Put Your Money Where Your Mind Is: 
EtherSphere v2.0

TL;DR In a short 27 day trial, EtherSphere v1.0, the blockchain based game, received $150,000 in bids. It’s instant success was due to a combination of game theory, excellent design, and a website packed with useful tools. Now the creators are gearing up to release v2.0 and it is gathering interest.

Games can be broken into two clear categories, those that include an element of chance, and those that depend entirely on the skills of the players. The random order in which cards are uncovered always makes card games a game of chance. This is true even though skill is frequently the decisive factor between winning and losing. Whereas board games that don’t use dice or turned cards are typically games of skill.

It is also possible to separate games of skill into two groups, though not as cleanly. Some games, like chess, are won by a combination of problem solving and memory. Others require little in the way of problem solving and rely heavily on bluffing or predicting the other player. One of the purest forms of these games is Paper-Scissors-Rock. To win this game more than probability would indicate, you need to understand your opponent so well that you can predict what they will do and win using choice.

Game Theory

The science behind getting into your opponent’s head and basing your move on what you think they will do is known as game theory. And while it might seem academic, the reason people love a game of skill is usually because it is designed to enable the use of game theory to win. And we benefit from playing these games because we also use game theory whenever we negotiate. A form of play-thinking that prepares us for real life.

The team behind the online game EtherSphere knew this and used their extensive understanding of game theory to develop their offering. Part of its charm is that it is incredibly simple on the surface, but enables the player to really delve into game theory for an advantage. But before we get into the game mechanics, a quick overview of how it relates to cryptocurrency is required.


EtherSphere is a distributed application, written in Solidity, and running on the Ethereum network. Using this technology as the foundation of this service has made development and release much simpler than could have been achieved previously.

Until now, the only way to be certain of what code you were running was to compile it yourself before installing it. By leveraging off Ethereum exposing the code, and by using understandable variable names, EtherSphere was able to gain the trust of players.

Betting on Yourself Isn’t Gambling

So to play a round in Ethersphere you have to have some Ether to put into it. Ether is the name of the currency that is traded on the Ethereum network, and it is similar and tradeable with Bitcoin. But just because it requires putting some ether into a location, and just because the winner of a round can get quite a lot more ether back out, does not mean it is gambling.

People who pay money to join a darts game or put money on a game of pool that they are playing would never call it gambling. The reason for this is that these are games of skill, not chance. It is generally recognized that gambling involves an element of chance, and the code behind EtherSphere leaves nothing to chance. And the rules published on their website can be matched to the code stored at the address.

The website also published details about the game including the highest aggregate bid. Adding to this transparency is the fact that the published address where you and all of the other players put their ether can be explored. So at any time, a player can see how much someone has put into the game as soon as the transaction is confirmed. For privacy, all they see are the transaction details including the from address on the blockchain. But this still tells a player everything they need to know.

EtherSphere v1.0

Before we talk about the upcoming release of version 2.0, let’s have a look at the original version. On Sunday, May 21, 2017 EtherSphere closed its doors for the last time on their prototype game. During its short twenty-seven days of operation it received US$150,000 in bids and paid out almost as much in rewards. As a proof of concept, it was incredibly successful and left players ripe with anticipation for their promised version 2.0

Despite a number of interdependent rules, the game was clearly a success. Part of the reason for this was that the developers took great care in making the game easy to understand. Their website not only explained the rules, but provided a trial sandpit in which to experiment without investing any money. They also provided a live calculator to explore the effect of an action being considered. And because they knew how important communication is to game theory, a chat window was built in. This clarity, openness, and concern for the player experience gained them a lot of trust, and made the game more fun.

The Game Play

To play the game, all you had to do was send a value to their provided address on the Ethereum blockchain. This single, simple action meant that you were in the running to receive a proportion of the value that exists in the reward pool. The money that you put in goes into the bid pool.

So the reward pool happened to have 10 ether in it for this round. First comes Sally, who puts 2.5 ether into the bid pool. Then Tom puts in 2 ether, and Jane puts in 0.5. Then the game is declared over at the end of the day. Their combined bids meant that the bid pool contained 5 ether. So Sally who put in half the value of the bid pool, gets sent half the value of the reward pool, 5 ether. Jane only put in 0.5 ether or 10% of the bid pool. So she gets sent 10% of the reward pool, being 1 ether. And Tom gets 4 ether for the 2 he put in.

