An interesting, but subtle view of market behavior is an indicator developed by Pascal Willain and which is available at www.effectivevolume.com. The indicator is the 20d Money Flow, which is a rolling indicator that reviews the past 20 days of large money flow across different sectors (industry groups, actually), operating under the premise that large funds spread risk across multiple stocks and ETFs that are focused on the same sector.
Timeframe and context are important in any trading system, and the 20d MF indicator is no exception. This is a short-term indicator, e.g., it fires quickly at oversold and overbought points, and it can easily reverse in a short-term. It tracks the choppiness of the markets, so it stands to reason that it is faster than my GGT or other moving-average based systems.
Pascal has combined the signals of the 20d MF signal with an overbought/oversold indicator, and when these two fire in close proximity, I typically make an effort to review the status of these two on a daily basis.
These two systems have fired a long signal within the last week:
The OBOS indicator has been rocketing upward the past few days, and crossed over Pascal's empirical level of -70 from below with the close of markets on 11/19 or the open on 11/20. Had you checked the site last weekend you would have read that we had the model transition LONG with the close of markets on 11/16 -- last Friday -- because the OBOS indicator moved upward 3% in 1 trading day.
Either way you look at this, specifically whether the signal occurred last Friday 11/16, Monday 11/19, or the open 11/20, from a money flow perspective the markets were heavily oversold and money moved in enough to cause us to wake up. Here's the view looking back in my rear-view mirror over the last 5 trading days, from 9:54 a.m. on Friday. It is clear that we have quickly resolved the oversold conditions that existed just one week ago.
IBD -- that newsrag followed by thousands and also anticipated by many more, called Friday 11/23 a follow-through day which triggers a number actions for folks who follow Bill O'Neil. The market has been in a confirmed correction for several days, with the issue printed Wednesday night (Friday's dated edition since Thursday was a holiday) indicating that the markets were in correction but leaders were crafting bases. Here's the latest from the close of markets Friday:
I'll leave it as an exercise for the reader on why DVA, RGR, and TSM are poor candidates for entry at this time, but BAX and CRM probably require some additional review. Post a note below if you think you know why DVA, RGR and TSM are dangerous or why BAX and CRM could be attractive at this point in time.
So with EV showing that we're on a buy signal, and with IBD now showing that the market is in a confirmed up trend, what's an investor to do?
I look at my GGT dashboard, of course.
Again, right click on the image to open in a new tab or window.
I've drawn a line to highlight the GGT model response on the 16th onward. Prices have been up day over day since 11/16 without exception. Volume has dropped day-over-day since the 16th without exception, which was anticipated due to Thanksgiving but is hardly a resounding call to throw everything you own in the market.
If you look closely, GGT's Database Strength indicator bottomed on 11/15 and has not looked back since. History is repleat with whipsaws on this indicator, as it is simply a day-over-day view, so by itself it isn't a signal. Nevertheless, it's moved to middle territory now with a database sitting at mid-scale -- 0.518 -- and this means that there's fuel to go higher but also free space to drop. It's hard to state anything more here except that the market is in true equilibrium.
The price slopes of my model started turning positive on 11/16, and even with Friday's shortened action moved all positive. On the right side of the table you see a sea of green. This data represents the day-over-day acceleration of prices -- this section always leads the left side of the table (why?). We have a considerable amount of green on the right, and this portends well for the short-term trend. If we start seeing red and negative numbers creep back in on the right side we know that prices will be under pressure, and that this signal could be short-lived.
Based on price behavior alone I would say that a toe in the water is warranted, but not so fast Kimosabe...
I place a considerable amount of value on my Long-Cash Ratio Slope Model (LCRSM), which considers price, volume, and price rate of change of over 3000 stocks. Prices can be moved around with algos, but it is more difficult to do this in combination of price/volume/ROC.
Here's the slope model dashboard:
Again, right click on the image to open in a new tab or window.
By my rules the close of markets Wednesday night triggered a "buy" signal. A "Buy" is signaled any time the model has:
- the 2d and 3d slopes of the LCR are positive or
- the 2d and 3d slopes of the LCR are simultaneously moving positive with
- the 5d slope of the LCR moves positive.
