Decentralized sports & event betting platform powered by Ethereum smart contracts and automated market making.
New betting events will initially be created by the founding team by calling a createEvent method on the smart contract. In the future, this event data could be requested from a Chainlink oracle.
Any Ethereum wallet can place bets on an event by calling the placeBet method on the smart contract and sending a bet option and wager amount in the form of ETH tokens.
This process will be facilitated by a web app which displays all events eligible for betting and allowing users to claim payouts for won bets.
Users will transfer funds and interact with the smart contract by connecting a Metamask wallet.
-
Apply the proof of stake model to outcome reporting. Automating result reporting is not scalable, brittle, and requires some level of trust. The best way to get dependable results is to marginally incentive someone to answer honestly and massively penalize them for acting in bad faith. It's a self governed system that requires no belabored software automation efforts.
The idea is that when an event ends, someone can report a result via an offchain oracle. When reporting the result, the reporter must stake an amount that is more than the sum of the bets for that event. After a result is reported, the report can be disputed or verified by other staked reporters. Once a predetermined verification threshold is met, the event can be confidently resolved. The reporters get a small reward for reporting honestly. If they answered dishonestly, their stake is forfeited and used to correct payout amounts.
A clear downside to this approach is that an event pool may contain a large sum of money. It would be unreasonable to require a reporter to stake potentially hundreds of ETH just to report an event result. One approach to overcome this is to break the pool into sub pools and have multiple reporters that stake the amount relative to their pool size.
Pros: self-governed, fast event resolutions, trustless, scalable Cons: requires motivated users to report/dispute results, requires a user to be flush with liquidity to report results
- Event ends
- Some percentage of nodes on the SBP network are randomly chosen to be event result voters
- Each voter calls a smart contract method to request a randomly generated proof-of-work assignment, which is used to force the voter to find some random nonce that they will reveal in a later step for vote verification purposes:
contract.requestProofOfWorkAssignment(eventId) => proofOfWorkAssignment- Voter completes the proof-of-work assignment locally, recording the nonce value discovered to fulfill the assignment (The necessity for this nonce will also be explained in a later step):
proofOfWorkAssignment(nonce) => proofOfWorkFulfillment- Voter executes a local hashing function, passing their vote and the nonce (discovered in step 4) as inputs, and signing the output hashed data using their private key. This signed data acts as a vote verification key, which will be explained in a later step:
sha256(vote, nonce) => hashedData
createSignature(privateKey, hashedData) => voteVerificationKey- Voter calls a smart contract method to cast their vote, passing the signed vote verification key produced in step 5:
contract.castVote(voteVerificationKey)Since this voteVerificationKey is a signed hash of the voter's vote and a privately held nonce, it does not reveal anything to public viewers (including other voters) about which way they voted, but it can be used later to ensure that the voter did not change their vote after others revealed their votes.
- Once a threshold of votes is reached on-chain, voting is closed and the smart contract will reject future votes for the event. This threshold is necessary to incentivize voters to vote quickly -- selected voters who do not vote before the threshold is reached will have their vote rejected and will thus not be able to collect voting rewards.
- After voting closes, voters can call a smart contract method to reveal their vote and discovered nonce:
contract.revealVote(vote, nonce)- The smart contract can verify the integrity of each vote (and ensure the voter did not change their vote after others revealed theirs) by checking the proof-of-work nonce and hashing the data (in the same way the voter did in step 5) and passing the output hashed data through a signature verification function:
proofOfWorkAssignment(nonce) => proofOfWorkFulfillment
sha256(vote, nonce) => hashedData
verifySignature(publicKey, hashedData, voteVerificationKey) => booleanSince the data returned from smart contract calls is publicly viewable, a nonce is necessary to prevent voters from easily brute forcing the hidden votes cast by other voters by just guessing through every possible vote and checking to see if that vote could be verified using their public key.
The nonce requires brute forcers to complete a proof-of-work and arrive at the identical nonce discovered by the voter, before they gain any information about their vote. Gaining access to other votes before casting a vote could cultivate consensus corruption, so it should be infeasible for one node to complete enough proof-of-work assignments assigned to other voters to gain a large amount of information before casting their vote.
