Algo trade system Warning: ftx, deribit, etc.

When ftx rolled out it’s automated trading software, I thought, ‘screw that.’ But then I stupidly tried it anyway. It cost me some moneey. How much? I’ll get to that. (See how I hooked you, reader/fish?)

I wanted to trade the daily Move contract on ftx. Just one and only if it dropped to $120. The outcomes looked good. I set it to trade 1 contract.

Trouble is, I don’t do algo trading. This fucking thing bought a contract for $120 every 15 seconds! as it cycled through. It stopped only because I had the system set to 20X leverage and it traded my leverage to max. I finally realized what was happening (it was over $1 million in notional value.) The notional value was pegged to Bitcoin.

Me without a face and with less money. Note the lack of a fine Scotch that day.

The thing had bought 220 contracts. The price had dropped by $15. I panicked. I started selling. And it was buying them right back! At $120. The bid-ask spreads were huge – like $2 or so. (I’m blocking out the full horror.)

The price dropped further. I messaged the Telegram group, even chatted with the CEO. He looked, told me the problem, disavowed responsibility and didn’t help at all.

I didn’t like him much at that point (but they did help later – more hooking. Talk about crafting expert content. You’re in up to your neck on this story, eh?)

Bitcoin Algo Trading Horror Story – (Gratuitous SEO heading: blame Yoast)

I managed to shut down the bot, then went back to closing down the trades. The actual $ value was most of my account. And I had seen that contract settle at $1 a few times. (Which is what it did later that day.)

FTX’s MOVE contract

I should explain the MOVE contract. It’s essentially a straddle contract on Bitcoin. That means it has an upside option and a downside option, but they’re inseparable for this contract. The strike price is preset as the Bitcoin price when the contract goes live. However, you can buy the contracts before the strike goes live.

I was trading the dailies – a MOVE contract expires every day at 12:00 UTC and a new one initiates. So it didn’t have a long life span. Basically, it was the Avacado of trades. I guess the quarterlies are more like a potato.

And I’m not explaining that metaphor either. That’s because there’s 10 kinds of people in this world: those who understand binary code and those who don’t.

If the BTC price (based on the index – another thing to maybe explain in another post, if I gave a sh*t, let’s just say it works pretty well for price discovery and I’m not quite sure if that’s the correct terminology for this situation, but now the sentence is too long, so I better end it before the next period unless something else needs to be said, but it doesn’t, so .) Ahem, if the BTC price is $10,513 at expiry and the MOVE has a strike of $10,205, then the seller pays the buyer $308.

If the BTC price and MOVE price are really close, then I just lost a BUNCH OF FUCKING MONEY. And Warren Bugfest told me rule #1 is not to do that, even though Berkshire Hate-away lost (hang on – I’m checking how much) 28% during the Covid Crash. I guess Oracles from Omaha are not failsafe.

I’m pretty disappointed, too, because everybody kept saying that he was really holding cash, so how in God’s name did he lose 28%?

Shitty trading day
What’s the 2 lines on the right side?
Why is his/her head a stop sign?

Anyway, I sold off those contracts in 10 lots until I was out. Having lost a cool $5.5k that day in a few hours, I felt SH*TTY (sometimes I bleep, like if my kids come in, I quickly edit the cuss words and they’re clueless) for a long time.

ftx did help out, though. I remembered that the fees were based on the bitcoin price (about $8k at the time) rather than the MOVE price (about $120). And I had traded in and out with the higher taker fees mostly in both directions. I had paid over $1000 in fees for this debacle. They agreed to a refund, but made sure to remind me they were not responsible for me not understanding how to use their FUCKING tool. I’d look up the email for the ‘xact phrasing, but I don’t want to and I really doubt you give a rat’s beehonkus. (new word generator functioning properly.)

Conclusion

When I was learning all about the deathsport known as financial blogging, SEO power ranger Neil Patel’s blogpost on how to write a blogpost said, when you get to the end, write a conclusion. But I don’t have one, so I’m just doing my best while trying not to slobber on the keyboard.

Hey, I take trading and investing seriously, but not too seriously.

Leave a Reply