Wednesday, June 3, 2020

What to Do While Not Counting Waves?

Most Elliott wave counts of importance take place on the Weekly, Daily and 4-hr time frames. Shorter waves may be countable or provide information on overlaps, but for some day-traders that can be a lot of time to wait trying to see if a wave count completes, and a reversal begins. What to do with the intervening time?

A while ago I published a Paraphrase of Ira Epstein's Guidelines for Trading at this LINK. One of the key guidelines is that "The 18-day SMA is the Line in the Sand". No short trades are taken above it and no long trades are taken below it. Now, have a look at today's intraday chart (ES 30-minutes) which I was following for my own account.

ES Futures - 30 Minutes - IntraDay

What do you notice? It looks remarkably similar to one of the same daily charts. It should, it has an 18-period SMA, and Bollinger Bands (18,2), and the Slow Stochastic Indicator on the chart.

First, what if a person were to screen the entire upcoming day by first realizing whether prices were over or under the daily 18-day SMA? In other words, if daily prices are above the 18-day SMA, then only long trades could be considered intraday. If prices are below the 18-day SMA, then only short trades could be considered.

That being the case, starting at the lower left look for the first two-bar fractal that breaks upward. That 'could be' an entry.. ummm.. IF you are awake, If you decide, etc. (Remember, nothing here is trading or investment advice. You are responsible for your own trading decisions).

Now, notice how the slow stochastic goes "embedded" while price is "riding" the upper Bollinger Band. This is the strongest technical signal. That being the case, this methodology would suggest at no point would a person consider being short - both for the daily chart, and the intraday chart.

But, then, as Ira suggests, notice how there are four consecutive closes above the upper daily Bollinger Band. The chances of being above the band at random at any time are only 5%. And, you subtract 1% for each further consecutive close. So, what does that get us? 1) 5%, 2) 4%, 3) 3%, and 4) 2%. Yes, after that fourth candle, there is only a 2% chance the next bar will close outside the band.  And, it doesn't.  In fact, there are a series of candles lower down to the lower band.

The guidelines indicate one should not "buy new" over the upper band, but one can "let profits run" - until the slow stochastic again closes under the 18-period SMA.

If one exits sometime after 07:00, then one still does not go short because the price is still above the 18-day on the daily chart. But one would look for fractals to break upwards after price recycled down to the lower Bollinger Band which is typically at the 50 to 62% retracement zone. Again price gets above the 18-period on the intraday chart, the slow stochastic goes embedded and price never gets to the lower band again - a sign of strength.

Then, before noon, there are fully six closes over the upper intraday Bollinger Band. These occur while the slow stochastic is embedded, and price is "riding the band". But, the six consecutive closes means that  there is a near 0% probability of the next bar being outside of the band, and again, it isn't. Prices sneak inside the bands for next seven candles, until the last candle of the day which again tags the upper band. The slow stochastic then loses the 80% level as prices fall off, with the expectation that price returns to the 18-period SMA on the intraday chart.

Does this sound familiar? It should. It's why I keep trying to separate "trading" from wave counting. But it is based on some simple ideas. First, why did I chose a 30-minute time period? Everyone will say, because you can divide the 24 hours of the day by 30 evenly, and you can evenly divide the cash session into 30 minute periods too. Yes, but that is not the complete answer.

The complete answer is that Bollinger Bands are based on "two standard deviations from the mean". And I once learned in statistics class that it is only accurate to calculate a standard deviation once you have 25 observations or more (like to calculate the standard deviation of the heights of dogs, you need at least 25 height measurements).

Well, 24 / 2 = 48, meaning that 48 half-hourly bars should allow for a very, very accurate calculation of standard deviation for a day's trading. The other realization is that if the Elliott Wave is truly fractal in nature, then if the daily Bollinger Bands are good guides on the daily chart, then some subset of them should also be good guides on the intraday chart.

But, next, how effective would it be if one only looked for new long trades when one is over the 18-day SMA, and not over the daily Bollinger Band? And how effective would it be, if one only looked for new short trades when one is under that 18-day "line in the sand"? Are you this disciplined? Or, are you looking to "top-tick" the wave count? From that paraphrase of the trading guidelines, at least one broker (Ira) suggests it might be better to see if the wind is at your back first.

And, how effective might it be if one only took trades as above, in combination with the same direction as shown in the 30-minute chart?

