300 Point Rally follow up

Barry…several days ago I posted the quote below from Crooks & Liars and your system kicked it out. But I’m posting it again because it’s important thant all these analysts start being held accountable for all their prognostications. These people just babble on and on without anyone ever calling them out, in the open media I mean. So here goes and hopefully this time the system does not kick it out.

this post with Michael Hanlon in mind: C&L’s Accountability for the Punditocracy Proposal

Here’s a few things the networks can do to clean up their act.

1) Set up an Ombudsman with a staff for each network that isn’t an employee of their corporation and have a weekly segment devoted to policing the media. They will also be available to take complaints reported by individual citizens and investigate them thoroughly.

2) Replay clips of each pundit when they’ve been proven wrong and let them explain their positions and why they thought they were right and ask them how they will correct their mistakes in the future.

3) Keep track of their infractions and set up a benchmark, like a 3 strikes your out rule for pundits. When they hit the benchmark, suspend them for a period of time so they can reflect on their mistakes.

4) When they return to work, ask them why they should be believed in the future.

5) It would be nice if they stopped using pundits that we know have been wrong over and over again.

The american consumer is broke. Talking about statistical noise if fun. But so are fundamentals.

There is one way this story ends. If you are still guessing, you haven’t been paying attention.

Justin….yeah agree. I’m just tired of all the bullshit that passes for analysis. I worked on the street for 30 years and was held accountable by my clients/bosses for things I said & did. Oh well, that’s my rant for the day…..going on to enjoy my retirement. Peace.

That’s an illuminating chart. What is it really telling us?

In the most general terms, it says that historical volatility tends to rise during bear markets, and to quickly taper off during bull markets. Thus, the remarkable run during 2003-2007, when 1% and 2% down days almost disappeared.

BUT, the launch of a new bull market often includes a 3%-plus upmove during the early days. So selling every sharp rally works until the final time. Then the runaway freight station leaving the station crushes you to dust on the tracks.

To research this in the most general framework, I would look for patterns in the rise and fall of historical volatility, which generally correspond with falling and rising indexes respectively. Point changes and even percentage changes will show secular drift, but changes in volatility will still move in the expected correlation with indexes.

Of course, there remains the big asterisk of the 1930s, when huge daily moves occurred both during the 1930-32 collapse, as well as in the monstrously volatile rally of 1932-33. Selling the giant pop the day after Frank R’s “bank holiday” finished would have been a career-ending move.

Barry, this is very interesting in light of how many financial advisors dangle the argument that you have to stay in the market no matter what, since X% of the market gain is made in Y number of days, ie the big 300 point gains. This logic is false when you realize that even being in the market on those big days means you can be way down in the hole at the end of the slide.

So funny that you posted this. Just yesterday I did an analysis of percentage move days going back through 2003. I’d go back further, but the results follow the Vix close enough. I measure the absolute value of the move. (Whereas the Vix pops on big down days, in my opinion a 3% up day is just as significant given that they’re so infrequent in bull markets.) Below are the results.

1st column = year
2nd column = 2% move days (this includes 2% down and 2% up)
3rd column = 3% move days (this includes 3% down and 3% up)

2003 15 3
2004 0 0
2005 0 0
2006 2 0
2007 17 1
2008 21 4

The volatility is clear to see. We’ve already had more 2% and 3% moves than last year, and we’re only seven months in. 2003 was obviously a big up year coming out of the bear market, and volatility was still high, so this is by no means a perfect indicator. Overall, though, it does say to beware of clumps of big moves, either up or down. It screams of investor/trader anxiety.

This is not a “typical” bear market. Anyone that states this will say anything to be proven right (Mr Kudlow). This may very well be the turning point for America. There are so many factors contributing to this downturn it is almost unbelievable. Would it be big surprise if the market moved down 50% next week. No, it’s not likely but afterward everyone will say “yes, we saw it coming”. Where are our strong leaders? Aha…we don’t have any.

the average move from the close of the 300 point move to the lowest low over those subsequent sixty days is -7.73%

not bad. curious what the subsequent highs looked like over that same time range.

Good work, Mista B. Volatility is the key, and as you point out, it is bidirectional.

I happen to agree with Barry that March 2003 was the real bull market blastoff, although the price low was set in Oct. 2002. On the second day of that rally, 13 Mar 2003, the Dow rose +3.57%. But since it was starting with a 7-handle, in Dow points it was only +269.68, not enough to qualify for the “300-point” study.

Yet at the 14,000 level which the Dow hit last year, a 300-point move was only +2.14%, far less than the Dow’s +3.57% launch in March 2003.

Framing the question in terms of “300 points” is just wrong, when the Dow index nearly doubled during the bull market. (Back in the Sixties and Seventies, it would have been “30 points.”)

And as mentioned before, bear market volatility tends to linger briefly into the blast-off phase of the next bull move. So the “sell the 300-point rallies” rule works just dandy, until it blows up in your face at the V-bottom secular trend change. Find the missing variable to tease out this vital distinction, and you can be the next Soros!

Jim,

I fully agree that the percentage move is far more relevant than the point move. It’s arguable that since 1998 or so it would be okay to look at just the point move if one were only looking at the Dow. I say this because for the most part ever since the Dow initially broke through 10,000 it has stayed there. When it dropped below, it bounced back fairly quickly. Thus a 300-point move has occurred pretty rarely. That’s really what matters, imo.

Based on the numbers I showed, I think the number of 2% move days actually says more than the 3% move days. The 3% days are pretty rare, even in a nasty bear. The 2% days, however, rarely occur in bull markets, but they sure show up in bear markets.

Whether one could develop a timing strategy based solely on this data is certainly up for debate. (I think you’d be likely to miss the turning points, perhaps by a wide margin.) Personally, it’s helpful to me in that it confirms my belief we’re in a bear. I prefer looking at many data sets–demographic, economic, fundamental, technical, etc.–even for my shorter-term trading strategies. Regardless, to put it mildly, the percentage-move data does say one thing: it sure is yucky out there!

“Heading close to 300 point day down on the Dow. Is
that bullish as well?”

No, that’s “profit-taking” !