From an EW perspective, I was calling a major wave ending in real time. See:
-- May 2
-- May 5 (and a Long Term View Update)
-- Regarding Tops and Sloppy / Misleading EW Practices
So while I had the expectations of a top, my system doesn't care about expectations, it only trades the trends present in the market. Same deal with the most recent bottom in October, while my expectations are that it is a significant bottom (see Revisiting the Large Count), my system doesn't trade expectations.
Here is the performance on the SPY:
What is interesting to see if you dig into the data is that it gave correct signals for the two biggest waves since May. It held both the July and October trends for 10% gains each, one being a short and one being a long.
And that is the philosophy I used when building the system. Try to find trends and give them the leeway to develop, but to cut them early if the entry setup turns out to be a fakeout. And the average winning trade rate is only 50%, but the average winning trade is much larger than the average losing. (And while this is only a 6 month sample size, the same pattern holds true on a backtest to 2001).
What I like about my system is that it consistently gives me an edge, especially when EW counts are ambiguous. And it helps me sort out what is happening in reality vs. my expectations: Regarding Trading Positions and Long Term / Macro Outlook
