The problem with this is that predictions should, ideally, be clearly specified, offer a percentage estimate (0-100% likelihood), and have some reliable method of validating the outcome.<p>The Good Judgement Project does all these things: <a href="https://www.gjopen.com/" rel="nofollow">https://www.gjopen.com/</a><p>This website, on the other hand, seems to encourage none of these epistemic virtues. It's people making more or less precise verbal predictions on twitter and then assessing themselves.
Suggestion: require "skin in the game" ie. make predictors risk something of value (with commensurate chance of gain if their prediction manifests).<p>This would likely increase prediction quality/accuracy.
My favorite prediction from a similar site:<p><a href="http://longbets.org/601/" rel="nofollow">http://longbets.org/601/</a><p>Jeremy Keith bet that “The original URL for this prediction (www.longbets.org/601) will no longer be available in eleven years.”<p>He looks set to lose, which is great news for him, honestly.
Hi there,
My friend Nathan and I recently launched Predibly.com: a social platform to publicly share your predictions of the future and have interesting conversations about them.
Nathan had the idea last week, I build the MVP and we'd love to know your thoughts!
A relevant episode of EconTalk[1]. If I remember correctly, Philip Tetlock is running a continuous betting project to try to determine what characterize good forecasters.<p>[1]:<a href="https://www.econtalk.org/philip-tetlock-on-superforecasting/?highlight=[%22tetlock%22,%22on%22,%22superforecasting%22,%22%22]" rel="nofollow">https://www.econtalk.org/philip-tetlock-on-superforecasting/...</a>
I like it! Love that you can sort by time of prediction or by time of post/popularity. Sort by "controversial" could be nice.<p>As soon as you have more than a dozen you will need tags or category. Tech, science, politics, sports... will cater to vastly different people and foster different discussions.<p>I don't like the fact that Twitter is required to sign up.
I have a similar site I've been working on, although ours is more focused on betting and quantifying accuracy. It's <a href="https://hunches.app" rel="nofollow">https://hunches.app</a>
Nice idea! In case you haven't seen it, this site uses a similar concept:<p><a href="http://longbets.org/" rel="nofollow">http://longbets.org/</a>
I really like the design!<p>I would like a filter for controversial, having people add: The earth will still exist in 2020, is not so interesting.<p>I normally don't like leaderboards, but in this case, it would be nice if and only if it was weighted by how controversial the prediction is.
Nobody suggest having a "prediction reputation" that keeps track of the quality of people's predictions? Like, after the even should have taken place, open it up for up/down votes or such?
If you haven’t seen it, I recommend predictionbook.com.<p>Honestly, not as nice as what was shared from a UI perspective. However, the user experience on predictionbook.com appears better.<p><a href="https://predictionbook.com/" rel="nofollow">https://predictionbook.com/</a>
Interesting, but there is no legal disclaimer, about page, corporation, cookie usage information, etc<p>Does that mean you don't log in anyway any information about visitor, and that you share all gathered information by contributors with the public ? For example, where can I download the whole database of prediction, with votes, comments and history ?<p>If I can't, I'm assuming this is not fully public and you potentially use my data as a visitor and/or contributor to make money, and you therefore need a legal entity to represent the website.