Brokerages

Kraken

Introduction

QuantConnect enables you to run your algorithms in live mode with real-time market data. We have successfully hosted more than 200,000 live algorithms and have had more than $22B in volume traded on our servers since 2015.

Kraken was founded by Jesse Powell in 2011 with the goal to "accelerate the adoption of cryptocurrency so that you and the rest of the world can achieve financial freedom and inclusion". Kraken provides access to trading Crypto through spot and Futures markets for clients with a minimum deposit of around $0-$150 USD for currency and Crypto deposits. Kraken also provides staking services, educational content, and a developer grant program.

To view the implementation of the Kraken brokerage integration, see the Lean.Brokerages.Kraken repository.

Account Types

Kraken supports cash and margin accounts. To set the account type in an algorithm, see the Kraken brokerage model documentation.

Create an Account

Follow the account creation wizard on the Kraken website to create a Kraken account.

You will need API credentials to deploy live algorithms with your brokerage account. After you open your account, create API credentials and store them somewhere safe.

Paper Trading

The Kraken brokerage doesn't support paper trading, but you can follow these steps to simulate it:

  1. In the Initializeinitialize method of your algorithm, set the Kraken brokerage model and your account type.
  2. Deploy your algorithm with the QuantConnect Paper Trading brokerage.

Asset Classes

Our Kraken integration supports trading Crypto.

Data Providers

The QuantConnect data provider provides Crypto data during live trading.

Orders

We model the Kraken API by supporting several order types, supporting order properties, and not supporting order updates. When you deploy live algorithms, you can place manual orders through the IDE.

Order Types

The following table describes the available order types for each asset class that our Kraken integration supports:

Order TypeCrypto
Marketgreen check
Limitgreen check
Limit if touchedgreen check
Stop marketgreen check
Stop limitgreen check

Order Properties

We model custom order properties from the Kraken API. The following table describes the members of the KrakenOrderProperties object that you can set to customize order execution:

PropertyData TypeDescriptionDefault Value
TimeInForcetime_in_forceTimeInForceA TimeInForce instruction to apply to the order. The following instructions are supported:
  • DayDAY
  • GoodTilCanceledGOOD_TIL_CANCELED
  • GoodTilDategood_til_date
TimeInForce.GoodTilCanceledTimeInForce.GOOD_TIL_CANCELED
PostOnlypost_onlyboolA flag to signal that the order must only add liquidity to the order book and not take liquidity from the order book. If part of the order results in taking liquidity rather than providing liquidity, the order is rejected without any part of it being filled.
FeeInBasefee_in_baseboolA flag to signal that the order fees should be paid in the base currency, which is the default behavior when selling. This flag must be the opposite of the FeeInQuotefee_in_quote flag.
FeeInQuotefee_in_quoteboolA flag to signal that the order fees should be paid in the quote currency, which is the default behavior when buying. This flag must be the opposite of the FeeInBasefee_in_base flag.
NoMarketPriceProtectionno_market_price_protectionboolA flag to signal that no Market Price Protection should be used.
ConditionalOrderconditional_orderOrderAn Order that's submitted when the primary order is executed. The ConditionalOrderconditional_order quantity must match the primary order quantity and the ConditionalOrderconditional_order direction must be the opposite of the primary order direction. This order property is only available for live algorithms.

Updates

We model the Kraken API by not supporting order updates, but you can cancel an existing order and then create a new order with the desired arguments. For more information about this workaround, see the Workaround for Brokerages That Don’t Support Updates.

Fees

To view the Kraken trading fees, see the Fee Schedule page on the Kraken website. To view how we model their fees, see Fees.

Margin

We model buying power and margin calls to ensure your algorithm stays within the margin requirements.

Slippage

Orders through Kraken do not experience slippage in backtests. In live trading, your orders may experience slippage.

To view how we model Kraken slippage, see Slippage.

Fills

We fill market orders immediately and completely in backtests. In live trading, if the quantity of your market orders exceeds the quantity available at the top of the order book, your orders are filled according to what is available in the order book.

To view how we model Kraken order fills, see Fills.

Settlements

Trades settle immediately after the transaction

To view how we model settlement for Kraken trades, see Settlement.

Security and Stability

Note the following security and stability aspects of our Kraken integration.

Account Credentials

When you deploy live algorithms with Kraken, we don't save your brokerage account credentials.

API Outages

We call the Kraken API to place live trades. Sometimes the API may be down. Check the Kraken status page to see if the API is currently working.

Deposits and Withdrawals

You can deposit and withdraw cash from your brokerage account while you run an algorithm that's connected to the account. We sync the algorithm's cash holdings with the cash holdings in your brokerage account every day at 7:45 AM Eastern Time (ET).

Demo Algorithm

The following algorithm demonstrates the functionality of the Kraken brokerage:

Virtual Pairs

All fiat and Crypto currencies are individual assets. When you buy a pair like BTCUSD, you trade USD for BTC. In this case, LEAN removes some USD from your portfolio cashbook and adds some BTC. The virtual pair BTCUSD represents your position in that trade, but the virtual pair doesn't actually exist. It simply represents an open trade. When you deploy a live algorithm, LEAN populates your cashbook with the quantity of each currency, but it can't get your position of each virtual pair.

Deploy Live Algorithms

You must have an available live trading node for each live trading algorithm you deploy.

Follow these steps to deploy a live algorithm:

  1. Open the project you want to deploy.
  2. Click the Lightning icon Deploy Live icon.
  3. On the Deploy Live page, click the Brokerage field and then click Kraken Exchange from the drop-down menu.
  4. Enter your Kraken API secret and key.

    5. Gather your API credentials from the API Management Settings page on the Kraken website. Your account details are not saved on QuantConnect.

  5. Click the Verification Tier field and then click your verification tier from the drop-down menu.

    6. For more information about verification tiers, see Verification levels explained on the Kraken website.

  6. Click the Node field and then click the live trading node that you want to use from the drop-down menu.
  7. (Optional) In the Data Provider section, click Show and change the data provider or add additional providers.
  8. (Optional) Set up notifications.
  9. Configure the Automatically restart algorithm setting.

    10. By enabling automatic restarts, the algorithm will use best efforts to restart the algorithm if it fails due to a runtime error. This can help improve the algorithm's resilience to temporary outages such as a brokerage API disconnection.

  10. Click Deploy.

The deployment process can take up to 5 minutes. When the algorithm deploys, the live results page displays. If you know your brokerage positions before you deployed, you can verify they have been loaded properly by checking your equity value in the runtime statistics, your cashbook holdings, and your position holdings.

You can also see our Videos. You can also get in touch with us via Discord.

Did you find this page helpful?

Contribute to the documentation: