5% Rules! How Can We Be So Right? (Members Only)
While I love a good Ouija board as much as the next guy, I'm not big on TA.
What I am a fan of is the law of large numbers, fundamental analysis, freakonomics, chaos theory, psychohistory and crowd psychology – all in the elusive search for the objective truth, which I have on good faith is 42.
Surprisingly enough, based on this outlook, I get a lot of consulting work and sometimes that consulting work led me to consult on trading systems and that led me to figuring out a very useful system for predicting the move of the markets without charts (though I'll use them to illustrate points) and without all the silly fuss that TA guys make about their "craft" (sorry Dave).
The 5% Rule does NOT tell you which way the market is going. It does tell you where the resistance points will be. Of course, knowing that and knowing what kind of bounces to expect and knowing where a proper breakdown or break-out occurs is kind of useful and, when it coincides with the tea leaves that are read by the "real" TA guys – you can really have something good to go by!
Unfortunately, the 5% Rule is not really a RULE because it requires a cynical background in statistics, especially regarding aberrant values or "outliers" and a general understanding of market history as well as current market events because all need to be taken into account in order to give you accurate "consolidation levels" from which we base out chart movement.
For example, not to get too into it now but a while ago I pointed out that our V-shaped recovery on the S&P, etc. wasn't all so impressive if you consider that the drop from 800 to 666 was simply a panic spike and was not statistically relevant. That means that the S&P is not up 80% at 1,200 but is, in fact, up 50% off what we call the "non-spike" floor at 800 after topping out at 1,576.09 in October of 2007:
There are a lot of things going on here but I wanted to impress on you that they are, to a large extent, the same thing! The Fibonacci series, which is a study of regressive sequencing that incorporates elements of chaos theory, statistics and the law of large numbers and tends to work very well when studying market movements. Without getting too off topic, the key point is you can expect to hit resistance points between two significant points (like a high and low) at 23.6%, 38.2%, 50% and 61.8% due to a tendency of larger and larger sequences to converge towards a mean. Suffice to say it works well enough that most charting programs include the tool…
Our 5% rule is fairly simple. We expect major market moves to trigger in 5% blocks. Why is this? Because all the clever programmers and all the captains of industry and all their little consultants all get together to design the ultimate trading system but they all end up rounding off here and rounding off there and, ultimately, the decision to push the button comes down to a person (or, even worse, a group) who then end up being affected by psychological resistance points as well as the usual mob psychology that dominates the markets. The reason Goldman's trade-bot is able to convert winning days 93% of the time is they are NOT trying to do any of these things – they are simply inserting themselves in the middle of a transaction to swipe pennies – that's not trading but it sure is profitable!
MOST market participants actually buy stocks and they try to buy low and sell high and they have profit targets and stop losses, etc. Over 1/2 the funds playing the market rely on various TA methods for entry and exit points and many funds use Fibonacci series in their calculations so it all becomes a self-fulfilling prophesy. We take advantage of that by effectively averaging out the results which gives us a general rule that stocks tend to move in 5% incriments before a 20% retrace (keeping in mind that Fibonacci looks for 23.6% so it's an "about" number).
The more liquid something is, the smaller the 5% Rule breaks down so when I say "the 2.5% Rule" or the "1.25% Rule" or the "20% Rule," it's all the same rule. What becomes critical is the bounce zone. If we're watching a movement, like the S&P drop above, we see a floor forming at 1,240, which we expect at 20% below the consolidated top at 1,550. We also expect a 20% bounce at least as a weak bounce while a strong bounce is a 50% retrace but neither one violates the trend. As the S&P is a large index, we got first a strong bounce, retracing 1/2 the 310-point drop from 1,550 to 1,240 and back to 1,395 and you can see that we consolidated there, then fell back to 1,240, consolidated (not bounced) there, then bounced 20% back to 1,302, then broke 1,240 and plunged to our doom.
How far did we plunge? Just about 25% to 930 (another 310 by the way) before weak bouncing to our expected test of 992. 20% bounces are a sign of weakness, not strength! In the 5% Rule, 10%, 20%, 25%, 40% and 50% tend to be the strongest turn indicators – 10% because a lot of pshychology kicks in, 20% and 25% because they straddle the first Fibonacci level at 23.6% and then 40% and 50% are also close to Fib levels at 38.2% and 50%. Never mistake the 5% Rule for an exact science, we are looking for consolidations around those points and we are very happy to throw out our targets in favor of exact numbers, like 800 on the S&P rather than 813.75, which is the 12.5% decelerating leg we expect if things are going to turn around.
