The past few weeks have seen a marked increase in interest in the Greedometers. Website traffic is up, and subscribers are way up. As is usually the case, there are many in disbelief, and some that are clearly drooling at the chance to find this approach wrong / a sham. I get it. Few are more skeptical and cynical than I am –especially about things related to the investment industry. To this end, I am annoyed by talking heads that forecast doom (stock market crash) but fail to provide important details, like:
- Why will this happen?
- When will it start (within a month)?
- When will it end / conditions for it to end?
- Can it be stopped? If so, what?
- How will it resemble previous crashes?
- Is it related to previous crashes?
This is why I take (what I consider to be) great lengths to share as much for free as practicable. Therein is the challenge: provide sufficient evidence of the work without compromising the newsletter business.
I’m not concerned with the Greedometers not working / with the forecast being wrong. This is not because I have faith (I’m not a man of faith), and it is not because I am dogmatic – I look forward to finding new data that allow further refinements and building a more robust toolset. Having devoted thousands of hours researching and refining these two algorithms over the past eight years, and applying them over the past 10-15 years of data (Greedometer -15, mini Greedometer – 10), it seems extraordinarily improbable that they would suddenly stop working now. This is particularly so since the sequences have been recurring with increasing regularity (not random!). Moreover, these risk gauges have been back-tested and applied through some of the most violent economic situations seen in the past century. I doubt recent Russian tensions /events are of the same order of magnitude compared to seeing the largest banks and insurance companies drop like flies — in terms of the impact to global stock, bond, currency, commodity markets.
Let’s not confuse precision with accuracy. These two gauges are based on weekly data, and are linked to weekly closing prices on the S&P500 index. (Actually there is daily, weekly, monthly, and quarterly data. But it is assembled into weekly units.) The Greedometer readings are posted 2 days after the fact (in the newsletter, Tuesdays around 5pm east coast US time). Furthermore, the application of these risk gauges to the S&P500 is -despite my efforts- an imprecise endeavor based on empirical evidence. Previous sequences saw a relationship exhibited in the S&P500 wherein the final peak was within 1% of the peak before it (true in 2000, 2007, 2011). Thus the forecast has been for the SPX to reach 1830-1870 — because the Dec 31st peak was 1848. The forecast was not 1848 +/- 1.00% (1829.52 -> 1866.48).
As I write this post, the futures are pointing to an open around 1876. Perhaps the SPX manages to reach 1878 or 1880 today. Many are suggesting the forecast is therefor wrong. 2 things.
- Thing 1: The week ain’t over yet. The gauges use weekly closing prices.
- Thing 2: If the week closes above last week’s close (1859), the forecast would not yet be wrong. Since mid/late January, the forecast has been for this week or the previous week to set a new secular peak in the SPX. That estimate was 1830-1870, and tightened to 1850-1870 more recently. Is the SPX setting a new all-time peak this week? Yes. Is it far from the forecast range? NO –> less than half of 1%. The missing ingredient is the SPX has yet to roll-over. It will.
Because of how previous mini Greedometer sequences have functioned, I estimate the odds are still 75% for the SPX to close under 1859 this week (the 1830s-1840s is my estimate). In the event the SPX closes over 1859 tomorrow (this week), a higher mini Greedometer reading will be generated for this week than last. That still aligns with the forecast made 6 weeks ago. With 6 previous mini Greedometer sequences as the reference library (and that’s not a lot), a higher reading this week would generate a larger mouth to the mini Greedometer funnel than would be anticipated. (The mini Greedometer funnel is the space between the topline and bottom line.) If this unfolds, so be it. The data is the data. My understanding and application of the gauge may have to be tweaked marginally. I’m very comfortable with this prospect. (don’t be dogmatic)
I would be deeply/wildly surprised if the mini Greedometer readings for this week AND next week were to climb higher. That would generate a funnel mouth as wide as the widest sequence yet: 2007-09. Given the current baseline is roughly twice the slope of the 2007-09 sequence, the mouth should be considerably less narrow. That’s a general theme of the 6 examples seen to date.
Keep this in mind: it takes months for the Greedometer and mini Greedometer sequence to build and begin decaying. All of this must occur (and does occur) PRIOR to the S&P500 peaking and initiating a collapse. Until there are 2 baseline points and 2 topline points set on the mini Greedometer, we remain in a mode where the stage has yet to be completely set. To date, the baseline has been set, but we’re still searching for the first topline point (much less the second). We will not know until next Tuesday if this week or the previous week are the initial points that set the topline. I have extremely high confidence (90%+) one of these two weeks will end of being the point that initiates the topline.
Take a deep breath. Here’s what’s going to happen:
- The SPX is going to roll over and drop 6-9% over the next 2-3-4 weeks.
- It is then going to bounce off the mini Greedometer baseline -again, further reinforcing the confidence in it.
- Then climb back to similar peak SPX levels. 1830-1860 is a reasonable range for now (you’re getting this for free, so that’s a decent forecast). That forecast be get tightened in the newsletter over the coming weeks.
- When the SPX rolls-over again in April, we’ll have the second topline point. THEN we’ll have the mini Greedometer funnel defined. Right in time for the next dip. A much larger dip — roughly 15-20%. Wash, rinse, repeat.
Note: If the Fed or ECB chime-in in the coming weeks with something new and exciting, the mini Greedometer baseline and topline will pick-up this change and reflect it in the readings. If the news is somewhat big, the baseline and topline may be slightly warped –as they were last year when Ben was threatening to taper QE, then backed away. This forecast is not set in stone. It is subject to the whims of central bankers. Let me be clear: the ONLY force that has warped or truncated the sequences has been large central banker intrusions. Not geopolitical events. In the event the ECB announces a QE program (and I expect it will in June-July) and if the printing press is large enough, the mini Greedometer sequence will get truncated. Only to grow -immediately- a new sequence with an even shorter build-up. If they don’t find their “inner Krugman”, things will get very ugly very quickly this fall.
If you are not an advanced access subscriber, you’re not going to see the play-by-play. I don’t intend to keep sharing this level of detail for free. (sorry that reads like a sales pitch).
- I am short the S&P500 with a 10% position in SDS (effectively acting like a net 20% short position). I’ve done all I can to provide advanced warning of this pending collapse — and put my money where my mouth is.