Trailing Strategy SL not working?

[bravegag]

Hi all, I've copy pasted the TrailingStop example from the documentation but I'm not understanding why the stop loss doesn't get activated and have massive losses. The example I copy-pasted is the SmaCross from here:
https://kernc.github.io/backtesting.py/doc/examples/Strategies%20Library.html

but I get massive drawdowns meaning the SL is not offering a true downside protection. How can I change the TrailingStop Strategy's SL to make it as tight as possible?

I am now using Miniconda 25.1 with Python 3.12 and backtesting 0.6.3

UPDATE
In general, I find the library very neat but the results counter-intuitive, I have a signal that gives me 70-80% precision with really good classification prediction and still losses almost everything while backtesting it ... the common wisdom says if you are able to build a signal with 52% hit rate you should be able to make money in trading. I am struggling to prove that wisdom right ...

Look at this beauty but overall I lose nearly all the initial equity: