Skip to content(if available)orjump to list(if available)

Show HN: Have fun betting virtual (not real) money on predictions from HN users

Show HN: Have fun betting virtual (not real) money on predictions from HN users


·November 24, 2022

I wanted to see how difficult it would be to build a web app using a sub-$300 android smartphone. Decided to build a fun predictions website where you could bet virtual (not real) money on predictions made by others, and also make predictions of your own.

Building it turned out to be considerably easier and more fun than I anticipated.

Primary tools used were:

# QuickEdit as the mobile code editor (Note: the free version of the QuickEdit app is riddled with ads, it shows an advert each time you close a tab, but it unfortunately had the best UI of the 3 or so Android code editors I tested. Ended up using NetGuard to block it from retrieving & displaying ads),

# PHP for the Backend ( custom PHP microframework I've used and built on over the past few years ).

# jQuery for the frontend js (cringing) - it appears I'm simply too lazy to learn React/Vue/et al. Every once in a while, I pick one of them to learn, but I always end up returning to jQuery - or time-permitting - amateur level vanilla JS.

# Bootstrap for the CSS - Battle-tested. For a purely backend dev with minimal design skills, good ol' Bootstrap (and in a growing number of cases, Tailwind) is always a life saver.

# Whole thing is hosted on 2 VMs (1 hosting the web app, and 1 hosting Redis & MySQL).

# As to the site itself, it turned out to be pretty cool to play around with. Go there, view the predictions, bet on the predictions you believe will come true, or against the ones you think will not. You get $50,000 to bet with (not real money). No signup is required to bet, but a quick signup is required to make a prediction. Hope you guys like it, and please be ruthless in telling me of any bugs you've found.

So go on here =>

And Have Fun!


Interesting site. How do you decide on the result of a bet? Where do you get the truth? What if the bet is described ambiguously? Or worse yet, something like "This prediction will be false".


I think the "market creator decides" works really well for play money markets. doe this. The risk of a "rug pull" is often discussed when assigning probabilities, and you can see people's history to decide if you trust them.

A good market will have a well thought out process of deciding the truth. For stuff where there is a betting market or money market, it is easy to refer to that. Otherwise for something like "Will Russia invade Ukraine", at the time the definition of "have they invaded yet" had to be figured out, especially as no one knew how far it would go.


Picking your brain for a slightly related question: how would you propose setting the initial odds, assuming the market creator can also bet?


Big question!

(And sorry I might have been confusing. I am not the OP or creator of the Show HN, but I jumped in and answered how I would do it)

First decide if you want the bets to be taken by an algorithmic market maker (so anyone who wants to place a bet can place it, but they get slippage, like Uniswap) or an order book (more like you say "Ill bet $500 at 33%" an then someone else says "I'll take the other side", like Betfair)

Let's assume you use a AMM. Then the way to do it is let the market maker pick the probability, and provide the initial liquidity to the AMM at those odds. So they are incentivized to get it right. To derisk them a bit (after all they are taking bets with the least information, right at market open) you give them commission or something like that on bets placed.

These are complex decisions. went though various iterations. They are erring on the side of simplicity for new users, so they fix the odds. You don't get a choice, it is 50%. Because it is an AMM you quickly revert to the right probability. Of course, since you time the creation of the market you could make those correcting bets yourself just in time.


It doesn't allow making prediction to 2030 so I just leave it here:

I predict that by 2030 none of the top 10 US car manufacturer will offer self driving in their cars (except for simple autobreaking). Even those manufacturers that offer some self driving now will stop offering it by 2030.


I think it’s unlikely we’ll see level 4 driving in fewer than 10 years, and I don’t think we’ll ever see level 5. It’s a really hard 80/20 problem. I think most peeler are going to be pretty happy with what you can get in a lot of vehicles today.

Disclosure: I worked for an Auto OEM but didn’t have any knowledge or communications with any folks working in the space.


...and how likely do you think this outcome is? 60 %? 90 %? Somewhere in between?


I'm interested in the argument for this!


When your neighbor Joe in his 1987 Buick run over your cat, you know he has 2 mortgages on his house and wife with cancer. When a self driving car runs over your cat or ding your fence post, you will gladly sue the multi billion dollar company, just for the settlement money. It will unleash the neverending stream of lawsuits and companies will be forced to stop offering self driving or go bankrupt. Yeah, they can afford best lawyers, but not 500 new lawsuits every day.


There was a case with Ford in the 70s were they calculated that a retribution cost per person would be around 200k - calculated the likely effect of the cars busting into flames because they had an design flaw. They realised they are still better off than to recall all vehicles.

As the story illustrates, there might be situations were expecteded litigation is worth it. I'm not saying this is morally right, but companies will try to maximize their profit anyway, even if there are (calculable) risks involved.

In the case of self-driving this might mean being a profitable car manufacturer vs becoming a niche, enthusiast car producer for people who still enjoy driving as a leisure.

Here is one of essay on this:


Wouldn't car companies just lobby for some law changes that would make the "drivers" responsible rather than the manufacturers?

Or alternatively, allow self driving only on highways and such, where it's much safer?

