This page has been archived and commenting is disabled.
What Do the Numbers Say?
It is rare for me to make a statement regarding the markets unless I have some data to back me up. In the weekly sentiment wrap up,
I stated: “since 1991 covering 40 unique instances, the SP500 bottomed
80% of the time with a reasonable risk (<6% draw down) to capital
after two consecutive weeks of extreme bearish sentiment; and 2)
failure of the market to bottom is a very ominous sign.” So let’s
drill down a little deeper and see what the numbers say and why I would
make such a statement.
Figure 1 is a weekly chart of the SP500
(symbol: $INX) with the “dumb money” indicator in the lower panel. So
here is the study: 1) buy the SP500 when the “dumb money” is bearish for
two consecutive weeks; 2) we sell our position when the “dumb money”
indicator turns neutral. In other words, we are only in the market when
the “dumb money” is green in color, which means investors are too
bearish. The data goes back to 1991, and all positions are executed at
the close of the week.
Figure 1. SP500/ weekly
Since 1990, there have been 40 unique occurrences or trades yielding
858 SP500 points. Buy and hold only generated 950 SP500 points over the
same time period. The strategy was only in the market 20% of the time,
so in essence, the strategy yielded an almost market beating
return with only 1/5th the market exposure. That is pretty strong. 80%
of the trades were profitable. The average trade lasted 5.5 weeks. So
far, the results are fairly compelling, but to get an even better
picture of why I would make those statements, let’s look at every trade
from this strategy. To do this we need to look at the maximum adverse
excursion (MAE) graph. See figure 2.
Figure 2. MAE graph
The MAE graph looks at every trade from a strategy and determines how
much a trade had to lose before being closed out as a winner or a
loser. If you are like me, you put on a trade and watch as it goes
adverse to its entry price. MAE is the angst factor. At some point,
the position is closed out for a win or a lost. For example, look at
the trade inside the blue box. This trade initially lost 5.35% (x axis)
before being closed out for a 1.77% (y axis) winner. We know this was a
winning trade because it is a green caret.
Out of the 40 trades, 33 had MAE’s less than 6%; this is to the left
of the orange vertical line. In addition, 31 of those 33 trades were
profitable. So if you buy the SP500 when the “dumb money” indicator is
bearish for two consecutive weeks, then you have an 80% chance your MAE
or trade draw down will be less than 6%. To the right of the orange
vertical line shows all the failed or bad signals. With 5 out of the 7,
the MAE was greater than 10%, and 6 out of 7 were losing trades.
That’s why I can state that bad signals are ominous.
In sum, I still believe the data supports betting with the market
when the “dumb money” indicator turns bearish. It doesn’t work every
time, but show me something that does.
If you would like to have TheTechnicalTake delivered to your email in box, please click here:
It’s free!!!
- advertisements -
Dumb money, smart money, I do not give a crap. I will not buy this shit for a market. It is rigged. It is false, it has nothing to do with reality, and I will not give it my confetti. OK, so I am dumb.
Got my physical, miners on watch and a couple of shorts. This govt. makes me want to puke.
+100 Why bother, who cares, you don't have to own anything in the stock market to make money.
Especially when 90% of trades shown are phony. Phony trades give you phony statistics and charts.
Dumb money proves itself when the market goes up, because in the end all you get is money, which just happens to be worth less than it did when you put it in. But that's what happens when the metals are going up.
1991?
Please.
How about going back to 1921?
1911?
Hmmm...?
HFT's are bearish? hmmmm
Technical analysis... you mean the technical VanGogh chart/painting?
Look in the mirror to see the dumb money.
Major false breakout, marked up distribution = dumb money buys, or dumb 'smart' money. Thin liquidity etc etc. Dow touching 12000 on air. i'll run copper and metal charts against futures. i think the china RRR hike, plus other interest rate hikes (HK etc) should send stocks reeling, that and EZ and US are a mess. the only dip buy is on huge CB money printing..
also watching risk crosses (FX) which set bull-traps last 24hrs
I don't think this is the time to go long more than intraday. All of QE II went to capitalize foreign banks and to prepare to push the PIIGS off a cliff. The rest of the run up seems to have come from banks taking on 380 billion in new liabilites and basically tricking investors into thinking that the market was being run up, when it was just a calculated head fake. This market is running on HFT algos, money that wasn't really put into it and expectations based upon deceptions. What news driver could possibly drive this market up and energize the bulls? I can't think of ANYTHING that would override the tide of bad news.
Stop me if I had this reversed! But is it technical or just some kind of contrarian to follow the (I think) Richard Ney "system"?
What he did was watch an entities volume and trend-line. If it declined on low volume it was a loser, then - go sell. But if volume is low and it is rising/flat he would accumulate.
