Aitor

[Aitor]

I uploaded a very detailed video tutorial about Backtesting in Spanish. I want to create a new one in English. So far you can check the Spanish tutorial here:

[Aitor]

If you add +1 to DU_CAP_COUNT the bot will make a new DCA buy (as far as conditions are met and you have enough funds).  Just make sure you have saved changes and the new strategy / changes are being taken into account by the bot.

On another note, the fact that the price went lower than price it doesn’t mean that it has to buy. It won’t buy if:

1- Price reached to the point very fast and reverted fast before the bot ended the cycle and processed the pair.

2- Trailme started working and price reverted fast. Thanks to trailme you will get huge sells / profits, but sometimes it can avoid a sell to have.

3- Any other indicator is not met, and Gunbot needs all the indicators to meet at the same time, else it won’t buy.

[Aitor]

Thanks for sharing your results. v22.0.1 works fine for me, except the PNL issue. v23.0.2 fixes PNL. It doesn’t need to stop and start the bot to get updated PNL stats. Well… in fact it does, but v23.0.2 auto updates PNL every few hours, so it partially fixes PNL issue.

[Aitor]

Thank you for sharing your results. Note that PNL is not updated automatically. The lastest bot version fixes it. At least it updates PNL automatically every few hours. If you are running older versions, you need to click “Stop trading” button and then “Start trading” again. It’s the only way to check updated PNL stats when running older bot versions.

[Aitor]

Very good results. Still trading with TL 40 and 3 pairs?

[Aitor]

[quote quote=10295]I also got a very very nice profit since I started it 2 weeks ago[/quote]

Glad to hear that. Nice job :thumbup:

[quote quote=10295]I post my last results in my thread but did not get approved yet. [/quote]

Sorry, my bad. We had a lot of spam posts so I had to set some restrictive settings. Every post that includes links (pics are considered links) is on hold until I manually approve it. I don’t like it but sadly is the only way to fight against spammers. And there was several post sharing results that I approved today. Apologies for the delay. I want to set a new forum role for contibutors so you will be able to post your results and be insta-approved.

[quote quote=10295]I really like this strategy, one of the best strategies since I start GB back in 2017 By the way, when can we get V4?[/quote]

So far v4.2 improvements are only implemented to C5, C6 and DB1 (most likely I will change the names so we know if it’s v3.5 or 4.x). Anyway v4 it’s still under development and so far only Javier and me are testing it. I don’t want to distribute it until it’s fully tested.

By the way, in the previous post the profit the bot got is 450. Ok, the thing is that at the moment of taking the picture, the total profit wasn’t updated. Now it is, and the real profit the bot got on 1st August is 505$

[Aitor]

Finally USDT-ETH pair closed its cycle with nice profit: 439$

After 101 days running SG Mutation (v2, v3, v3.5 and v4) I got 3747$. The average has been 37$ / day.

The bot did a nice work lowering avergare bought prices for all pairs, and it managed to end cycles with profit despite all pairs dropped by 60-70% recently. Now all the pairs are cycling fast and nicely.

Latest days have been very profitable. Specially 26th July, when it closed a big USDT-UNI cycle, and yesterday, 1st August, when it closed USDT-ETH cycle (Started the cycle in 4400 and ended it selling with profit at 2599).

From now on I will start posting the results in a new post I will made for v4 version.

[Aitor]

You are welcome :thumbup:

[Aitor]

USDT-BNB and USDT-UNI pairs closed their cycles with nice profits:

As you can see, USDT-UNI pair was sold at a very good price. Thanks to TRAIL_ME it sold much higher than the original sell target.

In the pic above you can see the profits from 22 April 2021 to 26 July 2021. For some reason after updating to latest Gunbot version I lost the stats from 20 March to 22 April, so this is all I can show. Anyway you can check the profits from 20 March onwards on other posts.

603 trades made in 3 months (95 days).

Profit per pair. As you can see ETH has been the less profitable. I know ETH is working very good for many people, but in my case it’s “stuck” in a big cycle starting at 4400, so it made less trades than it should under normal conditions.

The pic above shows the daily profit. As you can see the last day was very profitable as it ended both BNB and UNI cycles.

ETH cycle is close to be closed too, and it will be a huge one :thumbup:

[Aitor]

How to create the telegram bot:
(I recommend to stop Gunbot until the Telegram bot is created, then you can start it again).

Now open your Telegram and follow next steps:

1. Send a message to @Botfather and type /start and then /newbot to write the name and nickname you want for your bot.
2. Send a message to @myidbot and type /start and then /getid to get the ChatID of your telegram account.
3. Go to your Gunbot GUI by typing localhost:5000 in your browser. Then go to Profile – Telegram, activate the Telegram Notifications, paste the Token that @Botfather gave you previously, in “Chat Id” and in “Admin Id” paste the number you got from @Myidbot. Now click Save.
4. Click on Start trading button, and start a chat on your telegram with the username you chose for your bot in point 1 and type /start and you will see some buttons. Go to Settings – Notifications and enable bot alerts.

[Aitor]

[quote quote=10253] I made some modification to make it more aggressive: Period: 5m SMA: 21 Gain: 0.2[/quote]

In my opinion GAIN: 0.2 is too low. It usually gets 1%, 2%… Anyway, did you disabled “Autogain”? If you didn’t the bot is running Autogain despite you set any value to GAIN. In other words, if you want GAIN parameter to be taken into account by the bot, you need to set AUTOGAIN to “false”.

