To my knowledge, range voting is the most optimal way to run a single-winner election or poll. Aside from being dead simple (e.g. score each choice 0-9, highest sum wins) it's staggeringly unmatched in computer simulations calculating Bayesian regret compared to all other systems regardless of voter utility, strategy, honesty, ignorance, or noise.

Is there a web app for conducting polls with good support for range/score voting? A self-hosted solution is perfectly fine, although unnecessary. Any required architecture or platform will do.

Google has lead me to SurveyMonkey, PollDaddy, EasyPolls, StrawPoll, and none had this. Some came close but the votes weren't kept secret. StrawPoll allows approval voting, which is at least a step in the right direction but the results should be hidden until finished.

BetterPolls has the best implementation I've found but seems relatively small and lacks features such as optional single-IP tallying. (Although it does have a mild captcha.) OAuth login support (e.g. Google+, Facebook) would be ideal. I know it's impossible to stop all ballot stuffing but it would be great to see an app for this superior system, particularly for fast and easy straw polls.


We are working on an app which enables range voting (http://www.wecide.org).

The focus lays on easy group decisions. We don't have a login, because it's faster and because we expect a high level of trust among the group members. It's rather about friends who want to find the best decision for everybody, instead of cheating on each other.

Also, we show the votes of the other voters right after one-self has voted. The idea is to exchange each others opinion very fast and not hold it back. But if we like the idea, we could think about implementing it as an optional feature.

Please check it out. It's in an very early stage. So we can use lots of feedback. You can also write us at info at wecide.org

  • 1
    Looks good! Votes should probably be permanent though unless this is for extremely tight-knit groups. (Maybe its current state is temporary.) I don't think seeing results should be a part of the voting process as that distorts the data, "I see these 2 are in the lead, I'll vote 100% for one and 0% for another instead of how I really feel!" (This strategy works well for non-range-voting polls that are easily predicted by a small sample so it wouldn't be surprising if they do it.) Defeats the benefit of score voting where honesty is also highly strategic. :( Feb 13 '15 at 22:37
  • this domain is used by other site now. seems about skin care cream, japanese.
    – qdinar
    Jun 9 at 13:04

My recommendation is for https://github.com/amikula/rangevote by Alf Mikula. I know him personally, and have used the software extensively.


It's very functional and fairly polished. It has been possibly the most recommended web app among the range voting community. When it was a hosted solution years ago, it was clearly the best software for range voting available.


The author moved on to other projects, and while this project got fairly polished (not beta quality), it hasn't been worked on in many years, so it's using a very old version of Ruby on Rails and may have other age-related weaknesses.

If someone were to fork the project and update it to modern standards, that would be the best solution available.


I am one of the developers of https://pollunit.com and i think our tool meets all your requirements.

  1. Range voting: with star rating or dot voting you can define score ranges
  2. Secret votes: with a premium account you can hide votes or participants
  3. Login: your users can login with Google or Facebook. No account required to cast votes.

For more info you can take a look at our tutorial for range votings

  • Thank you. Next time I need a poll I will try it, and then revisit whether this question is finally "Answered"! Oct 10 '18 at 21:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.