Hey, you resurrected the thread buddy, not me
Had a quick look. These 'fractals' just look a bit like a 3 bar formation representing a local high or low. Perhaps it's more complicated than that.
Not clear that they have anything to do with fractal math though. So someones comment on the thread years back about them being fractal and so they apply to all timeframes seems a bit of a stretch.
Yes, I started it, but one or two actually replied, and that's interesting.
Yes, one of the ways of I have of understanding fractals is 5 bars. If the middle one has a wick or close which goes further up or down than the 2 to the left and the 2 to the right then that would become a fractal and holds until another swing high or swing low exceeds it - and that can take quite a few new candles sometimes.
In a low time frame eg M1 new fractals will be drawn more frequently than in a higher time frame even m15. That's fine, because of the geometry of the market.
I use them to the extent that I would consider to go long if in my chart, be it m5 or whatever, an old fractal gets broken on the upside. Vice versa on the downside, although whether to even trade also depends on the trend, I am cautious about trading against the trend which is measurable by the moving average of lets say 30 bars.
I would confirm my move with regard to a wave formation measured by such as MACD or the indicator which comes with the Bill Williams or Itradeaims systems, as it is useful to know whether one is in a wave 3 or 5 situation, going against the trend usually means you're trading the direction of a wave 4, and wave 5 usually offers the highest potential for a push towards fib extension targets.
You'd need to subscribe to one of the systems to really understand the setups
In summary, what I'd want to say is the system works now, it worked in backtests 2, 5,10,20 years ago, and it seems to be immune from the changing market conditions which kill off so many other algorithms