Tooting my own horn ...

[mmerlinn]

...also consider that Schiff might be taking into account, that each of his interviews might be someone's FIRST time he/she came across him, thus he might want to break through the initial "who the hell this joker thinks he is" attitude of his first-time viewers, by quoting his fulfilled predictions.

The above quote found in this thread has prompted me to toot my own horn.

Based on that quote, maybe I should be more aggressive in pointing out my fulfilled predictions.

I believe history repeats itself over and over and over again in a predictable manner. Being a numbers man, I am always looking for events that line up by the numbers in a predictable way. And I am often spouting ideas based on those assumptions, many of which are often wrong.

However, twice in a row I have accurately predicted both the timing, to the exact day, and the severity of the downdraft of the stock market, to almost the exact percentage, YEARS in advance. I predicted both the 1997 and the 2008 market crashes almost exactly, both to the day and to the depth.

In 1997 I made my first predictions about 8 months before the fact and on the predicted day I was at a brokerage at 5 in the morning (Monday, 27 October 1997) to watch the market reach the low almost exactly at my numbers, some 40% below the high 56 days earlier. At the end of the day, I took a screenshot which I printed out. I still have that hard proof.

In April 2005, I predicted a low on Friday, the 24th of October (in 2008, not in 2005), that would be 40% below the high 56 days earlier. In July 2008, fully 3 months before the markets crashed, I posted those predictions on the internet on a site that no longer exists. However, on the 16th of October, someone on another site found those predictions and presented them to the readers on that site. Anyone that is interested can view that very popular (13,090 views) thread here. The interesting part is that I made sure that I was in Austin, Texas on the 24th of October, 2008, the exact city I was in when the market crashed on Monday, October 19, 1987, 21 years earlier.

Note that much of what I predicated was incorrect, but that the underlying premise, that the market would bottom out on the 24th of October at about 900 basis the S&P was spot on. Could I do it again? Absolutely if and only if things ever line up the same way again. As of right now, I see nothing that prompts me to believe that the markets will crash on any specific day in the future, so, for now, if I make any predictions, they are generalized and not pinpointed predictions.

I full well believe that there will be a market crash next year for multitudinous reasons, some of which I have spelled out in other threads here. But I have done NO research that would indicate any specific date or dates. Therefore I won't make specific predictions, just generalized ones, which, of course, could mean more mud on my face.

When do I think we MIGHT have another crash like 1929 or 2008? Dates that come to mind are Friday, 28 October 2016, and Friday, 26 October 2018. I strongly suspect that one of those two dates will see a market low 40% under the high around the previous Labor Day. However, I do not have enough information to stick my neck out and say it will happen on either of those days, although I tend to lean towards 2018.

 

[pmbug]

So, late October is the witching week?

 

[Penn]

A crash for who? Most people are already wiped out.

 

[Mark]

So I guess that means that I go from 0 into the negative numbers.

 

[rblong2us]

Hey mmerlinn

brave statements as anyone prepared to call the future for big / distant events is setting themselves up for a fall. The odds of being right are so small ............

Yet youve done it for two market crashes.
Number patterns bother me because they seem to be anywhere you care to look ( at least to me ) and trying to put meaning on them bothers me even more.

How did you predict the crashes so asccurately, years in advance ?

 

[rblong2us]

... correlation with US debt is the best one can ever get on PM (gold) prices, and that is quite natural, as it is solvency of the USA and USD based system , not actual inflation as such (but expected inflation at the collapse/reset) that is driving the PM prices up.


In 2011- beginning of 2012 US debt growth SPEED ( increase/year or even better, % increase/year) has come markedly down from almost 2 trillion/year increase in 2009 and 2010. In fact, in 2011 it has been 1/2 of 2010 increase- about 1 trillion , so far the same low speed is maintained in 2012. Hence the drop in PM prices in 2011 - early 2012 as expectations of continuous fast debt growth which drove the overbuying in 2011 (but was quite logical on the back of 2009-2010 debt growth speed) of the USA debt did not materialize. However, this is just a temporary slowdown (it also happened after tech bubble before Iraq war) for many reasons, recession/USD strength being the most obvious one. War- if that happens- absolutely. These things all may happen before November, as well as QE as a response to them. But it will happen, I mean acceleration in the USA debt growth.



One more thing- for systems moving towards crash ( i.e. partial default on US debt) , it is quite characteristic to show log-periodic oscillations before the final spike and crash. Log-periodic oscillations in such case INCREASE in frequency (decrease in wave length) nearing the crash. These oscillations sit on top of exponential growth curve, so are more clearly visible in logarithmic chart. Close to crash, growth curve turns superexponential ( on log chart that would mean upwards deviation from straight line) and oscillations disappear.

Looking at US debt chart above, there are easily discernable two waves on top of exponential growth, with a natural beginning in 1971 when USD was moved of gold and debt accumulation became inevitable way of US operating mode. The first wave is 30 years long, ending in 2001 where the next wave starts. The next wave seems to end in 2011- 2012, making its length 10-11 years. I think this is the last wave before the crash, but of course there could be one more 3-4 years long delaying the default to 2016-2017-even 2018. ( I think it will be end of 2015).