Then, when he concluded the insiders are moving prices he would do the reverse by watching when volumes were very high. (That is when you should sell high items to the suckers like Wall Street does, and of course buy things cheap when high volumes and blood is in the street.)
I think Ney was one of those esteemed Street guys who got into it after a profession of sports.
Oversold indicators work in a bull trend or a trading range. On the weekly chart the bull market from 09 is still intact. Keep taking that signal until it doesn't work. Then it is probably time to start selling the over bought signal. Use your own stop since the author doesn't discuss money management.
Mkkby:
Actually the money management is right in front of you and it is the MAE graph; more than 6% loss tells you something is wrong
Also put in the length of time for each trade....what more do you want? I can't remember ever seeing this kind of info in article
I agree, we are oversold, but on a long term cyrclical PE basis the market is hugely overbought. 5 up days will change the tenor of the market
"It doesn’t work every time, but show me something that does."
The trading desks at the big Wall Street banks often boast of 100% positive trading days in each quarter. But I'm not recommending that you, or we, become crooks as well.
The graph in Figure 1. clearly shows that for the majority of the shown time period the 'dumb money' was correct and it would have been profitable to follow the dumb money! That's enough for me to ignore this contrarian junk.
Figure 1 shows seven completed green-to-blue cycles, where one buys the beginning of the green and sells the beginning of the blue. The candlesticks are weeklies, and difficult to align with the signals, but my best evaluation is thus:
1. '08, late June to Mid Aug: break even
2. '08, Aug to early Nov: big loss (depends upon stops)
3. '08, mid Nov to Dec: break even
4. '09, mid Feb to mid March: slight gain (1 to 3%)
5. '10, Late May to Mid June: break even
6. '10, July to Sept: 5% gain
7. '10, Sep to Oct: 3% gain
If one cherry-picks the entry points within the weekly candles, a few more percentage points of gain can be squeezed out. That's why I asked for the definitive buy point in the earlier comment.
Mr. TechnicalTake, your June 12 wrap-up says that the insider buying indicator is neutral, and that the Rydex funds are bullish (many more longs than shorts), and this is a contrary indicator which you take to be bearish. Therefore, two out of three indicators say "don't bother buying this dip", at least not yet. Why follow these indicators if you disregard them?
I'm not trying to be a smart-ass. As a structural engineer, I always try to know as much as I can about a project before we start pouring cement.
Skeptic:
Great question and glad you were paying attention....my interpretation of these discrepancies is that the market will bottom but the bounce will not be strong....it could be a 30 trading day rally (weak) v. 100 day rally
For us viewers at home, we need to know when you went long, and at what price. To evaluate your success, or lack thereof, we will need to know when the indicator flashes a sell signal, when you got out and at what price.
This is a real trade, and not just paper exercise, isn't it?
Not that we are disbelievers, but too often these back-tested models seem to break down when actually used.
It looks as though the "dumb money" is pretty smart. It got in in April 2009, got out a year later, then got in again in fall 2010.
...the 40,000 feet thing, pretty good actually.
I love the technical / fundamental discussions, both have taken a beating lately. Honestly...can anyone make out a banks financial statement today? Less likely. Does it matter? For the chartists...there is no doubt that patterns exist, reasons aside, just look back in horror as they repeat time and again seemingly no matter the circumstances or drama of the moment! But is the reason that we are up because of a technical formation?? At the end of the day, all that matters is the balance in your account.....that balance is moving along with the price of whatever you own. I can't bet on some business ratio that the company gives me, highly educated well paid liars, any more than what I believe the wave count to be! Sometimes the hardest thing to do is go with the price and manage risk accordingly. Is it impossible for the S&P to trade below 1,000? 400 may be a bit much but who knows. Why do people rule out a sustained downtrend in this market.
So, you're not in the market right now?
Who knows what this guy is doing. He contradicts his own stuff. Much like EWI's stuff. Garbage in, garbage out.
I have no clue what his stuff is trying to say. If you are like the MM's who determine reality. You could trade just like GS, JPM and BAC who had 90 straight days of perfect trading. I am willing to go out on a limb here. But, I bet they got the last week or two perfect also. Why not, they do not look at the charts. They are the markets determinate.
You can literally throw 'technical analysis' out the window when the aircraft has a giant hole in its fuselage at 40,000 feet high.
Such is the case now if the world were an aircraft. The hole is absolute and leverage debt.
what stops me is the "40 occurrences since 1990" (of a dip in the market that subsequently rose in the environment of dumb money bearishness). if you are a short term trader, that's important. if you are trying to avoid (or even profit from) the secular bear market since 2000, not so much.