Itchy Kinky Ho Charts

(Sorry about the title. It’s my internal name for the charts since I can’t ever remember how to spell the real name.) [Edit: Ok, I looked it up so google could index this article appropriately.... it's Ichimoku Kinko Hyo charts.]

I saw these charts in TAoS&C last year, but disregarded them pretty quickly. It seemed too arbitrary, and also the article said the two main lines were moving averages. Projecting moving averages forward and backward didn’t make much sense to me.

Well, now MWF is interested in them, so I spent some time today taking a closer look. It turns out, the two main lines are midpoints rather than moving averages, and I can see the logic in what they are doing now. It’s really not too much different than moving averages, except by projecting that cloud forward in time they get a nice visual effect.

I’m going to assume you know something about the itchy charts… if you don’t, then check out the article MWF linked.

I think of it like this… as far as I can tell, the three parameters should represent 1/3rd cycle, 1 cycle, and 2 cycles on whatever chart you use. Incidentally, this means that the traditional 9,26,52 are probably not very good choices. On ES I am using 7,21,42 as a start. Maybe I will make it adaptive in the future. Anyway, so given that, it’s clear why a cross of the 1/3rd cycle midpoint with the full cycle midpoint would give you a bullish feeling, and vice versa. It’s kinda like a cycle-width pivot point. So far so good.

Then, they take the average of those two values (though honestly I think it’d work just as well to just use the 1-cycle number, especially given that averaging them seems so arbitrary), and they push that number forward 1 cycle into the future. They also project a 2-cycle midpoint out 1 cycle into the future. The area inside these two points is “the cloud.”

Now, the party line on “the cloud” is that it represents support/resistance. I don’t really buy that. It works no better than a moving average in that regard. BUT, it does make sense that the cloud can help you gauge the trend. Why? Because if you are trending up, then 1 cycle from now you ought to be above the midpoint of this cycle, and you REALLY shouldn’t fall below the midpoint of the last two cycles. See?

So, price relationship to the cloud can help confirm the current trend. Moving averages with a 1-cycle and 2-cycle lag would do the same thing (which is why MA’s with periods around 80 are good at holding price down on ES… it’ll have a lag of 40 candles, or about 2 cycles). Don’t believe me? Here’s a chart with the Itchy cloud, along with 42 and 84 period MAs. See how the MAs and the cloud are very similar? Nothing magic about it.

Ok, so then there’s that line that they project backwards…. what’s that about? Well, they are projecting the current price backward 1 cycle. It stands to reason that, if the current price is above where we were one cycle ago, then buyers are active, and vice versa. That’s all it’s for, as far as I can tell… it adds strength to buy and sell signals if you were already heading in the right direction.

I put the TS code together in a few minutes… it’s very easy calculations. Instead of projecting that one line backward in time, I just put colored dots on top of the primary line. Green means that the backward line was above the price, and red means the backward line was below the price. It was WAY too messy to have a line that looked just like price action, only on a delay. If I want to see double, I’ll get (more) drunk, thank you very much.

For fun, I built a workspace out of the itchy ho charts, meant for scalping. The entry signal is the eotpro.com Patty B signal. The idea is to take the first PattyB after the Itchy lines cross, when the price is in the right relationship to the cloud. I also put on a very quick 5-minute rolling VWAP (and the associated paintbars) as secondary confirmation. I threw it on multiple time frames, as is my usual practice.

You can see a typical play highlighted on the chart. The 3 timeframes all agree about the trend, and we take a shot at a fresh push up on the fastest time frame. I looked over a couple days of data, and it looks pretty decent. Since I can only trade a partial day Tuesday, I may try this bad boy out! :-)