Simple 'Minor Bottom' again

[elann]

Does the bottom detector look at intra-day lows? I’m wondering because today was an unusual day in the market. The daily range between low and high was almost 200 points on the S&P. The day began with what looked like a bloodbath and reversed into a powerful rally, closing up 2.6% for the day.

Makes me wonder how many traders are watching for bottom signals. Is it possible that a lot of people saw a bottom signal early in the day, which triggered the huge market reversal and the strong close? How many people out there are front running us? If that’s the case, will the signal stop working?

Elan

[said2]

This is what I just heard from Monaco (Jim):

Despite today’s remarkable drop at the open, neither my “short term bottom detector” nor “major bottom detector” models has give even a weak version of a bottom signal since close Sept 29.
(the usual rule being to wait for a “no signal day” before investing, meaning the buy signal was end of day Sept 30)
That doesn’t mean they are right, of course. But they are silent for now.

[38Packard]

Here’s what seems to be the main reason the Market was so strong yesterday. Let’s see what today brings.

Stocks ROAR Back!

Nailed it!! Here’s a link that confirms your explanation. Thanks!! It was a massive rally, and one that came out of nowhere. And it’s left market observers like yours truly wondering what the heck just happened. There wasn’t any new data, no headline-making speeches, no event that occurred just after the open to spur such a move. It literally came out of nowhere—and left us grasping for possible reasons. “Today’s market reversal was a head-scratcher,” writes Oanda’s Edward Moya. And he’s n…

'38Packard

[aussi]

Elann

The platykurtic bottom detector uses new highs, new lows intra-day data. For Oct 13, from WSJ, NYSE, there were 934 new lows and 29 new highs. Total traded was 3402. I think platykurtic’s approximately one third new lows for major bottom uses total listed, not total traded.

Markets Diary (wsj.com)

[Rayvt]

@platykurtic

Here’s 2020 - there are no column labels but it should be self evident what the data represents (for some reason it interprets the label line from the s/sheet different to the rest of the table):

Expand Table

Humph. Here’s a great example of how this new format sucks. On the old boards, when I decided that I didn’t want/need to read the entirety of a long post all I had to do was click “next” to skip over the rest of the post.

Now, I can’t. Have to scroooooooollllll down and down and down to get to the next one.

[lohill]

You could also hit the browser’s back button or the n unread again.

[platykurtic]

@Rayvt

Humph. Here’s a great example of how this new format sucks. On the old boards, when I decided that I didn’t want/need to read the entirety of a long post all I had to do was click “next” to skip over the rest of the post.

So it turns out there’s a ‘Details’ tag underneath the ‘gear’ icon in the edit box. It allows ‘spoiler’-like collapse-able sections. And so I edited the long table based post to include these tags (see above).

I propose that we attempt to use that details tag moving forward for large data drops. I suspect there’s a few of these types of features that we’re going to need to learn to use the new system at our previous level.

[platykurtic]

@aussi

Total traded was 3402. I think platykurtic’s approximately one third new lows for major bottom uses total listed, not total traded.

Just a minor correction - in the simple bottom detector calculations ‘total traded’ or ‘issues traded’ is used.

This is for two reasons…

1. The data is readily available on the WSJ Markets page itself & I’m quite lazy (the main reason).
2. Whether a stock trades or not offers up information on that stock’s status. If in a heavy down move an issue doesn’t trade it’s not contributing to the fall, isn’t likely invest-able & isn’t likely that important.