KyberDeveloper · Powering Liquidity for the Ecosystem

KyberDeveloper · Powering Liquidity for the Ecosystem

  • Home
  • Integrations
  • Reserves
  • KyberPRO
  • Addresses
  • API/ABI
  • Trade Debugger
  • Tracker
  • GitHub

›Getting Started

Getting Started

  • Introduction
  • Market Making Primer
  • Infrastructure Details
  • Trading Strategy

Tutorials

  • Walkthrough 1
  • Walkthrough 2
  • Mainnet Staging Ready
  • Walkthrough 3

Additional

  • Reserve ID and Rebates
  • Javascript SDK API
  • FAQ's

Market Making Primer

Kyber: Professional On-Chain Market Making Primer

On-Chain Market Making

All Kyber Network liquidity related operations happen on-chain, from the price quoting to the instant settlement. This means that the market makers have to publish their prices on-chain, having the Ethereum chain reaction times.

Ethereum currently has an average block time of about 13 seconds, but this can vary between a few seconds and up to a minute sometimes. Pending transactions have to bid amongst each other via gas prices to be included in the next mined block.

Hence, with the volatility of gas prices and varying block times, the time that price updates actually take effect cannot be precisely calculated.

Kyber Network Matching

Kyber Network uses an RFQ model where the taker requests how many units of source tokens they want in exchange for a destination token.

The network iterates through all the reserves (liquidity sources) for that taker specified quantity and returns the rate from the reserve which has the best rates. Takers can filter in/out reserves in their query. Reserves are either professionals that run their own reserve (FPR), or automated reserves for providing liquidity.

Currently all trading pairs on Kyber Network use ETH as quote currency. When there is a tokenA to tokenB trade, there are two trades, tokenA > ETH and ETH > tokenB, occurring under the hood.

Kyber FPR Reserve

Kyber FPR is the reserve that is addressed to professional market makers, with many key design advantages. It has several differentiations to make on-chain marketing effective and powerful. Can emulate an orderbook effect without committing the inventory, and also has several safety features.

  1. Different pricing logic: Reserve sets a baseline price and then a function can be applied to that price that alters depending on the absolute (setQtyStepFunction) or net traded (setImbalanceStepFunction) amount. This can emulate an orderbook style pricing while keeping gas utilization for pricing very low.

  2. Efficient inventory usage: Tokens can reside in a wallet and can be used only when needed for a match. Inventory is not committed like when submitting a limit order. After the price is set and a taker wants to trade, available inventory is checked.

  3. Safety features: Price can be checked against a sanity price (SanityRatesInterface) and blocked if the configured limits are exceeded. Pricing can be changed according to quantity (setQtyStepFunction) of the trade or according to the net change on token (setImbalanceStepFunction) that this trade will bring. A TTL field can be configured to ensure prices will expire (setValidRateDurationInBlocks) if for any reason the reserve will not be able to push new prices on-chain. Limits of the per-block (maxPerBlockImbalance) or between-updates (maxTotalImbalance) net amount of each token to be traded can be set.

  4. Gas optimizations: Reserves can publish price updates for many tokens in a single tx and gas optimizations make sure that those updates cost as little as possible.

Trading On-Chain: Pros And Cons

One could argue that Ethereum on-chain trading is more beneficial for takers. Any arbitrage strategy has at least two legs and in the off-chain world you are never sure that you can get both legs without some price deterioration.

On-chain a smart contract can check for profitable trade conditions and because of the sequential processing of Ethereum, if the checks fail, no trades will be opened. Also,many fast traders will look for a spike in price on a centralized exchange that hasn't been reflected on-chain and will try to grab the lagging on-chain rate.

On the other hand, there are many Dapps that many times prioritize execution over price and provide a constant stream of trades that can be very beneficial to the makers. This can offset many of the issues with the on-chain pricing latency.

Our own FPR reserve was slightly more than 35 bps in ETH profit per volume traded for the year 2019. The percentage numbers are similar but the volumes are higher a bit before the end of the first half of 2020.

Because bringing liquidity on-chain has a cost, the profit margins are much higher on-chain, and lesser volumes can bring enough profitability to sustain making a business decision to invest in becoming a Kyber FPR reserve.

← IntroductionInfrastructure Details →
  • On-Chain Market Making
  • Kyber Network Matching
  • Kyber FPR Reserve
    • Trading On-Chain: Pros And Cons
KyberDeveloper · Powering Liquidity for the Ecosystem
EssentialsGetting StartedDesign PrinciplesProtocol OverviewSmart Contract Architecture
IntegrationsDAppsVendorsWalletsPlatform Fees
ReservesFed PriceAutomated Price
© 2019 Kyber Network. All rights reserved.
Follow us on Facebook
Follow us on Twitter
Follow us on Github
Follow us on Reddit
Join our Telegram group
Subscribe to our Youtube channel
Read our blog on Medium