David Merkel asked Ritholtz where exactly above a 52-week low he'd be willing to enter a stock, and Ritholtz explained Saturday how he prefers to wait for a trend to re-establish before entering a low-range stock.

This discussion between his colleagues prompted James Altucher to backtest all Nasdaq 100 stocks since 1996 to see what would have happened if one had bought stocks hitting 52-week lows and then sold them one quarter later. However, his column outlining the results did little to sway Ritholtz, who defended his original statement and responded to Altucher's research in another Columnist Conversation post.

Altucher addressed the queries in a second post, and now both he and Ritholtz are working on follow-up columns on the subject.

Today, Richard Suttmeier offered a technician's view of the debate with a column that argued buying a stock near a 52-week low is a good strategy so long as there's support nearby from a value level.

Look for more commentary in the coming days. In the meantime, read on to catch up on the action thus far.

Barry Ritholtz
7/21/2006 1:03 PM EDT

Some readers criticized the "never buy a stock making a 52-week low" statement I made in this space back in May.

I strongly believe that it is solid advice.

To reiterate why, stocks in downtrends tend to stay in downtrends for much longer than most people estimate. Since we do not know when technically weak stocks -- those under institutional distribution -- will stop being sold by the big boys, a 52-week-low purchase is merely a guess that the selling is just about finished.

A few specifics: Since that May 18 post, eBay (EBAY) is down 25% and Dell (DELL) is off 40% from roughly the same time.

The best performer has been Microsoft (MSFT), which is flat since this post.

I cannot take credit for the "Don't buy 52-week lows" rule, so this is not crowing -- rather, it is to point out that a well-crafted technical rule, which has been historically proven to have value, is not to be so blithely ignored.

Long cash, SPY puts

Click here to see this post as it originally published.

David Merkel
7/21/2006 3:43 PM EDT

I sometimes buy stocks at 52-week lows, small amounts to rebalance to target weights. It usually works for me, because I try to avoid companies with bad balance sheets. I don't tend to get many "death spirals." (I do get a few, though.)

Barry, is there some sort of cutoff for where you might buy a stock above the 52-week low? Does it have to be 10% above, or some other criterion?

Click here to view this item as it originally published.

Barry Ritholtz
7/22/2006 8:09 AM EDT

It's not the lows, but what they mean: I advise individuals not to buy stocks making 52-week lows for a wide variety of reasons -- these stocks:

• are often in a downtrend
• are part of a negative sector or group
• have fundamentals that may be decaying
• are possibly in markets that may be in a confirmed bear.

For most investors, it's better to miss that first 5% to 10% move off of the bottom and wait until a trend is re-established.

Consider H-P (HPQ): It had quite a few false bottoms and fake reversals, but you could have picked it up in early 2005 around $20 and watched it gain 50%.

No, you didn't bottom-tick it, but think of how much damage investors bottom-fishing the likes of AOL, CSCO, EMC, DELL, INTC, LU MSFT, SUNW, NT, QCOM, YHOO, etc. did to themselves.

Never say never, but for most retail investors (and quite a few funds), they may be better off saying "extremely rarely."

No positions in stocks mentioned

Click here to see this post as it originally published.

James Altucher
7/24/2006 3:00 PM EDT

There was an interesting discussion over the weekend in Columnist Conversation between David Merkel and Barry Ritholtz on whether or not to buy stocks at 52-week lows. Barry's conclusion was that investors should mostly avoid stocks making 52-week lows because they are often in a downtrend, are part of a negative sector, have fundamentals that might be decaying and are possibly in markets that might be in a confirmed bear phase. I have a couple of issues with these statements, but first let's do some back-testing.

I took all Nasdaq 100 stocks since 1996, including stocks that have been deleted from the index (to avoid survivorship bias). What happens if you buy stocks hitting 52-week lows that are trading for greater than $5 (avoiding penny stocks) and sell them one quarter later?

The results actually demonstrate that, over this period, the odds were on your side to outperform the market if you bought stocks at 52-week lows. The average return per trade was 7.34% (over 662 trades), including wins and losses. This far outperforms the average return per quarter of the Nasdaq during this period of 2.6%.

Some 60% of the trades turned out favorably, and 40% were failures.

You won't always catch the exact bottom, but that's impossible. The key is, for any investing, the maxim "buy low and sell high" can be extended to "buy 52-week lows and sell 52-week highs."

The other issue I have is with Barry's statement that stocks making 52-week lows "are possibly in markets that may be in a confirmed bear." What is a confirmed bear?

March 2003 was a dismal month, yet that was the beginning of a huge bull run. At the end of July 2002, the market gave every indication of breaking out, yet it fell to new lows. Ditto for the end of 2002 -- the slide into February-March 2003 was certainly scary.

