Almost a year ago, I made a program that predicts the price of bitcoin. I decided to go back and check the predictions, and they're not too far off.

Back in February 2022, I made a post here about a bitcoin predicting program I'd written.

Here's the original post: https://old.reddit.com/r/CryptoCurrency/comments/sz29b9/just_for_fun_i_made_a_program_that_predicts_the/

The TLDR is the bot predicted a drop from 50k to 20-30k levels then another drop to 13k levels with a bottom $13.3k in October, then a final price of $22.2k by February 20 2023.

Here you can see a side-by-side comparison of the prediction (red line, left, feb 2022 to feb 2023) with the actual price (right, last year of prices).

https://i.imgur.com/Nr95kug.png

The prediction seems to have kept pretty much the same pattern, or at least trend, for the last year. It's not a perfect match, obviously, but I think this is a pretty good prediction for a program I wrote when I was stoned in the middle of the night.

Here's the three-year graph from the original post: https://i.imgur.com/m76ks2c.png

Peak: 2025-08-30 $282,899.62

Bottom: 2022-10-06 $13,340.72

Final: 2025-12-13 $155,385.66

So if this graph is anything to go by, we could see $282k bitcoin in 2025.

With all this said, this is just a price predictor made by an amateur based on a very simple principle. The fact that it matches is probably mostly luck, but still, I thought it was interesting that it seems to have some ability to model the future market.

I did have the full list of prices for every day in the predictions, but I lost the text file when my hard drive died. It should be possible to get the original prediction by feeding it the same parameters and old market data from the date I made the original post, I'll see if I can recreate the original prediction at some point.

What are your thoughts on this?