HODL+ your crypto with Grid Bot on Binance. Make gains when the price fluctuates within a certain range

Edward Tsang
13 min readAug 21, 2022

(Updated 2023/12/23) Just sharing notes on my recent experiments with tradebot for crypto trading. Been running these bots since March 2023. The bot did over 16k trades for me and made around 3.5k profit on 28k capital, not too bad. It is what I wanted for crypto winter (horizontal market). The hardest part is still to pick the right crypto pair to use the bot with, and then “guess” the range that it will fall within. My bots been out of the range from time to time, but I got rather good at stopping them and start another bot.

Read the disclaimer at the bottom (if you really need to). This is not investment advice. And I might make money from affiliate links, etc, etc. If you are considering crypto bots and don’t know these by now, please don’t read on. :) DYOR (Do Your Own Research) is important for crypto. Be warned.

** Year End Summary. I have been running the HODL+ Grid Bots since April 2023. The bots did a total of about 23700 tranactions and made a net profit of about 23% in the 8 months. **

I have added a HODL++ article, you should read it after this one. It extends on the grid trading bot with a rebalancing bot to improve on the results.

So, let’s set the scene. I am not a full time trader, but I have been trading crypto for a few years. Buying and selling. I do/did crazy things with my crypto. I treat them just as funny money. I don’t put much cash into them, I have been lucky to earn most of my crypto play money from work. But lost most of them with trading, or just betting, on highly risky tokens. You name all the high risk stuff, I have tried them. Leveraged tokens, DeFi, LUCA/USTC, etc, etc. I have also lived through the first 2 gen of exchanges crashes/vanishes, MtGox and BTCChina; 3rd Gen now dominated by Binance, and I still haven’t learnt to de-risk by trading on multiple exchanges.

My recent attempt is to go back to basics. Simple trade bots with simple maths and logic. Grid Trading Strategies on Binance, a free feature on binance.com which is where I usually trade on, your favourite exchange should have it too. I think the crypto winter or just slow market will be here for a while, so this kind of automated trading during such horizontal market might be useful to pick up a few points here and there. Longer term, I am still positive about the crypto market, but mainly on top coins/tokens. Check out my other articles on my view about the wider crypto market and crypto trading. So I use the grid bot to make a little from what I am hodling and put the profit back into buying more of the crypto I want, hence, HODL+.

To start with, you should read this, it’s probably the most comprehensive guide on crypto grid trading on Medium. It’s not for Binance but for Pionex, you need about half an hour to read it, good for you to understand the logic and maths behind it. The formulas specifically for Binance Grid Bot is here, make sure you take a look too. You will see that you can’t really “lose” your crypto with grid trading as such, it can only do “buy low, sell high”, as long as you are within the range, it will keep trading for you to earn a little more. But if you are out of the range, you will either be left with all crypto or all stablecoin, and the bot will stop making any trades.

Background of Grid Trading and why is it useful (or not) for crypto winter

  • Grid Trading is not new, been around for a while and mostly for foreign exchange (forex or fx) trading. Crypto trading kind of re-discover this, forgetting that crypto was supposed to be currencies like to start with.
  • Sideways Market which Grid Trading is good for, is “exactly” what FX market should be. It’s not common to expect 10% up or down on national currencies as normal daily events.
  • Grid trading is mainly about profiting from fluctuations and not about compounding to gain more currencies on hand. That’s why automated compounding is not part of the maths and risk design.
  • These automated trading tools weren’t available to non institutional investors, not just on the client end, major forex exchanges didn’t have consumer/retail facing APIs for automations.

My Goals — Why I spent time playing around with this?

  • I have been looking out for automated trade bots for a while. There are many tradebot services which can control your account via APIs. And even the Binance one is not that new, been around since early 2022. But just got around to evaluating it properly. Seems to be working well, so I am sharing the notes now.
  • You just can’t trade 24 hours a day. You can’t actively monitor the market all hours and make decisions on every trade. You will need to set certain rules and strategies for it. And if you already have an idea on what these rules and formulas are, you might as well automate them so they are repeatable.
  • I was avoiding having to code and host it myself. Coding will likely to introduce my own bugs and errors, end up spending a lot of time fixing it, on something that will only earn me a little. Hosting 24/7 app is not fun, even with cloud services, it is hard to ensure that the system runs all hours, or 99.999% uptime (the five 9s SLA which I am used to building tech solutions for professionally).
  • Third party trade bots have many advanced features, but with crypto, I just can’t easily trust them with my API keys, even just for non-withdrawal keys. And I don’t want to pay them monthly fees, so I am trying to make the most out of the free feature from Binance where I do most of my trading anyway. Use the native Grid Bot feature on your favourite exchange to be “sure” that there are no external factors for downtime, miscalculations, etc. The exchanges just want to get more trade volume and make more from fees, simple.
  • I am just lazy. Want to set something up and leave it running. Only alert me when it is about to get out of range (mainly above the range, which is a good problem to have). There are system alerts, but you can also set one yourself at the last buy/sell order of the range.