I've tested thousands of market-timing systems, trend-following systems, breakout systems and mean-reversion systems, and I don't think there's any way to reliably market-time for the downside, although there have been several largely successful mean reversion-based systems that can help time for the upside and have worked in bear and bull markets.

No positions in stocks mentioned

Click here to see this column as it originally published.

Barry Ritholtz
7/24/2006 4:07 PM EDT

James Altucher wrote an interesting column today on buying at 52-week lows. As of now, I am unconvinced of the wisdom of buying 52-week lows by his back-testing.

But I am intrigued by the analysis James put together. So I'd like to ask the following questions:

1) How did non-Nasdaq stocks do under this methodology? How would the Wilshire 5000, the S&P 500 or the Russell 2000 stocks hold up when buying 52-week lows?

2) The 1996 to 2006 time frame is an aberrational period in history (to say the least), thanks to the bubble top and crash. How does this strategy work between, say 1950 and 2000? 1906 to 2006?

3) Which 52-week low do you buy? All of them? For example, have a look at Sun (SUNW) or Cisco (CSCO) from 2000 to 2002 -- they seem to be making 52-week lows every other week. Is your back-testing model buying all of them?

4) Lastly, why hold a quarter? How were the results for a week, month or a year?

You wrote a very interesting piece, and I'd be curious about how all these other elements play into your findings.

No positions in stocks mentioned

Click here to see this post as it originally published.

James Altucher
7/25/2006 7:35 AM EDT

Barry had a couple of good questions related to my article on buying 52-week lows. I have some answers:

Barry asks: "How did non-Nasdaq stocks do under this methodology? How would the Wilshire 5000, the S&P 500 or the Russell 2000 stocks hold up when buying 52-week lows?"

I haven't tested the non-Nasdaq (although there is 99% overlap between Nasdaq 100 and S&P 500). I don't believe there will be a huge difference in results, but in the next few days, I'll test out on the S&P 500.

Barry: 2) "The 1996 to 2006 time frame is an aberrational period in history (to say the least), thanks to the bubble top and crash. How does this strategy work between, say 1950 to 2000? 1906 to 2006?"

Well, if you can get me all the data on all stocks (including stocks that got bought or disappeared so we can deal with survivorship bias) from 1906 to 2006, then I'll test and maybe we can get a joint Ph.D., Professor Ritholtz! 1996 to 2006 is a good sample for a couple of reasons:

• It has a bull market, a bear market, choppy markets, rising interest rates, declining interest rates, etc.
• It's the most recent 10 years, so probably most reflective with how the market is today.

No, I'm not buying any 52-week lows that occur during the three-month holding period. In other words, I'm buying the first 52-week low.

Barry: "Lastly, why hold a quarter? How were the results for a week, month or a year?"

Arbitrary. I tested on six months, and the results were 9% for a six-month holding period. I tested on one month and the results were similarly good.

Rather than defining a thorough trading system, I'm mostly just trying to make the point that buying 52-week lows is probably better than not buying them or than simply buying the market. I'm sure there are many specific examples or periods where this approach does not work. But in general, it seems to work.

No positions in stocks mentioned

Click here to see this post as it originally published.

Richard Suttmeier
7/26/2006 7:52 AM EDT

My model is based on the premise that investors should consider buying weakness to a value level in stocks that are undervalued and oversold, and this often occurs when a stock is achieving a new 52-week low. Conversely, investors should consider selling strength to a risky level in stocks that are overvalued and overbought, and this usually occurs as a stock is forming a parabolic peak at a new 52-week high.

No one can buy at the exact low or sell at the exact high, but I believe the most effective strategy for generating profits from the market is to own core positions in stocks that are appropriate for your investment style and then trade around those core positions. One way to effectively make trading decisions is to use my three-pronged approach, which includes a fundamental valuation, a technical profile and the value and risky levels.

Let me show you an example of how this would work. Take the example of rebuilding a technology portfolio on weakness in July using the following screening guidelines. To be considered for purchase, the stock must:

• Rate a hold or better according to ValuEngine.
• Be at least 20% undervalued vs. its fair-value price.
• Have a market cap of more than $5 billion and trade above $10 per share.
• Have a weekly chart profile that shows oversold momentum (12x3 weekly slow stochastic).
• Trade near a 52-week low supported by a value level at which to buy on weakness.

In columns written in December and January, I suggested that investors consider reducing holdings in the highflying technology stocks by up to 50%, as shares rose above 52-week highs. At highs, investors should employ exit strategies to either reduce holdings on strength to a risky level or on a sell stop to protect gains after a 52-week high has been achieved.

Technology stocks are arguably the most volatile in the market, and when they became overvalued and overbought as they were in the first quarter, I suggested exit strategies. This month, the same technology stocks became undervalued and oversold, and I provided entry strategies.

Click here to see this column as it originally published.