# Just for fun, I made a program that predicts the price of bitcoin

I was bored trying to think of a python program to write and I decided to write a bitcoin predicting program.

How it works is it looks at the price graph for the last N (cycling through different values of N) days then tries to find the time the most similar pattern in price occurred in the past to the current pattern, then predicts the price based on what happened after that previous pattern occurred.

This is probably worthless and won’t be an accurate reflection of the future price at all, but just for fun, here’s what it says about the next few years of bitcoin.

It says that the current pattern we’re in is reminiscent of 2018-01-06.

On the graph, the left side of the green line is the recent price history (red line) and historical price history from 2018-01-06 (blue line), and right of the green line, the future price (red line) is modelled after the price history from 2018. The y-axis numbers on the left are the recent/future price and the y-axis numbers on the right are the historical prices from 2018.

**90-day prediction:**

Graph:

Peak: 2022-02-24 \$41,030.37

Bottom: 2022-04-19 \$24,538.98

Final: 2022-05-21 \$34,227.86

**1-year prediction:**

Graph:

Peak: 2022-02-24 \$41,030.37

Bottom: 2022-10-06 \$13,340.72

Final: 2023-02-20 \$22,262.68

**(just under) 3-year prediction:** (falls just short of 3 years, I think because it doesn’t have enough data)

Graph:

Peak: 2025-08-30 \$282,899.62

Bottom: 2022-10-06 \$13,340.72

Final: 2025-12-13 \$155,385.66

So there you have it. Just a bit of entertainment. Don’t go investing based on this. I’m gonna keep track and see how successful its predictions were.

View Source

### 47 thoughts on “Just for fun, I made a program that predicts the price of bitcoin”

1. *OP clearly indicates in the title that this is **for fun***

2. I feel like if it’s accurate we will never hear from OP again

3. So you are saying I have to wait a while to buy the peak?

4. Maybe you could let it run on a website? You could add other coins as well and accept donations. It’s interesting to have AI predict prices and it’ll be interesting to see its performance compared to real world graphs.

5. I did too. It was just a random number generator.

6. ##TL;DR : Bitcoin will hit 100k by the end of 2021.

7. You wrote that in Python? That’s very Basic. I give it a C, or C+ at best.

8. i will be taking this as financial advice

9. *sets limit buy order at \$13,350

10. !remindme 3 years

11. Calender marked for October, I’m going all in. Thank you for the investment advice! ![gif](emote|emo_pack_1|dancing_wojak)

12. Oh, I love a little Standard Deviation measurementâ€¦ but yeah you should put it up on a site and add the top 10 or 20 coins and see if you can find anomalies outside of the usual â€śBTC is the marketâ€ť

13. OP just gave me all the hopium I need till 2025

14. Bullish on Python

15. ![gif](giphy|j6uK36y32LxQs)

16. RemindMe! 6 October, 2022

17. Finally, someone tells me the EXACT price in the future.

Thank you, I’m off to the bank đźŹ¦.

18. That is pretty friggin awesome OP. I have a lot of respect for you taking the time and having the knowledge to do this. It was a fun read thanks.

19. I am a chartist and I can say that from a broad perspective this is realistic. It appears as if you have used the 2018-thru-current data to proportionately predict the next market cycle. The 2018/2019 bear pattern was unique due to events surrounding mining profits and tether creation, but overall the basic idea and price structure is intact in my opinion. I like your model, sir.

What you could do is look at the 2014 run and determine the change between the 2014 pattern to the 2018/2019 pattern to extrapolate the overall proportionate change to the NEXT pattern.

What may foul things up are wider market conditions. BTC has grown in a decade of loose monetary policy where risk-on assets have had uninterrupted gains. This may be an issue if a wider recession hits.

20. YOLO everything into btc at \$13k and sell at \$200k. Financial advice taken

21. 4-20 Iâ€™m YOLOâ€™ing bitcoin

22. If btc ever goes to 13k I will most likely invest more than I can afford to lose.

23. I will take this as financial advice and ape in all my life savings, thank you

24. Wow, quite a big difference in the 1-year and 3-year prediction. Guess it’s better to wait until 2024 to buy lol

25. **BTC to 200k confirmed**

26. Stills a better prediction than Plan B 100k EOY 2021!

27. Three year projected chart looks exactly like the current 1W chart, which makes me think your model may rely a little too much on history repeating itself.

28. Thanks OP – however please be careful bc this looks like a classical case of overfitting

29. Set buy order 282k. Set sell order at \$13k. Am I doing this right?

30. I bet it is damn close

31. Lol yeah đź‘ŤđźŹ˝

32. Damn, I wish I had programming knowlegde to do this. Must feel dope to be able to do this…

33. you can definitely add this to your portfolio if you wanna be a software enginner. or maybe you already are idk lol

34. This strokes with my idea though, we are clearly not in depression and after a long and boring period of consolidation we will go down once again to freak out the last hands but most of all to give the MMâ€™s the chance to buy everything up one last time since retail will have spent all their money. It makes sense that your bot â€śpredictsâ€ť the same since Iâ€™m basing this idea off of the last bear market pattern just like your bot does. 13k is damn low though, but I think \$20k isnâ€™t out of the picture and since support is close to non existant at \$20k a lower price is very much possible. Fuck the old oracle, this is the new oracle :D.

35. So basically, if the 90-day prediction turns out true, sell everything and wait 6 months before buying back in

37. Your 1 year prediction is depressing at best.

38. Like the 3 year prediction but I’m not so much a fan of the 90 day and 1 year ones

39. Lmao all these predictions are not going to happen.

40. Instructions unclear: YOLO all I have in BTC because of the 3 year prediction

41. â€śPredictsâ€ť