// this is too easy
sha256(vote) => hashedData
createSignature(publicKey, hashedData) => boolean// this requires computational work and luck
proofOfWorkAssignment(privateKey, nonce) => proofOfWorkFulfillment
sha256(vote, nonce) => hashedData
createSignature(publicKey, hashedData, voteVerificationKey) => booleanA vote guessing attack becomes even less financially feasible due to the voting punctuality incentivization facilitated by the threshold defined in step 7.
-
The smart contract discards any votes that were rejected by the
verifySignaturecall in step 9 (i.e. votes that were changed after reveal) and counts the remaining votes to determine the winner. -
Voters who voted as part of the majority will receive a voting reward. This reward could be in the form of a newly minted governance token, a cut of the winners' pot, or a combination of both.
-
The winning result is stored in the smart contract, unlocking the ability for winning bettors to claim payouts.
(need details here)
In order to create a highly liquid market, the platform will incentivize users to stake funds into a liquidity pool, which will be used to payout winning bets. Having a highly liquid pool ensures that, no matter which side wins a bet, there is enough ETH in the pool to pay out all the winners, even if the platform loses money overall.
To incentivize stakers, the platform can offer a percentage cut of the platform fee to the liquidity pool, where each liquidity provider earns a staking reward in proportion to the amount of ETH they are staking relative to the overall pool.
With this model, early stakers will potentially make up a large percentage of the liquidity pool and therefore earn high rewards, but as more stakers add liquidity, it will lower the early stakers' earnings, as they now make up a smaller percentage of the liquidity pool.
This creates a self-correcting system -- in scenarios where there is a low amount of liquidity, people will be more incentivized to stake because rewards will be higher; in contrast, in scenarios where there is excess liquidity, rewards will be lower, incentivizing stakers to exit the liquidity pool and deploy their capital into something more productive.
One of the main project challenges is defining the initial payout odds for an event, which in traditional sports betting is usually done by a combination of proprietary computer algorithms and "experts".
These initial odds could potentially be informed using Chainlink oracles to get the initial sports betting odds defined by Vegas and global sportsbooks.
Another approach is to allow anyone on the platform to create a new betting market for an event (for which the odds will be auto-adjusted as bets come in from both sides), and staking an amount of ETH into the pool to sufficiently cover payouts for either result.
With this model, odds makers will be incentivized to set the initial odds to align as closely as possible with the real world expected outcome in order to maximize their profit by having a balanced book; this is reflective of how most traditional sportsbooks make money.
The platform can move betting pool funds into a decentralized lending platform like Aave or Compound and earn interest from borrowers. This passive interest earning can be passed off to users of the platform in the form of better odds payouts and/or a betting rewards system.
The platform can incentivize users to place bets by rewarding every placed bet with a number of reward points proportional to the size of the bet. Points could later be redeemed for ETH by calling the redeemPointsForEth method on the smart contract. These points could be tracked directly on the blockchain in a mapping of address => pointsEarned. Alternatively, the rewards points could be tokenized, allowing users to trade rewards tokens freely across wallets.
The platform will initially be bootstrapped and governed in large part by the founding team. As the community grows and the platform matures, governance will be incrementally transferred over to the community in the form of a decentralized autonomous organization (DAO).
Holders of a platform-minted ERC20 governance token may cast votes to adopt and reject new proposals to the platform protocol.
The governance token could be distributed as a reward to odds makers for maintaining platform events, proportional to the number of events they maintain relative to the total number of events on the platform.
Clone the repository:
$ git clone https://github.com/angelle-sw/sbpCreate a .env file in the root of your repository (You can use Infura to spin up a test endpoint):
TEST_ENDPOINT=<endpoint>
WALLET_MNEMONIC=<wallet-mnemonic>
WALLET_INDEX=<wallet-index>Install dependencies:
$ yarnCompile contracts:
$ yarn compileBoot up local Ganache network:
$ yarn blockchain:localDeploy contract to local Ganache network:
$ yarn migrateDeploy contract to Ropsten testnet:
$ yarn migrate:testnetRun tests:
$ yarn testRun tests and report test coverage:
$ yarn coverageGenerate types for tests:
$ yarn types:truffleGenerate types for frontend:
$ yarn types:ethersChange into client directory:
$ cd clientStart app in development mode:
$ yarn startBuild app:
$ yarn buildLint app:
$ yarn lintRun tests:
$ yarn test