If you're into bot trading or even creating your own bots, you probably noticed that the platform you choose doesn’t really give you that "trading edge".
Recently, some colleagues and I thought it would be interesting to apply a volatility breakout strategy across different platforms to see how much it really matters.
We started with OKX and Binance, which are their go-to choices, but I figured we should also try out Bitget’s API with their GetAgent AI since I've seen some discussions about it.
Same strategy, same setup, all over a period of 90 days. The APY results are in the screenshot below.
I might create another thread to share my thoughts on why the results turned out the way they did, and maybe another one discussing AI in trading. But it all depends on how this thread goes.
So what is the goal of this thread? To let people know that bots are profitable or what?
You do not need to make another thread if what you want to discuss is about bot trading and the result, you can use this thread for it. Instead. We are seeing what you are posting.
As for me, I prefer manual trading.
How did you create your bots and what was the parameters set, also how long did it take to create this result?. I'm thinking of exploring bot trading further and this could potentially make my decision.. no need for separate thread.. just do it here.
From your SS the APR on Bitget is slightly better than Binance and Okx.. any reason for that?.
Short term will not mean anything, but if you are using 100% exactly the same parameters and same bot, then you are going to get exactly the same result in all of these places on the long enough term.
There could be a small arbitrage at times, but that would be short term, keep doing that over course of months and years, and you should be getting nearly the same.
Sure something like 11.32% to 11.23% could happen, that's tiny, but over course of long term that may mean something to you.
I would highly suggest keep doing this for long term and see what the gap is, I bet that it won't be much. Binance seems low at the moment compared to others, I am sensing that will be either gone, or it would at least make more sense because it's used more.
For this test we used 90 days, and if I do another comparison, 90 days would be the bench mark. I do agree that the gap becomes more significant when you do it long term.
I trade manually too (mostly FX but also Crypto) and to be fair I was a lot more comfortable using bots for Crypto trades. To answer your question the whole point of the thread is just to show that it is also an option and it worksjust as good on different platforms.
What I can say is, I noticed that it had less slippage and tighter spreads, it also helps that its less crowded, which means fewer aggressive bots fighting for the same entries.
For your other question, check Using Getagent to write quant trading strategies. Remember it's not something you should run into without properly understanding how it works.
I can see that Binance has the lowest score, though I didn't expect that. Still, I have seen and told people that Binance is a big market manipulator, a "bucket shop" that I have caught red-handed many times, especially during the volatile market conditions. Thank you for exposing this further.
Please do so, it will help further. But as for the results, I guess it will always change depending on the time and the prevailing market conditions.