Steam Hype

This repository contains Python code to find upcoming Steam games with many followers.

Requirements

• Install the latest version of Python 3.X.
• Install the required packages:

pip install -r requirements.txt

Usage

As of November 2020, SteamHype.com is down. Please refer to this Colab notebook to rely on SteamDB.

Alternatively:

python compare_to_top_wishlists.py

or:

import compare_to_top_wishlists

compare_to_top_wishlists.main(version=2)

---

Previously, when SteamHype.com was up, usage was as follows.

To download data:

python download_hype.py

To print formatted output:

python utils.py

To compare top followed games to top wishlisted games:

import compare_to_top_wishlists

compare_to_top_wishlists.main(version=1)

Results

Results are shown on the Wiki.

The rank-order correlation between the top followed games (depth=166) and the top wishlisted games is positive:

• Spearman rho coefficient ~ 0.28
• Kendall's tau coefficient ~ 0.19
• Weighted Kendall's tau coefficient ~ 0.60
• Average overlap ~ 61.7%
• Rank-biased overlap ~ 62.3%

As of November 2020, using SteamDB's hype (depth=250), as SteamHype.com is down:

• Spearman rho coefficient ~ 0.44
• Kendall's tau coefficient ~ 0.33
• Weighted Kendall's tau coefficient ~ 0.64
• Average overlap ~ 66.5%
• Rank-biased overlap ~ 68.0%

References

• A tweet by the game dev Lars Doucet
• A blog post by the indie consultant Simon Carless
• An official ranking of the most wishlisted upcoming Steam games
• SteamDB:
  • upcoming games clustered by release date,
  • upcoming games sorted by descending number of followers, after this tweet in November 2020.
• Steam Hype API, which is not my work, and which my script relies on. Edit: As of November 2020, the API is unavailable.
• Webber, William, et al. A similarity measure for indefinite rankings. ACM Trans. on Information Systems 2010.