Online Stock Trading

I should warn you in advance that this article is both long and tedious. However, through that length and tedium you will find that it also offers a rigorous education for those willing to step into the deep. The article includes a total of twenty-one graphs which illustrate various stock metrics. At the end of the article there is a link to a “.zip” file from where you can download all of them if you want to examine them on your own. Also take note, all the pictures are designed with code so that they will open in a new tab when clicked (so you can see larger, clearer images), there’s no need to right click on them to keep the page you’re on.

---

After sifting through data from every single stock listed on the NYSE, NASDAQ & AMEX exchanges from the end of July until the end of December, I can decidedly endorse what we already know, namely that the stock market has performed poorly. Even so, there are, upon an exhaustive examination, some clear trends which may be extremely valuable when looking for the best possible investments under less than best conditions.

The primary merit of this examination is its timeframe: July 20th through December 22nd. Why? Because in the long run the stock market is “on balance” normal, meaning that good stocks with strong fundamentals which are undervalued will perform better than the market as a whole. But in the short run they can get annihilated. Short runs like credit crunches. This comprehensive analysis will examine not how the market does over longer periods, with “normal” conditions, but rather, these short and awful last few months.

Don’t let the fact that between the end of July and the end of December the S & P 500 has only lost 3.24% of its value fool you into thinking that things aren’t so bad. Those are some of our best 500 companies. Between all the exchanges, all the stocks, they’ve lost on average 9.85%. That’s pretty bad. So let’s get started.

If we look at the yield distribution linearly, we can see the spread immediately:

There is a clear steepness for the positive territory in contrast to the large flat slope in the negative territory. Indeed, over 70% have fallen in value. The bias is clear. If we look at the yields of all stocks as a frequency distribution, we see the same information in a different way:

We have very shallow frequencies of stocks which have gained followed by much higher frequencies of stocks in negative territory. In a perfectly balanced, neutral market, this would resemble a bell curve distribution. This shows the same heavy bias towards losing yields. The next nine graphs will show relationships of fundamental metrics as primary catalysts, with the resulting yields over the ranges of those metrics. These will also be represented using a short smooth (200-stock smooth) to extract greater precision at the loss of functionality. The ten after those will be many of the same metrics, only inverted (primary catalyst will be yield, with the fundamental metrics being measured on the y-axes instead) and will use a much larger smooth (1000-stock smooth) which will lose some precision but will achieve clearer trends where there are trends, which is ultimately what we’re after. Also, keep in mind that the fundamentals here are based off July 20th data. In essence, we’re going back in time to see what metrics then performed in the ways they have since.

Looking at market cap, we can see a clear trend:

This is the S & P 500 issue we talked about a moment ago. There is a very clear bias for larger market caps. The biggest companies have fared the best. However, “on balance” no subset would seem to have guaranteed gains, except the very extreme right, the very, very largest companies have, on average, squeaked out mild gains. On the other end, the smallest market caps have been annihilated.

Ok, so this is a credit crunch, correct? Let’s look at current ratios. We would expect that the companies with the largest quick ratios (those who can pay their immediate bills for longer) would have fared better. But this isn’t true:

As the ratios go up, the average yields do definitively tend down (we’ll return to current ratio later during the second set—when we do, this will be confirmed).

Still on the credit issue, let’s look at cash per share:

Here there is a clear bias towards greater cash per share: those with cash per share of 3-5 dollars fared best, but on balance still had negative gains. This metric is difficult because it is hard to know whether more cash per share is a result of simply a company having more cash generally, or just fewer shares. In the next section I introduce a rather quirky statistic (price per cash per share) which intends to test for the “expensiveness” of the amount of relative cash per share a company has.

Next, let’s look at debt to equity:

This one’s interesting. At this level of smoothing, the yields are sporadic through all ranges with the exception of the companies with the worst (greatest) debt to equity ratios. The very worst debt to equity ratios fared the very worst with the rest faring approximately equally.

Let’s next look at some valuations. The following is book value:

This metric seems to confirm the market cap analysis but has the same dilemma as the cash per share we looked at. Later, we’ll look at price per book to dissect the arbitrary dimension of book value outright versus number of shares outright.

This next one is really interesting. This is the 52-week changes of all stocks related to yields:

Look at the left extreme (x-axis). Those are stocks that, as of the end of July, already lost 30%. Since then though, they’ve since lost another 30% or so. Ouch. I guess this confirms what Danny and Bob already know, that stocks doing well will continue to do well, and those doing poorly, well, can get slaughtered.

To see another example of why valuation outright matters nil in this market, let’s look at price to earnings ratio:

Under normal circumstances, we want to find good companies with relatively low PE ratios, meaning we’re buying more earnings with our money and less expectation of future earnings. In this market though, you can pitch that. There is a clear bias toward higher PEs, but not too high—the very highest PEs get slaughtered. We can assume that the extreme highest PEs lost all their speculative punch through risk aversion.

Earlier we looked at book value but didn’t see what we would have expected (namely because book outright may not be relative enough to be useful). Now let’s look at price-per-book to see if this as a valuation is more meaningful:

This would seem to confirm what we found with price-to-earnings: throw valuation out right now. The more expensive stocks in terms of how much book value you are buying per share actually clearly did better, in fact, the “best” valued ones (ones where you’re buying much more book for every dollar you spend) did terribly.

This article is part I of a three-part series.

Here is Part II

Here is Part III