One year ago yesterday, I wrote a small article about the derivation of my Predictive Modeling for the S&P 500, included its output for the coming year of 2006. I also gave a list of stock picks, to boot! Not just any list, mind you, but four lists of 12 stocks broken into my expectations for the year to come.

You can find the original 2006 predictions on MarketThoughts or on Safehaven. Look them up!

My S&P 500 projection was as follows:

The 52-week model for the S&P 500 is currently reading in the 6th decile. The 6th decile is neutral in the sense that it suggests an average 52-week return. Median and average returns for this reading are 13.4% and 11.7% vs. standard median and average returns of 12.4% and 11.8%. The 6th decile implies a 79% chance of gains, with about a 52% chance of above-average gains, and a still-hefty 13% chance of ending the year down more than 10% from where we are today. Overall, the model output suggests a U.S. stock market year that is somewhat typical and not anything to be especially excited about or afraid of.

We finished this year with a +13.6% gain on the S&P 500, not including dividends, an above-average gain and just barely higher than I thought. If you took the midpoint of my "median and average" range of +13.4% and +11.7%, making that +12.55%, I was 1.07% off, which was closer than all but 8 of the 76 analysts that Business Week polled in Dec 2005.

I update my modeling on the S&P 500 every week, under the category heading General Market Commentary. Look for posts labeled "Predictive Model Output." I will review the charts and the projections for the next month and year.

Individual stock predictions

Here's a graph with the performance of all four "portfolios" versus the SPY. Note: all returns and stock prices are back-adjusted for the impact of dividends. For the index, the dividend on the SPY was used.

LCOP = Large Cap Out Perform
SCOP = Small Cap Out Perform
LCUP = Large Cap Under Perform
SHRT = Short These

All of these portfolios were assembled using valuations that I call "fundatechnicals" as well as some purely fundamental screening. The LCOP and SCOP lists were directly based on a "value" approach, back-tested through historical data. The SHRT list was based on an "anti-value" approach, similarly back-tested. The LCUP was a stretch of the SHRT ideas, loosening the criteria until I could find a dozen large caps that were close to the "anti-value" concept. In retrospect, I'd like to have that list back, but the other three performed up to expectations over the course of the year.

Large caps that I thought would OUT perform

Of the large cap stocks that I picked, six outperformed the S&P 500 and the equal-weighted portfolio of all twelve would have beaten a similar portfolio of SPY, chosen because I wanted to include dividends in the returns.

The average large cap return on my picks was +24.5%.

Small caps that I thought would OUT perform

Of the small cap stocks that I picked, five outperformed the S&P 500 and the equal-weighted portfolio of all twelve would have been beaten by a similar portfolio of SPY, but by less than a full percent.

The average small cap return on my picks was +14.6%. RCKY and CRMT killed this group, although in real trading a trailing stop loss would have preserved profit in CRMT.

If you combined the returns of the 24 "value" picks I gave on Dec 29, 2005, into an equal-weighted portfolio, the gain would be +19.6%, outperforming the S&P 500 by 4.3%. 20 of the 24 posted gains, 11 of the 24 beat the index, and 8 of the 24 posted returns more than double the index's return.

Large caps that I thought would UNDER perform

This was the one I was least comfortable with, as I felt I had a good handle on the metric for long-term underperformance, but very few "name" large caps met it; so I stretched the metric to find these. Mistake! These were growth stories that were large and overvalued, but not overvalued enough to trigger my "anti-value" screener. Oh, bother.

Of the large cap stocks that I picked to underperform, six outperformed the S&P 500 and the equal-weighted portfolio of all twelve would have beaten a similar portfolio of SPY.

The average large cap return on these picks was +22.0%.

Interestingly, these are the large caps that got hammered the most during the summer swoon, and they didn't start recovering until September (see chart above). Swoon aside, many of these stocks presented excellent shorting opportunities, and some of the big "you're crazy, Bill!" names on the list, like DNA and GOOG, were huge disappointments to own.

My SHORT list

From my back-testing, I was pretty confident that these would give underperformance as a group over a year's time, and provide good shorting opportunities. I was right.

An equal-weighted portfolio of these stocks would have lost money in a year that the markets returned double-digits! Combined return of -1.5%, with seven of the twelve losing money, six of those losing into the double-digits! Only three of the twelve beat the SPY, although the ones that did, thrashed it. Those were BIDU, SLW, and THRX. BIDU actually would have been a good short through February, although my broker didn't have any. Oh, bother.

These were exactly the stocks you wanted to be short on during the "summer swoon" (see chart above). These guys got murderized! As of June 13, this group was 31.5% off its May high as a portfolio.

One of the worst stocks on this list in terms of annual performance actually didn't provide any obvious short opportunities, considering its -74% return. THLD collapsed as many a biotech can, and unless you were short before the close on May 11, too bad, so sad, call your dad, he'll be mad. Other than that, just about every other stock on this list provided a class act short opportunity at one point or another.

The most interesting thing, to me at least, about the "Short" portfolio was that it is generally better to trade these stocks for the round trip, going long when they first hit the screener, and shorting them with abandon when they break, or on a market turn. In 2007, I will be far more aggressive on these picks.

If you combined the returns of the 24 "anti-value" picks I gave on Dec 29, 2005, into an equal-weighted portfolio, the gain would be +10.3%, underperforming the S&P 500 by 5.0%.

No stock predictions for the coming year, folks! I am keeping a public watch list on StockCharts.com, however, so you can follow along with my picks and my trades as I make them.

As a reminder, you can find the original 2006 predictions on MarketThoughts or on Safehaven. Look them up!

Best of luck and happy trading to you in the New Year!