BitcoinEcon

Bitcoin aficionado & minimalist. I'm trading, not hodling. Less narratives, more data. Personal & algorithmic opinions only. No investment advice.

[BitcoinEcon @BitcoinEcon]

Algo Trading Article

Here is an extensive article that I wrote about algorithmic trading:
https://t.co/xsDtHTpm2U

I wrote it for the manuals of my trading algos. But makes more sense to make it available separately for everyone.

About my trading algos:
https://t.co/qF2J6aC0vw

[BitcoinEcon @BitcoinEcon]

@peter_vijn From biology to linguistics?
Check my DM!

[BitcoinEcon @BitcoinEcon]

Bob finally accepting the bear market hypothesis 2 months after me 😉. The 4-year-cycle is dead. Long live the the 4-year-cycle!

[Bob Loukas 🗽 @BobLoukas]

Most of crypto is behaving as they would, in a bear market.

Every Cycle we tell ourselves, "I won't round trip this time".

Yet we're conditioned to respond with inaction, saving our fear not for losses, but for the possibility of missing out.

[BitcoinEcon @BitcoinEcon]

The PLC model from 2019 predicted the BTC bottom nicely at 15k for 2022. For end 2026 it predicts a 50k-60k bottom. 52-60% down. Sounds fine with me.
It does not predict tops very well because of scarce data. There is a lot more bottom data, why it works better for them.

[hcburger @hcburger1]

What to make of the current price of bitcoin?
In Sep 2019 I predicted that btc would continue to move in the corridor defined by the red and green bands in the future.
You can see in red how the price has fared compared to those predictions since then. 🧵
https://t.co/oqCDPxyg1H

[BitcoinEcon @BitcoinEcon]

BTC did not hold 90k. Price action is very weak. 32% down already. First time below the 50 weekly MA since the start of this bull market in 2023. Second week below it.

[BitcoinEcon @BitcoinEcon]

Hold & Sell Dillema

Both my brother & a friend are still holding BTC. I told them a while ago that next year will be a bear market and that 126k was likely the top.

They don't want to believe it and think "This time is different" or that the bull market will last longer.

[BitcoinEcon @BitcoinEcon]

Personally, I already stopped my CHARON algo on the live account as I know that it has no edge in a bear market. Also too little data for this type of algo.

My new AI daytrading algo works during bear markets. But still not ready. I will surely use it next year.

[BitcoinEcon @BitcoinEcon]

Hold & Sell Dillema

Both my brother & a friend are still holding BTC. I told them a while ago that next year will be a bear market and that 126k was likely the top.

They don't want to believe it and think "This time is different" or that the bull market will last longer.

[BitcoinEcon @BitcoinEcon]

... tied to the position. Exiting too difficult & feels too final. Coming up with a plan on the fly does not work. Needs to be done before you are in the situation.

The 4-year cycle has repeated too many times to dismiss. Still, some people don't want to believe it.

[BitcoinEcon @BitcoinEcon]

@FireflyTA Congrats! I hope it will make you rich while sleeping. Huge amount of work you put into it.

[BitcoinEcon @BitcoinEcon]

I expect to see below 100k soon. No bullrun this year so far. Only bigger rally in Apr/May.
CHARON did some dip trades with profits in Oct. But I made more money selling AI models than trading this year. Looking forward to do AI daytrading when I find the time to get it ready.

[BitcoinEcon @BitcoinEcon]

@FireflyTA The 4 year presidential cycle even exists for stocks, although less visible with blank eyes. Very visible on BTC.

[BitcoinEcon @BitcoinEcon]

People piled into BTC as gold broke down yesterday. Didn't help. Back down again. My algos got stopped out. I exited in time.

We might have seen the BTC peak at 126k this cycle. Still a small chance that it will do a new ATH this year. But I don't like the price action so far.

[BitcoinEcon @BitcoinEcon]

All algos out again. BECKI very likely exits on today's close. I did not like the price action & manually exited 1.5 days ago in profit. 80% of the time my gut warns me correctly.
Either no bullrun this autumn or in 2-4 weeks. BTC did not behave like 2023/24 this year.

[BitcoinEcon @BitcoinEcon]

Perfect conditions for a late year & final bullrun of this 4 year cycle. And BTC made a new ATH on a weekend.

All my algos are in now. CHARON only entered today as it missed it by a tiny bit 4 days ago. On live I manually made CHARON enter with DRAGON. Now I got a 2.5x position.

[BitcoinEcon @BitcoinEcon]

BEYOND also entered one day after BECKI, BEAST & DRAGON.

I will try to add a feature that will make CHARON enter nevertheless if there is a signal very close below a threshold. Might add some losses but should avoid getting late into a bullrun.

[BitcoinEcon @BitcoinEcon]

Perfect conditions for a late year & final bullrun of this 4 year cycle. And BTC made a new ATH on a weekend.

All my algos are in now. CHARON only entered today as it missed it by a tiny bit 4 days ago. On live I manually made CHARON enter with DRAGON. Now I got a 2.5x position.

[BitcoinEcon @BitcoinEcon]

That signal below from 2 weeks ago was early. Got stopped out.

Now BECKI is 🟢 again - valid at daily close. BEAST with 100% progress already confirmed the signal. DRAGON entered already. CHARON would have entered if price had moved a tiny bit higher earlier. BEYOND is silent.

[BitcoinEcon @BitcoinEcon]

The nice thing about it: It converts model v2.2 from a scalping into a daytrading model without retraining. And it additionally reduces drawdown & losses. I did not think that these AI models would be so robust to allow that.

[BitcoinEcon @BitcoinEcon]

AI Trading

Model v2.2 might still work. After a lot of debugging it turned out that ChatGPT messed up the live trading code. I fired ChatGPT, because it made too many mistakes & was too bossy. Annoying combo. Now Claude & I clean up the faulty code to make it work. Almost there.

[BitcoinEcon @BitcoinEcon]

AI Trading

As suspected the trillionaire model v2.2 is a loser. Still did like 400% per month in latest internal backtests, but now in a real backtest it just loses money (see below). Means that the trading simulation (used for training & internal backtests) is still imperfect.

[BitcoinEcon @BitcoinEcon]

Claude added a bug that showed me how to make the AI model stay longer in trades and not exit on smaller fluctuations. I love these bugs that tell you how to finally make a trading algo or model work.

[BitcoinEcon @BitcoinEcon]

That signal below from 2 weeks ago was early. Got stopped out.

Now BECKI is 🟢 again - valid at daily close. BEAST with 100% progress already confirmed the signal. DRAGON entered already. CHARON would have entered if price had moved a tiny bit higher earlier. BEYOND is silent.

[BitcoinEcon @BitcoinEcon]

My BECKI algo triggered yesterday. But none of my 4 other algos. Could be early or a fake like last time. Price action so far not convincing.