Here is a short list of what I learnt so far:

  • The “new” trading bot (Since October 2023 or so) feature on Binance is a bit annoying. They are trying to split the bot trading into another “wallet” rather than sharing with SPOT. Not against the idea as such, but now you can’t easily take the profit out of your grid trading bot without stopping/ending the grid bot. Previously, all profit just goes back into your SPOT wallet directly. And the fees calculation is also annoying now, not clear when they will deduct the fees, so you need to estimate and have BNB “upfront”. The money is still good, but rather annoying to have to keep starting and stopping bots, more fees, etc. I wish Binance can fix it.
  • After running the “new” grid trading bots for a while, I am starting to use the following “strategy” to do my own compounding. Basically, you need to get very good at stopping and starting the bots, remembering that each stop and start will cost you fees, up to 2 x 0.1%. If I have any grid profit of 3% and positive total profit, I compound it. Or after 30 days or so of running a bit, if it nets over ~2% and positive total profit, I compound it. To compound it, I just end the bot and manually handle the tokens, sell all the tokens at the current/high price, then setup a new bot, with similar/current range, put all the money I got from the ended bot and sold tokens to the new bot. Aim at a -0.5% — -1% or so for the trigger price so I will even cover for the fees.
  • Binance been pushing for FDUSD and TUSD with zero trading fees. It’s very good for grid trading bot, you don’t have to worry about fees, so you can just max out the number of grids. FDUSD is getting most of the attention now, their trading volume is much higher than TUSD, you need good trade volume for the bot.
  • Don’t use Grid Bot on a token pair that you wouldn’t have HODL anyway. It’s not for short term gains. It is hard to outperform HODL for short pumps and dumps. Talking about 3-6 months or more. If you are confident about the moon schedule, then you should look at Moonbot on some exchanges rather than Gridbot only on Binance. (I am deliberately avoiding Futures Grid for now, or anything that might set me back to zero overnight.)
  • In addition to the above, there is now a Moonbot-like feature on Grid bot called “Trailing Up”. Basically, it works by moving the grid higher as it approaches the top range and removing from the bottom, shifting the whole grid upwards. It is a very useful feature which can make more money from a trade pair growing above your inital range. But be careful about setting too small a range, the range should still be bigger than your expected spike, otherwise it can push the grid very high, but leave it out of range as it comes back down. (It only adjusts upwards, not downwards.)
  • My lazy way to pick the range now, including using the Trailing Up, is to look at the 30 days auto setup values as reference for the lower range, and use the 7 days as reference the top range. Then pick a suitable number of grids to get a sensible Profit Range that is sensible to the fees rate (in my case, I aim at 0.3%-0.8% Profit Range).
  • No, this is not a get rich quick strategy. Nor is it the only strategy you need to know for trading crypto. But it is a very useful one during a boring horizontal market. When you expect the price to be within a certain range and keep fluctuating within that range. If you use it right, you might be able to add 15–30% in a year.
  • Reminder: Make sure you buy some BNB to reduce the fees by 25%. (See below, over 1000 trades in 80 days, 66/88 Grids, that means I have traded about 12-15x my “full” investment in 7 days or so. Good to save on fees.) As you will be doing many many trades. If you signed up to Binance without using any referral codes, you are missing out on trading fees kickbacks. But these are only available to “new” accounts, your ID can only be used for KYC with 1 account. Deleting (not just suspending) your account can done via the mobile app easily, just follow the instructions on this page, but click the Delete Account rather than Disable Account open. But this can be very annoying as you lose all previous trading records and you need to transfer all crypto out, etc. But then again, you will get to save an extra 10% on trading fees forever, if you use it together with paying for fees with BNB, you get to save a total of 35% on fees, which is significant if you trade often. (Use my code to get your 10% fees kickback. You will see the 10% icon when you sign up. I get referral bonus from Binance, you will get 10% of all your trading fees forever as a kickback.)
  • It only works well for volatile pairs. So you need to check the trade volume and range carefully. Most of the volatile ones are still USDT based, followed by FDUSD. Grid strategy can be seen as not very “efficient” in using your capital. You are spreading it very thin across many levels and in small amounts. The bot needs to make many transactions to make the same money compared to you holding 100% into a single crypto and gain a few points.
  • Do your own research. Take a look at what other people are trading with. There is a Strategy Leaderboard page on Binance. You can click around to study the most successful or most copied parameters. Or use the Auto feature to calculate the numbers for you, check the charts, copy it to manual settings and tweak it for yourself. Do your own experiments. Start with a few hundred bucks first.
  • I am sharing a screenshot of a moderately “successful” bot I was testing with. Not making money as SHIB is currently down a lot, but you can see that after over 1000 trades or so with 2 bots, it made around 5–6% average gain in around 80 days, I plan to leave this one running for a few months.
The bots I have been running for 80 days or so.
  • Basic maths, you will need the same number of Matched Trades as the Grid Number to achieve the Profit/Grid range for your whole investment. In example above, needed 88 Matched Trades for a 0.4% — 1.07% gain. It doesn’t seem like a lot of profit for so many trades, but it’s all automatic and relatively safe.
  • To set the number of Grids to use, make sure the per grid profit range is sensible. You can get many more trades with ramp it up to 150 grids, but the profit range might be too small after fees to generate sensible profit. I usually aim at around 0.3–0.4 as the average (mid point) of the profit range.
  • Take a look at the Auto values recommended by the bot. Decide if you want to run it for a long time or not. Can mix and match these recommendations. For example, if you don’t think the market is heading up, you can mix the 30D or 180D peak with the 7D bottom.
  • The starting price is important. Measured by the range you think it will fluctuate between and where you are within that range. Try not to start when you are far above the middle of that range. The bot will need to buy all the parts above your current level at the trigger price when you start the bot. And you will only start making money from the matched trades above your starting price as it will keep to the buy low sell high rule. Try to visualise what is happening, when you start the bot, everyone it will buy for all the grids above your starting price, so when it goes up, the bot will sell and make money. For the grids below your starting price, it will only buy if the fluctuate to cross that price line. So it will make money next time it goes higher than that grid.
  • Assuming that you are using it on a crypto to stablecoin pair. If it falls below your range, then you have it all in crypto. If you it goes above your range, then you have it all in stablecoin. Then you might have new decisions to make. Do you end the bot and start at a new range? Or are you confident that it will get back in the range soon? Try to use the Auto mode suggestions to confirm if the range are still valid for 7D/30D/180D predictions.
  • Get good at stopping and starting the new bots. Track the performance actively. If you messed up on the settings, don’t just hang on to it. Stop the bot, tweak the parameters, and start again. You will lose fees. But a well tuned bot can easily outperform and make the lost fees back.
  • Another way to look at this, is why I called it HODL+, is that when it falls below your range, you are actually hodling your full amount in crypto, so that is no worse that HODL. But if it exceeded your range at the top, you will be left with all stablecoin, but if your start/trigger price is mid/low in the range, you should have made some gains, together with the Grid Profit, you should be able to buy back more or your crypto if you don’t leave it for too long. Crypto mooning seems “quick”, but real moon, even the recent rise of BTC from sub-10k to 60k, took around 6 months or so. You have plenty of time to decide what to do if it goes above your stated range. (All on the assumption that you set the initial range properly.)
  • There is no auto compounding. But you can do that manually, every now and then when you have accumulated enough Grid Profit, you just end the bot and start again with same/similar settings with the new total investment. If you are using it for HODL+, then just use the Grid Profit to buy more of the crypto you are hodling.
  • There is no back-testing feature with Binance Bot, other platforms might have it.
  • Not recommending this, but I am using grid bots with SHIB/USDT and MATIC/USDT to get more USDT, relatively volatile, and every 10-20 USDT earned, I buy more ETH/BTC, depending which one is “lower” at the time. I also have another ETH/BTC pair going too, but for longer term, 180days+. Basically, I am hoping to get more ETH or BTC for HODL. Have your own strategy and pick the coins you want to collect and HODL.

