PLS chart, page-15013

Hmm. Probably improve accuracy in the dailies, and might explain why I tend to exit early on the multi- day runs (which is how I can loose tens of thousands quickly - when I trade out early then trade back in blind at a higher price because it didn't turn when I expected it to, but then f*'n reverses straight after I trade back in) but intra-day will be a very different story. My approach gives about 90% accuracy on the first wave to third wave intraday (haven't actually tracked it, but when I miss it is a big surprise) and usually matches the prediction from the old flag pole method. The number of times I get the top of the third wave exactly right intra-day I can't count, and if it misses the target on the third wave it almost certainly gets it on the 5th (eg today where it hit the 160% target on the 5th wave but not quite the third - which of course also means the pattern was not an EW pattern). Interestingly, when you do it in reverse on a falling intraday wave it picks the bottom incredibly well - at least for the 3 step correction waves.

So it looks like intraday there are a slightly different set of rules applying and we need a modified EW theory - because using wave 2 as the start for the third wave projection would result in the price being set too high, given my method usually gets the price exactly right. I will however experiment and see whether there ends up being a practical difference between the two intraday, where we are talking whole cents and much smaller ranges (compared to daily movements over tens of cents), which thus has a built in rounding error, that may result in the two approaches yielding essentially the same result when rounded to the cent.