They all doubled their ether and it all sounds too good to be true. But this can happen, and did quite often during version 1. But here’s the trick. If too many people bid too much, then the bid pool becomes larger than the reward pool. This would mean that a pro rata payment from the reward pool would be less than their original bid. Imagine the Bidding Pool in the above diagram being larger than the Reward Pool so that each share shrinks before payout. Everyone loses.

And here comes the game theory. Who would do that? Who would look at the game, with all of its current bids available for inspection, knowing that the bid pool is getting close to being larger than the reward pool? Nobody who wants to hang onto their money. So the people bidding should be confident that nobody will ever bid so much that they all lose. Once this realization occurs, where nobody can move without hurting their own chances, we have established what is known as a Nash equilibrium.

Now John Forbes Nash, Jr. was a smart man and he understood people. He knew there were fools who do irrational things, or people who thought they understood something when they didn’t. So he added, that a Nash equilibrium only exists when those involved are rational, working in their best interests, and understand the rules. So just like Nash predicted, sometimes when it was best to bid no further, people did.


If the developers had wanted to, they could have left it at this, but they wanted a highly playable game and added some extra spice. They introduced a jackpot to encourage people to act against the group and try to get more for themselves. But not all of it, because EtherSphere cared about the other players too. So they set the jackpot to hand out 35% of the reward pool to the daring and greedy one who brought the castle down for personal gain. And left 65% to be shared according to the proportion of their bids, including the person who bid highest for the jackpot.

At first glance, it might seem like going for the jackpot is as easy as bidding more than the rest, once the total bid pool size starts getting close to the reward pool size. But nothing is ever that easy. Someone else will instantly see your total bid amount and be able to bid higher. But there’s a logical limit. The reward pool is only so big, and no matter how much you throw into the bid pool, the reward pool doesn’t get any bigger in that round. So just like the initial limit after which not everyone wins, there is a second limit after which nobody wins. Even if you win the jackpot and a lot of the reward pool, your total payout can be less than your total bid. That too would seem a stupid thing to do. But what if that meant you didn’t lose so much?

By this stage you can see how there are layers of game theory at play here and more than one point where a Nash equilibrium can be established. And that is where the success and allure of the game lay. It was fun, exciting, occasionally profitable, and the outcome was always a result of the players’ decisions. With everything going so well, you might be wondering why it was shut down.

Version 2.0

The developers are confident that this game can be polished into something even better. The original game ran over twenty-four hour instances. This favored the players who were up when the game ended and was reset for the next day. Version 2.0 will run two twelve hour games a day. This means that there will be a convenient time to watch the money roll in for people in an opposing timezone. And as you might have realized, once the bid pool outgrew the reward pool, people who had more money to be used for fun had the advantage. So a dynamic capping system is on the way to limit bid sizes.

The jackpot system is also in for a big overhaul. Instead of the jackpot diminishing the reward pool, it comes from its own pool. An existing player suggested that this pool should roll over each game when it isn’t claimed, so the temptation to go for it grows until it can’t be resisted. An excellent idea which is now being incorporated. And by not diminishing the reward pool, it seems like there’s an even greater chance of turning a profit and coming out lucky, or should we call that clever?

A Piece of the Pie

Either way, the game dynamics are certain to change as a result. At times like this, it is common to wonder why you didn’t think of this yourself. Or just as likely, wonder how you can get in on the action. The developers are planning on rewarding players, even when they lose, with a part of the action. And it looks like there’s even going to be an option to buy into a share of the house takings directly.

So with the old players eagerly awaiting the new release, and new players lining up because the capping system levels-out the playing field, it looks like EtherSphere are on the verge of bringing to the market a very interesting new form of entertainment. And sitting on the fringe where the US$40 billion online gambling industry intersects with the US$100 billion gaming industry, their potential for growth is astounding.

To learn more, you can visit the v1.0 website, even read some of the final chat window, or hop on to Reddit to join the discussion about the future.