Risks

  • Keeping crypto on a particular exchange for too long can be a risk. Binance is the biggest of 3rd gen exchanges, but there is no such thing as “too big to fail”.
  • Stablecoins are not that stable. Can’t remember LUNA/UST already? Even Tether or USDC might not be that stable. But I am a fan of Circle USDC because the guy behind it was a world class investment banker, but for trade volume, I still have to choose USDT.

Links to check out

Understanding Spot Grid Parameters
https://www.binance.com/en/support/faq/76bd4effa3c4456c971a1c6835762742
Upper & Lower Price
Upper Band = MA + BBM * Standard Deviation
Lower Band = MA — BBM * Standard Deviation
*What is BBM: https://academy.binance.com/en/articles/bollinger-bands-explained
Grid Number
Grid Number = (grid_upper_limit — grid_lower_limit)/ATR
*Average True Range (ATR) for the past pre-defined hours of the selected symbol

This is an early draft, I am still editing this. I will be updating this note every now and then. Please feel free to leave me with comments, etc.

Disclaimer

No financial advice, no warranty of any kind. This article is provided as is, just some random words on a page, not financial advice or any orders or instructions for you to do anything with it. Read my other articles if you want to know about my views on crypto. Yes, there might be affiliate links or ads on these which I make money from.

--

--

Edward Tsang

Experienced technologist, focused on selective combinations of blockchain and AI. I use Medium to repost what I share on https://evolvingviews.com.