So the gravity of a big number like 800, especially when it's nearly 50% off the top, trumps our expected 813.75 line but I tend to use both and look for strong intersections that line up from two points. It's also important to note that a 20% move off the bottom will NOT get you halfway back from a 40% drop from the top so, again, we need to look for intersecting points and defer to Fibonacci as well as psychological levels in those areas. Still, we got out 12.5% drop, which was 1/2 of the 25% drop and then we're looking for 3 bounces: A bounce of 20% is always expected and anything less than that (non spike) means we're weak and probably legging down. A bounce of 50% is strong but no consolidation and falling back below 50% means at least a re-test of the 20% bounce.
For the S&P at 800, we're looking for a 20% bounce off 775 (real support from 1,550) back to 930. Since 930 was already a magic number on the way down, we are very excited to see it coming into play on the way back up as that makes it "significant." And, of course it's significant, it's a 20% long-term bounce off a 50% drop of a major index! Also, our last drop from consolidation was from a 1,300 top to 800 (38.5% Fib series) and that's an expectation of a 100-point bounce back to 900 and then we expect a bounce off our fall from 930 to 800 and that's 26 points back to 826.
As you can see, we bounced back hard to 930, consolidated there but failed, then found some support at 826, made another run that failed at the halfway mark between the two and then fell about 25% more but came back exactly as fast as we fell so we can ignore the whole thing (although that's hard to do while you are in the middle of it, isn't it?).
You can see on the way back up how nicely everything lines back up and we have multiple lines of convergence in this zone as the 50% move up from 800 is 1,200, the 20% drop from the 1,550, non-spike top is 1,240 and the 38.2% Fib level is 1,228 so you are in for a slog around that zone and we need consolidation there in order to break back up. That's why I was not at all impressed with anything that didn't end up with the S&P over 1,200 – that's still 40 points shy of showing proper strength and, of course, fundamentally, I don't see that the market is "worth" 80% of what we were at the top at the moment.
So, 5% Rule is VERY useful for long-term planning. What about short-term? Hopefully we've establised that 1,200 is our consolidation point for the S&P and, if that's true, then we can expect that a 2.5% move up or down from there would be significant and even a 1.25% move should hold some weight. Those levels around 1,200 are 1,230, 1215, 1,200, 1,185 and 1,170 so, if 1,200 is our mid-point, we'll be looking to see where we're getting resistance and which side of our series the index is trending to:
.jpg)
See – this isn't hard (other than the fact that I accidentally used 1,218)! Also note that our last consolidation point at 1,100 would have had a 5% series of 1,155, 1,114, 1,100, 1,073 and 1,045 and you can see how that range worked out on the chart. The funny thing about 5% Rule numbers is that we could have written these numbers down in 2008 (I did!) if someone wanted to know how far the markets would fall and where we were likely to recover to.
So what lies ahead? Most likely a retrace back to 1,100 (25% of our run) but if that holds and we consolidate a bit, I will be downright bullish. I will also be impressed if we hold 1,145, which was our last breakout line but, for now, we have a 3.75% drop from 1,218 but a poor bounce yesterday indicates we are likely to get down to a 5% pullback from 1,218 to 1,157 and not holding that is going to be nasty. For now, we have the rising 50 dma and the 2.5% line at 1,170 so VERY BAD if we can't pull a bounce together here. We'll call it a 50-point drop from 1,220 and figure a 10-point bounce to 1,180 would be lame (a very important 5% rule term) and 25 points, 1,195, has got to be retaken before we get all impressed with a "rally."
Consider this a work in progress. Let's keep all 5% Rule quesitons on this post and, over time, hopefully I can refine my explanation as this is pretty much the first time I've tried to lay it out. Like all "systems," it's important to pick one that fits the way you think. This one is perfect for the way my brain is wired as I can do these numbers in my head as soon as I look at a chart so it let's me know if a stock or an index is "behaving" itself or not. That last bit is an important point – no matter what system you have, there are going to be some things it works on and some things it doesn't. DON'T force your system on a stock – find stocks that work with your system and you'll both be happier!
 Twitter
|
LinkedIn
|
 del.icio.us
|
|