I've got another thing that's been bugging me, and wondering if other guys have the same problem. That is, scoring charts for products that trade round the clock. On gold for instance, I'm using a 5min OR beginning at the 8:20 EST pit open. I find myself pausing when action later in the evening hours confirms something, and asking myself is it right to give this a 2 score? It seems too far away from that small 5 minute window that was 12 hours earlier. One example, on June 24 it failed at the A down and drifted sideways. But then confirmed an A down around 11:00 at night (eastern time). So I wondered whether or not to give a -2 to that day. Another day it may be a failed A down and then make an A up at 1:00am the next day. So I think too late to give a +3 day, but then consider if I lived on the west coast it would be 10:00pm and not too late for that days score. So I struggle with when to officially end the day for scoring purposes. It's so much easier for stocks because they have a distinct closing time where they really do "close". I'm tempted to simply end the day at the 5pm NY time rollover when most charts start printing a new daily bar, and then create a separate Asian session OR in case any big moves happen then. Any advice?
I try to keep everything relative. So I use the same times. I want to know how Gold is acting during US hours. Because that is where I am watching the spoos and AAPL and Copper and Bonds, etc. Relative performance is very important to me because it gives me more info than looking at something in isolation. Having said, I did find a way to create another indicator that corrects that for that (from the ACD playbook). So that acts as a modifier for 24 hour markets that takes into account the 24 hour cycle.
Yeah, I had read some comments from you earlier in the thread about keeping things relative. So I had also considered using GLD and USO instead of GC and CL. I did take a quick look at GC vs GLD and they seem pretty equal. Some days one will give a 2 and the other a 0, but once you broaden out the data set those differences seem to even out. Which jives with what you said earlier in the thread, that they are pretty equal with rare instances where you get a big divergence between the two. Will give this some more thought... Do you use a similar approach to currencies in terms of keeping things relative?
I started with Fisher's domestic OR, switched to London when you wrote about it and that has worked very well. What I'm still working on is the nature of currencies. I'm the 'let your winners run' type and I have the patience to let them run forever and a day. Doesn't work with currencies because absent a catalyst that moves them to a new range, they seem to stay in a fairly narrow range, compared to what I'm used to at least.
Great call on that USDJPY. I bought puts on 6J which is JPY.USD at 0.009880 will sell them if we hit 0.009797 area
Forgot to mention Cocoa looks good to me. Bounced off the monthly A down last week. Gold *was* looking good until today! Is a big move like this something to get freaked out over, or is it still good as long as the monthly A holds?
That is the exact big problem technical analysis has. This happend this morning : "Gold Slumps Most In 2014 As "Someone" Dumps $1.37 Billion In Futures At US Open" You can't do numbers lines, support levels, technical levels, analyses, voodoo, whatever when somebody on a Monday just dumps $1.37 bil in futures at US open.
Gold had a perfect failed monthly A up (chart attached). Never try to buy something that wicks like that at the failed monthly A up. Of course now it could be a nice buy at the failed monthly A down.
Coco has been the only shining star in commodity land this year. Although Coffee is looking interesting here at the monthly and QTR A down if it holds.