Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

Alternate voting systems

A topic by mechabit created Oct 11, 2025 Views: 689 Replies: 13
Viewing posts 1 to 8
Submitted (9 edits) (+6)(-1)

Hey so lots of people complaining about the comments and voting system but no one talking about alternatives. 

The current star average system favours games with few votes. A game with 10 5-star rating beats a game with 100 ratings with 4.9 average. You can just get 10 friends or 10 alt accounts to win this way (above the median votes threshold). Similarly a great game can get hidden by a few 1 star ratings because of small sample randomness. 

Here's a couple ideas for alternatives to the 5 star average system. Feel free to add your own in the comments.

1) Advocacy voting 

  • Instead of 5 stars system you click on the categories you think this game should win in
  • Can vote for multiple games in multiple categories
  • Could add more categories for things like best student game or best first game or made me laugh etc.
  • Used to select the pope sometimes

2) Bayesian average 

  • Instead of the current mean-average star rating. 
  • Weights scores based on the number of votes 
  • Games with fewer ratings are adjusted toward the global average.
  • Add a minimum threshold of votes to get ranked. 
  • Prevents games with few 5-star votes from topping the chart.
  • Doesn't eliminate 1-star bombing
  • Used by sites like IMDB for movie rankings.

3. Pairwise (Head-to-Head) Voting

  • Voters see two random games (or from the games that they've played) and pick which one is better. 
  • Repeat many times; ranks are computed via Elo or Bradley-Terry systems.
  • Removes rating inflation bias.
  • Easier for players, just “which did you prefer?”
  • Builds a robust ranking with enough comparisons.
  • Used in Ludum Dare’s “smart randomizer” for a while in testing phases.

4. Ranked-Choice / Borda Count

  • Each voter ranks their favorite games (1st, 2nd, 3rd, etc.). Points are assigned (e.g., 1st = 5 pts, 2nd = 3 pts, 3rd = 1 pt), and totals determine winners.
  • Can't down-vote to boost yourself


I like advocacy voting or pairwise head to head voting best. But option 1, 3 and 4 prevents the 1 star bombing trolls.

Submitted(-1)

Your assumption is actually incorrect. Games with fewer votes do get an adjustment - you could call it a penalty, but what it does is balance the ratings so the games with fewer ratings aren't automatic winners.

The formula for jam ratings was explained by an admin. According to them, the formula is:adjustment = sqrt(min(median, votes_received) / median)

This means that the adjustment for a game that got 19 votes in a jam with median of 20 is = 0.97467, about a 2.5% reduction in average score, eg. a 4.5 rating would go to 4.386 for the purpose of comparison during ranking.

Submitted(+1)

Yeah I know, these are alternatives

Submitted(+3)

I think, if anything, it's good we had a game jam this big to spur on this conversation. There certainly seems to have been some disconnect between comments and ratings. I also would vote for advocacy ratings, but brackets could be a very interesting way to handle things AND could help you players discover games they otherwise wouldn't.

Submitted (1 edit) (+3)

exactly, I would personally say make it either 1 and 2 and then a private rate with judges after, maybe take the top 20 or so games, that way then we still have a voice but also anything that slips through the cracks or people consider rigged in that top 20 can also be filtered through a set of people who wouldnt be biased, as they would not have submitted into the jam, therefore actually have incentive to intentionally 1 star bomb games/fake reviews. etc

Submitted(+1)

GMTK game jam has both a public vote and he picks a couple as his personal favourites 

Submitted(+1)

yea that's fair, just my personal pref really, I think they're doing something similar here now, where they're picking through their own favs too for jack to play now

Submitted(+2)

I think you could only vote if you were a part of the jam. So no 10 friends. Some people do mistakenly assume that visibility equals high rating, but that isn't necessarily true. I mean, this point system isn't perfect because 5 stars is a subjective measure and people have different views of what "3 stars" mean.. but any solution that benefits high rating numbers above anything else is just turning it into a marketing contest. 

Submitted(+2)

to me thats also the big thing, we could only vote, which meant other people, who wanted a better chance, would be faking "good" reviews and then 1 star bombing, its already been shown some people/devs have done that (they did get banned I think which is good but still!)

also fair, its always tricky to get a rating system right

Submitted (1 edit) (+1)

much harder to cheat the system with Pairwise voting and voting doesn't necessarily need to just be for jammers

Submitted(+1)

I think there should be penalties for everyone under the maximum rating count, but it should be logarithmic in some sense. So if the top game has 200 ratings games with 100 ratings receive a -1% penalty, 50 ratings receive -2%, 25 ratings receive -4%, 12 ratings receive -8%, and so on. Something like that.

Submitted (1 edit)

Yeah that deals with small sample randomness but trolls can still 1-star bomb.

I like the posts people make of their own favourites lists. They should just make that a mechanic.

Submitted(+1)

Oh, I really like the 3rd option! I was brainstorming with some of the top placements (who were also kinda bugged by this even though they won!) and we all agreed that there had to be some way to fix the 1 star bomb ratings, as well as prevent the amount of reviews problem. This head to head option is great, and we also felt that reviews and comments should be hidden until the end of the voting period since you can't change your game anyway. 

Also, my heart breaks after looking at the full list of corrected ratings under the bayesian average. So many fan favorites jumped up about 40 spots, while the games many are complaining about dropped by tons. It certainly does seem more fair, even without accounting for the review bombs. Since this is my first game ever, I had no idea itch had such a strange rating system that people have wanted changed for over 10 years, but experiencing it first hand, I want to make a change too! 

I felt so bad watching some of the most beloved games drop below me -- I'd have given them my spot if I could! While I'm extremely proud of my game, I definitely don't think I deserved to be higher than Grim Was The Acre, that's insane! 

Regardless, there's likely nothing that will be done to remedy this. I just hope that, in the spirit of the game we made, people will continue to pursue their dreams despite what puts them down. There were hundreds of spectacular games this jam with so much heart and soul put into them, and I truly hope no one gives up because of their itch rating! If anyone needs someone to talk to, you can message me on here, or under the same name on discord, youtube, or bluesky! Never give up, dreamers! And good luck! <3 

Submitted (3 edits) (+1)

I've emailed itch to add more voting options for game jams, if enough people do it it might happen.

Vote queue is one of the options in the current system which does help spread the votes a bit, could get itch to add pairwise voting as another option