Frequently Asked Questions (FAQ)
Before creating any order (using trading simulator or real trading), OctoBot asks the exchange for its minimal (and maximal) requirements for any order. When creating an order (following a buy or sell signal), these order requirements are checked. If the order is not compliant, it will not be pushed to the exchange.
The most common case of signals without created orders is when there is not enough funds of the required asset to proceed with an order.
Example: not enough USD to buy BTC for a BTC/USD buy signal.
In trading simulator and backtesting modes, OctoBot uses a simulated portfolio called
"starting-portfolio"that is defined in the trading simulator configuration.
It can be once in a week or 5 times a minute, this depends on the strategy your OctoBot is using.
For example: when using the default settings, the simple mixed strategy evaluator is using the 1 hour timeframe as the shortest one. Since it's a technical evaluator based strategy, it will update every hour. In this setup, your OctoBot will create new trades every hour it sees an opportunity. There might be hours with no opportunity and no order creation.
This is probably due to an issue in your tentacles folder. Try removing it and restarting your OctoBot, it will download the latest versions of each tentacle and should fix the problem.
When your OctoBot places an order or has a order that is filled, it will appear on the web interface. The web interface displays the list of open orders and the list of filled orders.
You can also receive Telegram and soon Discord notifications on orders placement and trades.
OctoBot will consider it can trade 100% of the portfolio you give it. However how this funds will be used (size of orders, orders frequency, ...) depends on your risk setting and the trading mode you are using.
Note: In 0.3.X, when the reference market is changed during a backtesting, its value will always be 10000. This changes in 0.4.X where it takes the value in starting portfolio.
The reference market is automatically switched to the base of the traded pair in backtesting to compute more accurate profitability.
Example: a backtesting on ETH/BNB would make BNB the temporary reference market for this backtesting.
For now, OctoBot uses all the available funds to trade. Therefore it's possible that 100% of the exchange funds on an account will be traded.
OctoBot backtesting is always using the maximum available data allowing to keep a realistic simulation.
However exchange are usually not giving all of their data: they give the last X candles (500 for binance). Therefore a regular backtesting data file has 500 1hour (1h) candles, 500 1minute (1m) candles etc. These candles are always the most recent ones. That means that when running a backtesting on 1h and 1d time frames, the maximum backtesting range is not 1h and 1d with 500 candles each but the time range where both 1h and 1d have data: there the past 500 hours (500 1h candles and approximately 20 1d candles).
As an example, in a backtesting with 1m and 1d candles: the common time range in 1d is
500/(60*24) = 0.35which means the whole backtesting is carried out with the data of one day: the last daily candle of the 500 1d candle only while using 100% of the shortest time frame: 1m (which all happened during this one day).
This error (or something very similar) might appear in your OctoBot's logs:
<function _ProactorBasePipeTransport.del at 0x000001064DE8A310>
Traceback (most recent call last):
File "asyncio\proactor_events.py", line 116, in del
File "asyncio\proactor_events.py", line 108, in close
File "asyncio\base_events.py", line 719, in call_soon
File "asyncio\base_events.py", line 508, in _check_closed
RuntimeError: Event loop is closed
This is a minor issue with the current Windows implementation of the asynchronous libraries OctoBot is using. It has absolutely no effect and can be completely ignored.