
        e.Sell(symbol,sell_price,pct*hold_value/sell_price)
        e.Update({symbol:row.close})
        buy_price = (1-pct)*hold_value/e.account[symbol]['amount']
        sell_price = (1+pct)*hold_value/e.account[symbol]['amount']
    if int(row.time)%(60*60*1000) == 0:
        e.Update({symbol:row.close})
        res_list.append([row.time, row.close, e.account[symbol]['amount'],e.account[symbol]['amount']*row.close, e.account['USDT']['total']-e.initial_balance])
res_waves = pd.DataFrame(data=res_list, columns=['time','price','amount','value','profit'])
res_waves.index = pd.to_datetime(res_waves.time,unit='ms')
print(pct,e.account['USDT']['realised_profit']+e.account['USDT']['unrealised_profit'] ,round(e.account['USDT']['fee'],0))

0.01 4945.149323437233 178.0

df_waves.profit.plot(figsize=(15,6),grid=True);

By the way, the performance of the grid strategy is measured backwards, the grid interval is 0.01, and the grid value is 10. WAVES and TRX both experienced huge pullbacks, with WAVES pulling back 5000U and TRX exceeding 3000U, which would basically blow up if the initial capital was less.

#网格策略
pct = 0.01
value = 10*pct/0.01
e = Exchange([symbol], fee=0.0002, initial_balance=1000)
init_price =  df_waves.iloc[0].open
res_list = [] #用于储存中间结果
for row in df_waves.itertuples():
    buy_price = (value / pct - value) / (value / (pct * init_price) + e.account[symbol]['amount']) 
    sell_price = (value / pct + value) / (value / (pct *init_price) + e.account[symbol]['amount'])

    while row.low < buy_price:
        e.Buy(symbol,buy_price,value/buy_price)
        e.Update({symbol:row.close})
        buy_price = (value / pct - value) / (value / (pct * init_price) + e.account[symbol]['amount']) #买单价格，由于是挂单成交，也是最终的撮合价格=
    while row.high > sell_price:
        e.Sell(symbol,sell_price,value/sell_price)
        e.Update({symbol:row.close})
        sell_price = (value / pct + value) / (value / (pct *init_price) + e.account[symbol]['amount'])
    if int(row.time)%(60*60*1000) == 0:
        e.Update({symbol:row.close})
        res_list.append([row.time, row.close, e.account[symbol]['amount'],e.account[symbol]['amount']*row.close, e.account['USDT']['total']-e.initial_balance])
res_waves_net = pd.DataFrame(data=res_list, columns=['time','price','amount','value','profit'])
res_waves_net.index = pd.to_datetime(res_waves_net.time,unit='ms')
print(pct,e.account['USDT']['realised_profit']+e.account['USDT']['unrealised_profit'] ,round(e.account['USDT']['fee'],0))

0.01 1678.0516101975015 70.0

res_waves_net.profit.plot(figsize=(15,6),grid=True);

#网格策略
pct = 0.01
value = 10*pct/0.01
e = Exchange([symbol], fee=0.0002, initial_balance=1000)
init_price =  df_trx.iloc[0].open
res_list = [] #用于储存中间结果
for row in df_trx.itertuples():
    buy_price = (value / pct - value) / (value / (pct * init_price) + e.account[symbol]['amount']) 
    sell_price = (value / pct + value) / (value / (pct *init_price) + e.account[symbol]['amount'])

    while row.low < buy_price:
        e.Buy(symbol,buy_price,value/buy_price)
        e.Update({symbol:row.close})
        buy_price = (value / pct - value) / (value / (pct * init_price) + e.account[symbol]['amount']) 
    while row.high > sell_price:
        e.Sell(symbol,sell_price,value/sell_price)
        e.Update({symbol:row.close})
        sell_price = (value / pct + value) / (value / (pct *init_price) + e.account[symbol]['amount'])
    if int(row.time)%(60*60*1000) == 0:
        e.Update({symbol:row.close})
        res_list.append([row.time, row.close, e.account[symbol]['amount'],e.account[symbol]['amount']*row.close, e.account['USDT']['total']-e.initial_balance])
res_trx_net = pd.DataFrame(data=res_list, columns=['time','price','amount','value','profit'])
res_trx_net.index = pd.to_datetime(res_trx_net.time,unit='ms')
print(pct,e.account['USDT']['realised_profit']+e.account['USDT']['unrealised_profit'] ,round(e.account['USDT']['fee'],0))

0.01 -161.06952570521656 37.0

res_trx_net.profit.plot(figsize=(15,6),grid=True);

Summary

This retrospective analysis uses a 5minK line, the middle fluctuations are not fully simulated, so the actual returns should be slightly higher. Overall, the balance strategy is relatively low in risk, not afraid of a downturn, does not need to adjust parameters, is relatively convenient to use, suitable for novice users. The grid strategy is very sensitive to the initial price setting, requires a certain judgment of the market situation, long-term view, the risk of doing a gap is high. This year's Battle of the Bands will offer free use of the Sustainable Balance Strategy, and everyone is welcome to experience it.