- Despite the market PRICE moves of Friday, 11/23, the database as a whole did not participate in volume and/or price rate of change. Most likely volume due to the shortened day, but at this level of observation there is no real way to know.
- These are "huge" jumps in LCR values, all things considered. I would conclude that prices are being driven up faster than the market can bear, and I would expect a consolidation to occur in the following week so that the gains can be digested.
Last year's behavior, just as another datapoint, was far more optimistic than this year. Thanksgiving Friday was 11/25/11, and the markets fell (per GGT) -0.54% on volume that was -65% lower than the 50d MA of volume at that time. Leading up to this period was bearish, with the markets quite oversold. In fact, the GGT database strength indicator was 0.112, compared to the value of 0.518 that we have now, so we were at a very different set of conditions last year relative to this year.
I think it's important to review last year's behavior, although I freely acknowledge that last year has no real bearing on this year:
Thanksgiving Friday, 11/25/11, is the top line in the LCRSM table above. Monday, 11/28/11, is where the markets went PRICE bullish, with a GGT price gain of 3.56% (not shown) and a "ho-hum" LCR change of +36%. Note that the 2d LCR slope moved positive, but all the others were negative. Note more importantly that the right side acceleration values were already positive (green), which was good.
On Wednesday, 11/30/11, the markets jumped ANOTHER +4.61% (not shown), as measured by GGT, and the LCR registered the all-time high daily change of 202%. THIS was a significant event, and this was a buy signal if there ever was one. Note the sea of green on the right side of the table, but also note a bit of dark clouds on the far right.
On the far right of the LCR table is a summation sign and a change in summation sign. This is a daily indicator that tells me just how powerful the daily move is -- think of it as an average of an average. The time lengths vary of the averages, and consequently, when the results are summed, we get a value that can range between +14 -- very strong, bullish move -- to -14 -- a very strong bearish move.
Note that the summation values all stayed around 0 -- the markets were moving on incredibly strong prices, but in context of previous moves, the strength wasn't there.
Indeed, on 12/8 we had a shot across the bow. The markets fell -2.48% as measured by GGT (not shown), and we registered the 64th strongest down day since September 2008. Weakness in the LCR had been creeping in -- look at the red developing on the right side of the table as early as 12/1 as profits were taken off the 11/28 and 11/30 moves.
The buy signal was completely negated shortly after 12/8 -- the LCR model turned completely bearish with the close of markets on 12/13, and volatility was moving upward. Despite this, a general thawing started occuring on 12/15/11 and 12/16, and on 12/20 slopes started moving green again. On 12/21/11 we received a new buy signal, and as most recall, we were off to the races.
If you look closely at the far right of the 2011 LCR table you'll also see that the summation values were very strong starting on 12/20/11 -- 14-12-6-10-8 -- and this was telling us to have a bit more confidence in the markets.
So, back to now. Here's the 2012 LCR table from above:
We have a buy signal with the close of markets on 11/21/12. We have a sea of green on the right, with just a tad bit of weakness on the 2d slope-of-slopes. We have strong summation values from the 16th: 8-10-4-6-8. I would have liked to have seen some +14's in there, but overall, this is a valid buy signal.
This isn't to say that the next week will not be down. It could be. I don't try to predict the future, I only try to establish what the current trend is.
I'm looking carefully at stocks to take a position within; my short list for review is the following:
Many of these stocks have held up quite well in the recent downturn, so already have some form of "Long" status. The stocks that are rated as "Cash" are the ones I'm watching like a hawk, as these could signal a "New Long" and I'd want to move on them (especially if they have favorable EV).
Overall, I think it important to state that we are very early in this signal, so it is prone to failure. My intermediate timer is in MIXED state off of being in CASH, indicating that the markets WERE in cash and are now transitioning in a bullish fashion, and my long-term timer is in CASH, indicating that the long-term trend is still DOWN. I wouldn't bet the farm right now, although 50% may be a good starting point if the right opportunities present themselves.
As always, do your diligence, and take ownership for your actions. Let the trend be your friend.