[quote quote=10253]The strategy is great, but sometimes I feel like it’s not good at capturing the bottom of some moves. [/quote]

Capturing the bottom is very difficult. It’s easy to say “I would have bought here at the bottom” once you see the graph, but very hard to do before it happens, as no one has a crystal ball . Also note that DCA 1, 2 and 3 and quite fast, but from 4 onwards, the bot needs to increase the distance between buy points for obvious reasons.

[quote quote=10253]also it always buy from top of the moves, I know it’s not from the bot itself, the strategy is doing what it supposed to do. just thinking maybe this is because of changing the period to 5m and SMA to 21.[/quote]

This is what strategy does, and it makes to get very good profits when the market is going up, and many times, when price is going up and down in a range, something very common, it gets lots of good and profitable trades very fast.  There are other strategies that try to find a good entry point when several conditions are met, but they make much less trades. Its a different concept of what strategy does.

[quote quote=10253]unfortunately, as I expected the P&L is showing wrong profits, I did manual calculations and found that my gains are higher than what PNL shows.[/quote]

How are you calculating the PNL? I recommend you to take a look at this post, maybe it can clear you some things:

https://gunbot.shop/forum/topic/pnl-calculation-gunbot-pnl-or-binance-pnl-which-one-is-better/

[quote quote=10253]So I was wondering can we increase the distance between the first 3 DCA rounds? for example first DCA occurs after 1.5% down in the price (the current setting is around 0.5%), second & third one each occurs after an extra 1% down (the current setting is around 0.75%), the rest can be the same as the current setting.

I think if my idea is applicable then I can higher my TL and down the DCA rounds to 10 or 12 without worrying about the big dumps[/quote]

Yes, it’s possible to increase the distance between the first 3 DCA. DCA distance is not something fixed, it changes dynamically depending on several factors. By changing Grid Multiplier you can increase the distance %. For example, setting Grid Multiplier to 1.5 would make the bot to increase distance between buy points by 50%.

DB1 already changes it dynamically depending on the situation. Anyway, I don’t think it’s a good idea to increase the distance for the first 3 DCAs and it would make you to lose a lot of fast and profitable trades.

[Aitor]

Quick update: Last week there was another big dump. BNB dropped by 16%, ETH dropped by 15% and UNI dropped by 23%. The distance to my break points at this moment are: USDT-BNB: 11%, USDT-UNI: 22% and USDT-ETH: 30%.

I’m relaxed as I feel the bot is doing a great job. ETH is involved in a cycle that started in 4400 and it managed to lower break even price to 2500, and it keeps doing its job, so I know it’s a matter of time… and patience.  We are facing huge dumps and today all the market had a big “crash”, including crypto markets.

However I know that sooner or later prices will bounce, and considering the good job the bot did lowering average bought prices, I’m sure it will be ending these big cycles with huge profits very soon, and I will be reporting it here ;) :thumbup:

By the way, I’m testing v4.0 and I’m very happy with v4 performance, as now CT sales are more profitable, and at the same time it’s more balanced than previous versions. Sometimes it needed to be more conservative and wait for a better buy point, and sometimes it needed to buy as the price was already a very good buy point and it had no sense to keep waiting and lose the good opportunity. Definitely v4 has much better balance and better adaptation to the different market situations. Here you can see some improved CT sales while getting better buy points, something that helps to reduce the average buy points at a faster pace while saving funds:

[Aitor]

Just a super quick update: USDT-BNB pair just ended the cycle with nice profit: 51,78$ :clap:

 

[Aitor]

New update:

The last 20 days the market has been an ass. BTC dropped like 25% and altcoins dropped even more and they all kept going up and down in a very small range.

Despite of this UNI ended a big cycle with good profit: 41,50$

After ending this cycle, the pair USDT-UNI is cycling nicely now:

Regarding other pairs: USDT-BNB is about to close another big cycle, it’s 4% below break even price and the bot is lowering ABP constantly.

USDT-ETH is still involved in the huge cycle that started when price was in the peak, 4400. The bot keeps doing it’s work and it managed to lower ABP to 2600, which is really close to current ETH price (2300), so hopefully it will close very soon this huge cycle with huge profits.

Here you can see the profit I got for each pair:

And daily profit:

Latest days the profits have been lower than usual. Nothing to worry about, it’s normal with current market conditions. And patience pays off, you will see ;)

– Regarding v3.5 and newer v4 versions:

v3.5 is definitely better than previous versions. It fixes the issue we found with original and v1 and v2 version that made the bot trying to find a lower buy point when the price falls very fast, and it made the bot not to buy even if the price dropped 30% in some situations. That was a big bug and it’s fixed in v3.5. On the other hand, when the price falls very slow and goes sideways, v3.5 was a bit too aggressive in some circumstances, and that something I have already fixed for v4. That way v4 has a much better balance between being conservative and aggressive depending on the situation.

Other improvement I’m about to implement to v4 is to increase the profit with Continuous Trading sales. CT is really good and it helps to get lower buy points while saving funds, but at the same time we noticed that many times the profit for each CT trade was very low, so in v4 I’m going to force a higher minimum profit. This way we will get a more powerful CT. Most likely the new post will be the continuation of this one but running v4.

[Aitor]

Increasing trailme could lead to lower buy points… Also note that the easiest and most direct way to get lower buy points is to increase Grid Multiplier parameter.

[Author]

Posts