@Anony So for instance if I'm using RSI ...would it make sense to use another OBOS indicator along with it...to "verify/confirm" the security is "actually" overbought vs oversold? And to "confirm" the RSI signal?

Very good way of laying it out Anthony! This is similar to looking at a valuation metric of a single stock. This data point is almost meaningless until it is compared to something (competitor, overall market, industry, etc..)

So would you suggest to use RSI as compared to a competitor or overall market VS comparing recent gains to recent losses to itself (basic calculation of RSI)?

@Anony I appreciate your input on this. I'm hesitant to exclusively use a simple RSI calculation to determine entry points. This, and this alone seems to be missing the component of being compared to other competitors/market/etc. I like the idea to doing some kind of blend, based on perhaps my basket of stocks vs each other or vs the index etf.

Going back to the notion of entries, would you say Volume should also be considered for entry points? Such as an "X" day moving avg above some trigger. I've heard args for and against.

Hello Anony,

Interesting comment, "This site makes it fairly easy to play at trading; to dabble and fiddle and tweak and learn. But it's not conducive to the actual practice. It's great for exploration and sharing and that's cool, to a point." When you get the chance, I'd be interested in hearing what sort of setup you'd recommend for more serious work.

Grant

Thanks Anony,

I too wish the Quantopian team well...I've definitely enjoyed participating in their start-up effort.

Fundamentally it doesn't make sense that annualized commissions for active trading should be all that much higher than those of passive investing. By "automated" I don't mean that traders aren't needing to keep an eye on activity (playing "man-in-the-loop") but rather that Quantopian/IB are providing what should be a kind of commodity service, since they just have code running on servers and electronic record keeping--nothing particularly expensive in today's world. The idea of paying $1 to electronically buy 100 shares of stock is sort of ridiculous in my mind, given the computing and communications infrastructure. The mechanics of trading should be asymptotically approaching zero cost as time goes on (unless perhaps there is a bunch of regulatory overhead that keeps the price artificially high).

By the way, "sacrificial cash" in the range of $25k to $250k is largely unrealistic for prudent, working middle class folks with expenses (e.g. kids/home/cars/camp/daycare/college/retirement/insurance/taxes...the list goes on). Of course, if one knows how to make a low-risk profit rather than sacrifice the cash, it is a different story. Any tips?

Grant

@anony for the RSI_14 which i'm using in my algo...what would you suggest would be a good "trigger" for determining ob/os? I was using 30 and 70 for my trigger levels...however should i using something closer to the theoretical 0 and 100? Basically i'm calculating (min) absolute distance for my basket of stocks to determine which has highest priority to long/short...