I created, organized, and ran an unofficial, prizeless awards event from December 2018 through mid-February 2019 celebrating cartridges made in the PICO-8 fantasy console. Many of these were games, but some are just interesting visual toys or experiences. The full details of the event and the results can be found here: https://www.lexaloffle.com/bbs/?tid=32427

PICO-8 has a relatively small but active and dedicated community. I created this event as a way to encourage the community to get together and celebrate our collective creative works.
While the event had ultimately mixed results, I believe it did succeed in this objective and was a good learning experience. Here are my main takeaways:

Lesson 1: Lighten the load where possible.

Planning, running, and tallying the results for the event proved to be a huge amount of work for one person to do manually.

How to Address This:

• Collaborate with other people to divide the work. - In retrospect, having a community event run exclusively by one person is rather backwards and inefficient.
• Find ways to automate response processing, especially for nominations. - I had resisted this due to the unpredictable nature of write-in responses, but writing scripts to automatically attempt to process nominations or to automatically format nominee lists for posting them could have saved a significant amount of time.

Lesson 2: Small samples are challenging to work with.

The relatively small size of the PICO-8 community contributed to several issues with the event.
When collecting cart nominations, there were only 29 responses total, leading to there being very few repeat nominations and thus no way to narrow the voting down to a few top nominees per category.
There was a much better turnout for the actual voting process (186 responses), but the results still proved problematic. A couple carts received disproportionate numbers of votes from people outside the main PICO-8 community (one had recieved media attention in France, while another was a student project supported by the creators' classmates), and due to the small number of votes recieved overall this gave these games a disproportionate edge.

How to Address This:

• Allow or require people to do ranked voting. - This will give a better idea of people's preferences, keep people from supporting only a single game, and help ensure all games have the chance to recieve significant numbers of votes.
• Enlist the help of prominent community members. - I organized the event over Twitter but only had a few hundred followers at the time. Although more prominent community members helped spread the word, I think collaborating with them more closely could have heightened the event's reach.
• Collect nominees and votes over a longer period. - The issues of limited turnout likely could have been reduced if people had more time to spread the word about the nominations and voting.
• Promote more aggressively. - While I posted about the nominations and voting weekly on Twitter, more frequent posts could have helped ensure a broader range of people heard about it.

Lesson 3: Transparency is key for damage control.

When people got (understandably) frustrated that these games won many of the awards seemingly out of nowhere, I opted to explain the situation as openly as I could, including posting the response summary charts from the Google form I used for voting.
People overall proved to be understanding, and some offered advice for how to improve it in the future. The unofficial nature of the awards and lack of concrete prizes also worked in its favor, as there was no real stakes in winning or losing. People were also openly appreciative of my efforts and intentions in running the event.