Last month I wrote an article about the fat tail distribution of the stock market and how that can mean huge, statistically improbable moves for stocks. In a follow-up article, I went on to talk about how we as investors should view those moves. Today I'm back at it again, in an ongoing effort to highlight the companies making these "5-sigma" moves.

Now just to briefly rehash, a 5-sigma move is when a stock has a one-day price move that is five standard deviations or more from the stock's average one-day change. Because we're looking at the price change relative to the stock's historical volatility, it's more than just a look at the same ol' jittery stocks making the biggest absolute moves. So although CIENA, Redback Networks, and AMD had some big percentage changes last week, you're not going to see them here, because of their higher average volatility.

As I showed in my original articles on 5-sigma moves, working with these stocks isn't as easy as selling every stock that makes a big move up or buying every one that does the opposite (or vice versa). It's crucial to understand the circumstances of the move and figure out whether it is indicative of fundamental weakness in the business or just short-term trading dynamics. In some of the historical cases that I looked at, stocks continued on a major upward march even after a huge one-day move -- similarly, there were stocks that managed to continue to lose ground even after a huge one-day fall.

Here are a few of the 5-sigma's from the past week:

Stock

Date

Change

Sigmas

Nuvelo (NASDAQ:NUVO)

12/11/06

(79.3%)

16.4

Hubbell (NYSE:HUB-B)

12/11/06

(12.7%)

10.9

EPIX Pharmaceuticals (NASDAQ:EPIX)

12/12/06

30.3%

7.2

Spartech (NYSE:SEH)

12/12/06

(15.5%)

6.6

Companhia de Saneamento Basico do Estado de Sao Paulo (NYSE:SBS)

12/13/06

15.1%

6.4

Alfimeprase is out
Want to see a grown man cry? Show him the chart for Nuvelo. The stock dropped a breathtaking 79% as the stock was cut from $19.55 to $4.05 and the company went from having a $1 billion market cap to being a sub-$250 million company. This was all at the hands of an announcement of poor results from tests of their phase 3 drug, alfimeprase.

So how exactly do poor test results on one drug cause such massive value destruction? Part of it is the fact that Nuvelo had quite a good year in the market up to December 11. At the very beginning of the year, the company announced a partnership with Bayer (NYSE:BAY) that gave Bayer the overseas rights to alfimeprase. For Nuvelo, the partnership meant royalty payments from Bayer for future overseas sales, but more importantly it meant cold, hard cash -- $50 million up front and more when certain milestones were reached. The stock jumped 41% the day the partnership was announced, and continued to climb through the rest of the year. By the time last Monday rolled around, the stock was up well over 100% for the year.

Alas, it wasn't to be. Alfimeprase was being tested in two separate phase 3 trials for different applications of the drug and both trials failed to meet either primary or secondary endpoints. In addition, as the rock star of Nuvelo's pipeline, alfimeprase was also being primed for trials for other treatments -- trials that may now be reconsidered. Sure, Nuvelo has some other compounds coming up through the minor leagues to try to make it big, but alfimeprase was in the driver's seat, and investors appear to be putting very little faith in any of the others making a big splash.

Dimming lights at Hubbell
Hubbell, a manufacturer and distributor of various electrical products -- including lighting fixtures and wiring systems -- and typically a pretty stable stock, saw a swift 13% haircut last Monday when the company announced that it was cutting its estimates for 2006. In a press release, the company announced that it is now expecting $40 million less on the top line and that EPS would be about 10% lower, between $2.53 and $2.58 versus previous guidance of $2.80 and $2.90.

CEO Timothy Powers attributed the expected shortfall to a faster than expected drop-off in residential construction spending, softening nonresidential construction spending, and high inventories at utility customers. Because the lower revenue was unexpected, overhead was not reduced and the effect on profit was significant. Though Powers said in the press release that management is disappointed with how the final quarter of the year is unfolding, it's worth noting that in Hubbell's third-quarter earnings press release Powers called 2006 the "peak year of initiatives and change."

Over the past few years, Hubbell has been in the process of streamlining the company and integrating a number of acquisitions, particularly the $250 million acquisition of U.S. Industries' lighting group in 2002. Something seems to be working, as the company has grown revenue and net income at a 12% and 36% CAGR, respectively. Of course, much of that time included the residential construction boom that has just started cooling in the last year or so.

Earnings in the company's new range would mean a drop of 4% year over year versus growth of about 7% with the previous guidance. Foresight to scale back production, and therefore overhead costs, could help Hubbell deal with a tougher business environment, but with no immediate remedy expected in the residential construction market, further softness in the nonresidential market could put a big wet blanket on Hubbell.

For more statistical outliers:

Fool contributor Matt Koppenheffer is happy with consistent returns, but wouldn't mind seeing one of his stocks on the 5-sigma list to the upside. He does not own shares of any of the companies mentioned. The Fool's disclosure policy is always statistically sound.