Intro

Haas Labs is the perfect place for the "mad scientists" who want to find the most optimal settings for their HaasScripts. Currently it only provides the possibility to search and fine-tune optimal parameters for any HaasBot script, but in the future it will expand and offer more tools for you to really get the best out of any script.

Backtest Algorithms

Haas Labs supports multiple tuning algorithms. Each algorithms has their own pros and cons.

Brute Force

The Brute Force algorithm will make and test all possible combinations of the parameter ranges. No combination is left untested.

Pros

• Every possible combination is tested

• Best used for fine-tuning

Cons

• Testing all possible combinations is slow and inefficient

• Adding/expanding parameter search space has exponential effect on the total possible combinations

Intelligent

The Intelligent algorithm should be utilized when the search space goes above 100k combinations (max. limit of Brute Force). Imagine you have script that has too much parameters to start tinkering with; Intelligent algorithm is your best friend.

This algorithm works best on semi-broad and very broad parameter ranges (from over 100k to way above millions of possible combinations, for example 1,290,240,000,000,000), but not so much for fine-tuning. For that, you should consider using Brute Force instead.

Pros

• Can search through very broad parameter ranges

• Can be used for broad "fine-tune"

• No combination limit

Cons

• Doesn't test all possible combinations

• Tends to get stuck around initially best settings with low Adjust Rate

• Does not work well for narrow parameter ranges (below 100k possible combinations)

Random

The Random Search algorithm is nothing more than pure random. Every parameter set tested will be randomly picked from all possible combinations available.

Pros

• Can search through very broad parameter ranges

• You never know what you might find

• No combination limit

Cons

• Doesn't test all possible combinations