The ratio between the periods, 30/10=3 or 30/11 = 2,72 which is quite close to basic constant in time related processes in nature, e = 2,718... Such ratios between log-periodic oscillation period lengths occur quite often ( close to 3) and , while there are no proofs, can be considered quite Natural in time related processes in the same sense as Fibonacci ratios are natural and often met in Nature in size/number related values.

So either

1) this was the last oscillation and next is the superexponential spike into default ( in which case default comes 2014-2015 end) ,

2)or there will be still one more oscillation of about 3-3,5 year length , leading to INCREASED rate of debt growth in 2012-2015 and crash (partial default) in 2016-7-8.

In both cases, the PM prices shall move up , though in first case it will happen much faster and the end of normalcy will come much faster as well.

Choose between these two , my modeling here http://www.tfmetalsreport.com/comment/83050#comment-83050 pointed more to the case 1), but that modeling is not definitive as fit is not perfect. Lack the knowledge/matehmatical resources. I wonder why Sornette has not done this projection in his Financial Crisis Observatory site-looks so obvious and very interesting target to me to predict:

http://www.er.ethz.ch/fco/index

interesting comment from Ivars ( eastern european mathematician who has been pretty good with predictions ) on turds site tonight, regarding the timing of the next crash, and its similar to mmerlinns

see comments - http://www.tfmetalsreport.com/blog/4026/metals-grains-crude-oh-my

 

[mmerlinn]

So, late October is the witching week?

Only in certain years averaging roughly 10 years apart. The rest of the time, as best I can tell, there is no possibility of a crash in late October. That does not rule out crashes at some other time of the year.

For example:

1987 + 10 = 1997
1997 + 11 = 2008

Mathematically crashes were not possible in late October of 2006, 2007, & 2009, while 2005 was too early, though possible.

2008 + 10 = 2018

A crash is not mathematically possible in late October 2017, 2019, & 2020, but I think 2016, though possible, is too early. That leaves only 2018 as a viable date. A lot of things can happen in the next few years. Come 2015, I will probably be ready to stick my head in the guillotine and predict the date of the next crash.

 

[ancona]

IMHOI, crahses are emotional and not subject to numerology, fibonaci or otherwise.

 

[Jetstream]

IMHOI, crahses are emotional and not subject to numerology, fibonaci or otherwise.

Particularly if those crashes involve airplanes.

 

[bushi]

IMHOI, crahses are emotional and not subject to numerology, fibonaci or otherwise.

true, they are emotional, but you imply here, that our emotions cannot be numbers-related - while they POSSIBLY could be related/described in numbers, at least some of their characteristics... Like for example some natural/statistically prevailing sense of timing. Example - ever tried to talk on a walkie-talkie without employing an official protocol (or over a connection with a significant time lag)? People tend to wait for a response from the other side of the line for very similar period of time. Both of them. Then BOTH of them start talking, simultaneously. Then they stop, when they realize that the other person also just started. And again, they instinctively tend to wait for a very similar period of time, before starting talking together again

So while emotions/impulses themselves are non mathematical constructs, but their timings, for example, or maybe even their strength, could possibly be correlated to some functions. I don't know, but why not. This all might be even linked to physiology, like average time required to comprehend the state of communication vs. protocol lags, or maybe even to things like nerves length/speed of neural impulse travelling through our bio-computers. Statistically, they might be correlated to some numbers/functions, with more or less accuracy.

That would be a hell of a job to find these correlations by analyzing every possible combinations and thesis, but maybe it is possible to focus on MANIFESTATIONS, through collective ACTIONS/decisions (market moves), as mmerlin seems to be doing (?).

Still, you would need some important inputs to your functions, I believe (like news that have potential to move markets - and how to determine, which news can move markets and how?), I don't believe that numbers alone would be enough to predict the market moves. I simply don't.

 

[mmerlinn]

I went back and checked my figures more closely. Upon doing so I found that I made a mistake in two of my posts here.

When do I think we MIGHT have another crash like 1929 or 2008? Dates that come to mind are Friday, 28 October 2016, and Friday, 26 October 2018. I strongly suspect that one of those two dates will see a market low 40% under the high around the previous Labor Day. However, I do not have enough information to stick my neck out and say it will happen on either of those days, although I tend to lean towards 2018.

The above should have been:

When do I think we MIGHT have another crash like 1929 or 2008? Dates that come to mind are Friday, 28 October 2016, and Friday, 25 October 2019. I strongly suspect that one of those two dates will see a market low 40% under the high around the previous Labor Day. However, I do not have enough information to stick my neck out and say it will happen on either of those days, although I tend to lean towards 2019.

Needless to say, the following quote was dependent upon the above quote, so it, too, was wrong.

2008 + 10 = 2018

A crash is not mathematically possible in late October 2017, 2019, & 2020, but I think 2016, though possible, is too early. That leaves only 2018 as a viable date. A lot of things can happen in the next few years. Come 2015, I will probably be ready to stick my head in the guillotine and predict the date of the next crash.

The above should have been:

2008 + 11 = 2019

A crash is not mathematically possible in late October 2017, 2018, & 2020, but I think 2016, though possible, is too early. That leaves only 2019 as a viable date. A lot of things can happen in the next few years. Come 2015, I will probably be ready to stick my head in the guillotine and predict the date of the next crash.

The only reason I caught this is that I was doing the math for another site and the numbers did not match.