New Members: Be sure to confirm your email address by clicking on the link that was sent to your email inbox. You will not be able to post messages until you click that link.

Writing scan for a 20 day test

Tom Bowley and John Hopkins often mention scan for stocks testing the 20 day with RSI between 40 and 50. The RSI part is easy to figure out, but I'm struggling with how to define a 20 day test. Is it when the close is within a certain percent of the EMA for example?

Any suggestions would be appreciated.

Comments

  • markdmarkd mod
    edited February 2018
    I don't watch them (not a judgement) so I'm not familiar.

    "Test" is kind of a loose term. If you want it to mean "getting close to", you could choose an arbitrary per cent, or maybe an ATR of some length (e.g. the current bar low or close is less than the ema plus some per cent, or plus 1 ATR, whichever you prefer).

    A stricter interpretation is that at least the low of a bar has to cross below the ema ( e.g. ema(20, close) x low). If the close of that same bar is above the ema, and stays above it, that would be a successful (and strong) test. Some people would also allow one, or two or three closes below the ema and still call it a test, provided the close stays pretty close to the ema (say, less than 1 ATR under).

    Whether the test becomes a break depends on subsequent bars.

    In any case, a down leg in a strong up trend usually reverses in the area of a strong up bar (heavy volume and a long body) in the preceding up leg, or a congestion area where several bars overlap so that there was a lot of volume in that area. If the ema 20 is hit while it is in the range of that prior volume, it appears to be the ema that is offering support when really it's that prior volume. Nevertheless, it seems to be a good thing when a test of volume occurs together with a time-based event, like price intersecting a 20 period ema, or maybe a 20 period price channel or stochastic. Gann made a big deal of price and time working together and there is some truth to it.
  • Thanks for that mark. I'll do some more reading in the direction of Gann, that sounds like it would be worth it from an education standpoint. In the meantime I will watch for the dynamics you mention in my chartlist and see if I can identify some ways of developing some scan code for it.
Sign In or Register to comment.