Welcome to TabletopRecommender.app
Enter your boardgamegeek, rpggeek, or videogamegeek username above to get personalized game recommendations based on your ratings.
Your recommendations will be formatted like this sample list of popular games:
| 1 | 9.7 | Monster Manual (2025)(2025) Dungeons & Dragons (5th Edition) Core Rules (min needed to play) |
| 2 | 9.5 | The Elder Scrolls: Betrayal of the Second Era(2025) 1-4 players Cooperative Game Solo Play |
| 3 | 9.4 | Clair Obscur: Expedition 33(2025) Sandfall Interactive Single-player |
- Blue entries are role playing games.
- Gold entries are board games.
- Red entries are video games.
- The decimal number is the recommender's best guess at what you would rate the game.
- Use the filter buttons and menus on the recommendations page to customize the recommendation list.
- Search for games in the recommendation list with the search bar at the top of the recommendations page.
- Share your recommendations using the Share ⤴ button at the bottom of the recommendations page.
A lot of thought and care has gone into the accuracy, usability, and aesthetics of TabletopRecommender. Thank you to the rpggeek regulars for your invaluable suggestions and feedback.
Have fun!
FAQ:
How long will it take to get my recommendations?
The first time your recommendations are generated, it could take up to 24 hours--or even longer for a very large collection. On the other hand, updating your recommendations may take anywhere from 10 seconds to several hours depending on the size of your collection. Feel free to contact me if you have trouble getting your recommendations.
Why am I seeing games that I own or have rated in my recommendations?
Use the filter buttons and menus on your recommendations page to filter out games that you own, that you've rated, and to access other filtering options.
How does it work?
The recommendation algorithm places a strong emphasis on objectivity, discoverability, and surprise. It uses a direct form of collaborative filtering--gathering the other users who have rated games in your collection and using their preferences to inform your recommendations. The recommender is blind to factors like genre, game mechanics, or series associations--instead relying purely on preferential associations derived from the game ratings of other users. The goal of the recommender is not necessarily to recommend games that you're already interested in, but rather to recommend the games that you would most like if you played them, based on the experiences of users with preferences like (and in opposition to) your own.
The recommender gathers your 'board game', 'RPG item', and 'video game' ratings, the ratings of related users, and game information using the BoardGameGeek API. Note that if you have only rated RPG systems on RPGGeek you will not receive any recommendations--the recommender uses 'RPG Item' ratings.
What does the number next to each recommendation mean?
The decimal number is the "recommender score" and represents the recommender's best guess at what you (or the user whose recommendations you're viewing) would rate the game in question.
Why are scores able to go over 10.0?
All else equal, the more ratings and associations a game has in the recommender the lower it's average score will be. For this reason the recommender uses a custom Bayesian model to determine how much game scores are penalized at certain popularity levels (more accurately, association count levels) and compensates with an equivalent score bonus. This is necessary for the recommender to be able (potentially) to accurately estimate what a user may rate a game. If a game has a score over 10.0, it's because the game had a very high recommender score before the score bonus was applied, in most cases suggesting a very strong recommendation.
How does search work?
- Search can only return games that are actually in your recommendation list.
- Search is case and accent insensitive and includes partial matches. Searching 'poke' could return 'Poker' and 'Pokémon Red and Blue'.
- Search includes role playing game systems. Searching 'cthulhu' could return the item 'Keeper Rulebook' from the RPG 'Call of Cthulhu'.
- Search includes video game developers. Searching 'nintendo' could return 'Star Fox 64' from the developer 'Nintendo Co., Ltd.'.
- Use quotes to perform an exact search (case insensitive). Searching "ra" (in quotes) could return the game 'Ra' but not 'Terraforming Mars'. Searching "7 wonders" (in quotes) could return '7 Wonders', but not '7 Wonders Duel'.
What is the list of games and numbers under each recommendation? (recommendation sources/connections)
These are the games you've rated that have had the most significant influence on a given recommendation (not all of the games you've rated that influenced a recommendation). The number (x) to the right of each game is the "connection count"--the number of times the rated game has been connected with that recommendation.
Only the relatively most significant and impactful connections for a given recommendation will appear in the list. In most cases the list will only display a fraction of all the connected games for that recommendation.
How do board game player counts work?
The player counts for board game entries are derived from boardgamegeek's community recommended player count polls. If there aren't enough votes (at least one for each player number), the game's "box" player count is used instead. The recommender counts a player number as valid if the combined boardgamegeek "best" and "recommended" votes are greater than the "not recommended" votes.
Contact:
Do you have a bug report, feature request, or other feedback? Send me a message on boardgamegeek!
Latest updates:
June 22, 2025
- Overhauled the filter system. The primary filters are now directly on the recommendations page and filtering is done without page refreshes or additional windows.
- Added a release year filter.
June 11, 2025
- Added a Share ⤴ button to the bottom of the recommendations page for sharing recommendations on the boardgamegeek forums or in plain text format.
June 9, 2025
- Fixed a bug that has been affecting the quality of recommendations. Requires an update.
June 6, 2025
- Added recommendation sources/connections--see the FAQ for details. Requires an update.
- Fixed a bug where expansions could sometimes be categorized incorrectly.
June 5, 2025
- Algorithm improvements (requires update).