Opps, needed to login first. Thanks again!

Added! Golang is nice, have picked up a little bit but not enough to be comfortable with it yet.

• "app.py" is the web server
  • HTML is in the template folder
• "enrich.py" is a one-time run which enriches the data a little (would be great to have the cover art be a part of the PriceCharting API, I would be surprised if there isn't a feature request open on that already)
  • This script requires two environmental vars "PC_EMAIL" and "PC_PASS", which I use to grab the cover art of the game.
• Ignore the fact that I have data in my Git repo :D
  • Would never typically do that but just wanted to avoid having to setup a DB since this was just a quick prototype (is also a big reason why the repo is private honestly; and data was paid for).
• The Pandas library in Python handles a lot of the numerics (sampling, etc.)
• Repo was originally used for a different game idea I had (is why the .gitignore is 3 years old... :D)

Yes I did, I used Python/Flask because that's what I know for my day job. If you send me your Github username I can add you to repo (was a quick prototype so pardon the mess). Is only about ~80 lines of Python and a little more of HTML.

Absolutely! Would make sense to have it on PriceCharting (could be a sub feature) and the data it would generate would be interesting too! Perhaps could help find insights like, just an example, "People with > 70% overall game price accuracy typically underestimate the price of combat games.", etc.

Some things I have found just playing with my prototype:

• Weight random game pick by sale volume
• Prices to pick from should be > ~5$ apart at least (otherwise feels a little unfair)
• Can show recent 10 picks but player's accurate score should be based on a much larger sample (I think ~200 samples typically coveraged to ~33% accurate if you picked the middle price every time)

If there is interest I would love to participate and provide feedback, etc!