I don't know. I don't know what you will decide to do with this information and this example. I only offer it for your consideration. It is not to be considered as trading or investment advice. I don't know your skill level. I don't know your financial situation, I don't know how familiar you are with fractals. I don't know know how much of an appreciation for standard deviation you have. So, your decisions have to be your own.

To be more complete, I add the daily (classic calculation) floor trader's pivot points to see where there may be targets, but I don't know if you have access to them, or an appreciation of them. Many do. Some don't.

So, that's what I do when not counting waves. How about you?

P.S. This is the second post today. The shorter term wave count of minute ((b)) is on the prior post.

Have a great rest of the evening.
TraderJoe


23 comments:

  1. Thanks Joe. IMO, this is one of the best posts you've written, and I've been following for a couple of years. As the parable says, "Give a man a fish, he eats once. Teach a man a fish and he eats for his whole life."

    Getting back to current events and based on my chart, prices on ES have fallen under the upper wedge boundary. It hit a low of 3111. At 18:05 pst it's back up to 3117.

    ReplyDelete
  2. Great info! Thanks!
    I wish I had as good an explanation behind the raison d'etre of my own non-wave counting time spent looking at round numbers which has worked well for me. I am working on it. I cannot fully explain why, but I was very confident after the decisive "conquest" of 3K that 3100.00 would be in play. Don't fully know why the pattern obtains, just that it does!
    Again, not trading advice and for conversational purposes only.

    ReplyDelete
  3. What is the significance of 18-day SMA over the conventional 20-day SMA?

    ReplyDelete
    Replies
    1. There isnt any.

      Delete
    2. Cycle analysis does indeed show that there are a LOT of 20 - 21 day cycles in the stock market. However, it also shows there are plenty of 16 - 19 day cycles, as well. So, the 18-day is a blend of the two.

      Delete
    3. ira epsteins answer for why the 18 day sma is ....

      over 40 years watching this ..
      . traders watch this average
      so i watch it
      because it becomes a self fulfilling prophesy
      if everyone is using it

      Delete
  4. "The guidelines indicate one should not "buy new" over the upper band, but one can "let profits run" - until the slow stochastic again closes under the 18-period SMA."

    Shouldn't that read - until both of the slow stochastic moving averages again close under 80?

    ReplyDelete
  5. Oh, yes. Although "Mom & Pop" via the AAII consensus are not bullish, the professionals are, and I think largely they have been forced into it by the FED.

    ReplyDelete
  6. So, I showed very clearly in the chart in this post, the alternate red Minor B very early on. That remains a possibility.

    http://studyofcycles.blogspot.com/2020/05/first-vaccine-day.html

    But currently minute ((b)) is not longer in price and/or time than Minor A. Therefore, by degree labeling, it may still fit as a sub-wave. If it somehow becomes longer in price or time, then it will be required to adopt the Minor B count as shown. The same is still true on the djta. But, there, a clearer reasonable count is diagonal A, and then a non-overlapping C after it. But even there, a minute ((b)) can not be ruled out.

    That's why flexibility, patience and calm are still needed. There are simply times when Elliott Wave presents us with two equally good alternates.

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete
  8. As I pointed out a few days before that post, the minor B could have been done at the 2,760 low. Since then that has been my primary count, and I have conviction that we're currently completing (or have completed) minor C. The overwhelming bullish sentiment supports that thesis - the 'pain trade' has now flipped to downwards rather than upwards in the 'unpopular rally'.

    ReplyDelete
  9. Excellent post. "The trend is your friend" - identify it and trade with it.

    ReplyDelete
  10. This comment has been removed by the author.

    ReplyDelete
  11. Chock full of usefull info, Thanks Joe!

    ReplyDelete
  12. Thanks TJ. What you have shared are very good tools for considerations and analysis. Have a good day and week! Stay safe.

    ReplyDelete
  13. Trader Joe,

    Excellent Post, I like using the Bollinger bands as a squeeze indicator also.

    ReplyDelete
  14. This is all I see so far.

    https://invst.ly/r0rpe

    TJ

    ReplyDelete
    Replies
    1. break of channel upwards, and upward overlap suggests only (A), (B), (C) down this point.

      https://invst.ly/r0ryz

      TJ

      Delete
    2. ES now lower low, could start a diagonal or wave three.

      Delete
    3. ES futures took out the overnight low.

      Delete
    4. ..keep your eye on measurements, IF below 3,087, then it's a third wave 'of some type'. Like iii OR c; if not then can be a sub-wave.

      Delete