Matthias c4fb0fd6ca Don't run the bot with python3 freqtrade

* we can either use `python3 -m freqtrade ...` or `freqtrade ...` - and
shorter should be better.

2019-07-08 17:01:25 +02:00

3.5 KiB

Raw Blame History

Freqtrade FAQ

Freqtrade commons

I have waited 5 minutes, why hasn't the bot made any trades yet?!

Depending on the buy strategy, the amount of whitelisted coins, the situation of the market etc, it can take up to hours to find good entry position for a trade. Be patient!

I have made 12 trades already, why is my total profit negative?!

I understand your disappointment but unfortunately 12 trades is just not enough to say anything. If you run backtesting, you can see that our current algorithm does leave you on the plus side, but that is after thousands of trades and even there, you will be left with losses on specific coins that you have traded tens if not hundreds of times. We of course constantly aim to improve the bot but it will always be a gamble, which should leave you with modest wins on monthly basis but you can't say much from few trades.

I’d like to change the stake amount. Can I just stop the bot with /stop and then change the config.json and run it again?

Not quite. Trades are persisted to a database but the configuration is currently only read when the bot is killed and restarted. /stop more like pauses. You can stop your bot, adjust settings and start it again.

I want to improve the bot with a new strategy

That's great. We have a nice backtesting and hyperoptimizing setup. See the tutorial here|Testing-new-strategies-with-Hyperopt.

Is there a setting to only SELL the coins being held and not perform anymore BUYS?

You can use the /forcesell all command from Telegram.

Hyperopt module

How many epoch do I need to get a good Hyperopt result?

Per default Hyperopts without -e or --epochs parameter will only run 100 epochs, means 100 evals of your triggers, guards, ... Too few to find a great result (unless if you are very lucky), so you probably have to run it for 10.000 or more. But it will take an eternity to compute.

We recommend you to run it at least 10.000 epochs:

freqtrade hyperopt -e 10000

or if you want intermediate result to see

for i in {1..100}; do freqtrade hyperopt -e 100; done

Why it is so long to run hyperopt?

Finding a great Hyperopt results takes time.

If you wonder why it takes a while to find great hyperopt results

This answer was written during the under the release 0.15.1, when we had:

8 triggers
9 guards: let's say we evaluate even 10 values from each
1 stoploss calculation: let's say we want 10 values from that too to be evaluated

The following calculation is still very rough and not very precise but it will give the idea. With only these triggers and guards there is already 8*10^9*10 evaluations. A roughly total of 80 billion evals. Did you run 100 000 evals? Congrats, you've done roughly 1 / 100 000 th of the search space.

Edge module

Edge implements interesting approach for controlling position size, is there any theory behind it?

The Edge module is mostly a result of brainstorming of @mishaker and @creslinux freqtrade team members.

You can find further info on expectancy, winrate, risk management and position size in the following sources:

3.5 KiB Raw Blame History Unescape Escape