I don't think progress is stopped that easily


I'm not sure the capability for FSD will be ready by 2030, but I don't think these reasons are compelling. Insurance will solve these problems


There is _literally_ nothing wrong with your javascript on the page, FYI. stick with your toolset.


Sure, for an mvp. Quick and clean output

But what happens as you start adding stuff. Toaster notification, user badges, swipe actions.

I an not a JS expert, but it seems a modern library will have these built in ?

EDIT: would be interesting to hear the original poster's opinion


> But what happens as you start adding stuff. Toaster notification

Sounds like an amusing corollary to Zawinski's law. "Every program attempts to expand until it can notify you your toast is done..."


Npm would be a 1000x increase in codebase size… for notifications.


Not original poster, but I also sort of don't understand the question. You do what you would do to evolve any MVP: swap out one part at a time to something that scales better. It's just a frontend.


What if someone posts a very likely or unlikely bet (e.g. "the next Mega Millions numbers will be ..."? Can everyone bet $30,000 against and then everyone gets $30,000?

You may want a system where, the more skewed a bet is, the less money gets paid out by the majority side and more money gets paid to the minority. That way, bets like those don't really get you anything, unless you bet the unlikely case and win, then you get a well-deserved jackpot.


A parimutuel system would solve this. When settling the bet, the pool of money is divided among those that backed the winning combination, in the ratio they contributed. You can also display the odds based on how much money is behind each combination.


You lot discussing this would get a kick out of:

I wish I could post a link, but in their Discord they had some really interesting chats and I think they pretty much invented a novel AMM for this. They are open source, so you can take a look here:

I forget whereabouts in that repo the algorithm is though but have a hunt!


Hey! Austin from Manifold here, thanks for the links! I wanted to note that our novel mechanism which we've termed "Maniswap" is indeed pretty cool, but for most intents and purposes a simpler Uniswap constant product market maker would already do quite well as a mechanism, with the advantage that it's really simple (explainable on the back of a napkin!)

Happy to chat more about prediction markets/AMMs, feel free to reach out to


One characteristic of a parimutuel system for a relatively illiquid market is that the odds can swing quite drastically whenever anyone bets. In fact, even the amount one person bets is going to affect the odds they will get, so you need to display a table of odds for various bet sizes.

This may or may not be a problem, but I've found it requires more... infrastructure (notifications etc) for people to avoid being confused by it, and give people a chance to withdraw their bet if the odds are no longer favourable (alternatively put in stops, but that also complicates the system.)


The issue is that this discourages popular bets, not just heavily-biased ones.

If you divide by ratio, and add a fixed number (e.g. 2) to the "for" and "against", there aren't really false positives, and true negatives (e.g. a reasonable bet where people just happen to all bet one side, or a biased one that doesn't get much attention) are unlikely - unless there are too many bets or people are gaming the system, that is.


I don't think I understand.

This pari-mutuel system (I always knew the system but did not know the name) seems near canonical to me. And the incentives seem clear cut. It seems to me that if you feel the odds for or against any prediction are wrongly calculated, then if you submit money to push those odds towards the correct ratio, then you should, on average, expect a return.

... I maybe should actually run simulations to check that, that's just what my intuition is telling me as to how it would work.

I'm trying to understand what you are proposing and the motivations for it.

By ratio, I assume you mean the ratio of the pool for and the pool against? So if prop A has $100 and prop B has $10, then the ratio would be 110/10, or 11:1. So your procedure would be to add a fixed number to this? Either 112/12 or 13:3? or perhaps you meant Prop A $100/(110/10)+2 and Prop B $10(110/10)+2

  So Prop A is ~11.1
     Prop B is  ~2.9
Am I understanding this right? What do we do with these numbers?


This assumes no one is trying to make money. In practice, this is virtually never the case.


That's a pretty good point, though in this case the 'winnings' are simply a fixed multiple of the virtual amounts bet.


A quick way out of this issue is to ask users not for how much to stake, but how confidently they believe something to be true, and then reward/penalise based on a score function derived from this and the actual outcome.


Order by most active bets at the top.

Right now the page starts with dead claims and you find pretty active and good ones while scrolling down.


I can't not make this joke: Where do I send my BTC?


Is there a good reason to use this over Manifold (


The difference afaict is that Manifold both requires a google account and is effectively just using cents as an abstraction over USD. In comparison the bets in the submission are "truly" based on virtual, nin-exhangable money. Additionally, the submission is focused on HN in comparison to general bets like Manifold.


You can't get the money out, it all goes to charity. So while you're not betting free money, you are not betting your money either, it is consumed before you bet either way.


Seems like Manifold requires a Google account which is definitely a mark against it for many


I predict this platform of yours will still be up and running in a year from now.


Can't count the number of times I've wanted something like this for Hacker News!

There have been quite a few hot takes (especially about Apple) where I would have bet against that person's prediction happening. Glad it's an option now.


Consider letting users add an explanation for each prediction so that we know the reasoning behind it.


It's this just mining HN hive mind for "insider" information?


The insider information being the wisdom of the crowd. "Outsider information" would be a better description.


even if it was, everyone would benefit from the information (it being public)


I predict that your site will be renamed as HNLadsBroke in 2023!!!