A Dash of Insight

Is This a Tradable Bottom?

We have had some inquiries about our "Gong Model."  They say that no one rings a gong at the bottom, so our marketing department thought this was a cool name.  This article showed a good description of the last time the Gong sounded.

Many traders are seeking "oversold signals" and calling the bottom.

The Gong is not now signaling a bottom, and it is not close.  The Gong model has two parts.  First the hammer must be drawn back, and we are not yet at that stage, nor close to it.  Second, the mallet must come forward.  We'll provide some updates.

Can We Be Wrong?

Of course.  We can certainly be wrong.  If tomorrow's payroll number is surprisingly good, given the +/- 100K confidence interval, the market could rally by a couple of hundred Dow points on a report showing surprising strength.  Our report on The Gong, and other methods, are available to readers on request.

Our intermediate-term outlook has grown increasingly bearish over the last month or so, as documented in our participation on the TIckerSense blogger sentiment poll.  We have reported this both there, and on the weekly updates of our TCA-ETF system.

It is entirely possible that we will have a rally without The Gong.  Also, the gong model gives an entry signal, but not an exit.  We have searched hard for the ultimate bottom-calling method, but we are realistic.  It is not easy.

The Importance of Time Frames

While our system signals have been negative, we have been less convinced by the fundamentals.  In our programs, we have the system (affectionately called the "Vince Model") and the fundamentals, (called the "Jeff Model").  The time frames are different.  The "Jeff'" model is geared to the long-term investor and has a great long-term record.  It is thematic, and the themes have worked over a period of more than ten years.  It is not a trading system, although we obviously try to find the most promising stocks and sectors.  We currently believe that people have become far too negative about the economy and economically sensitive stocks.  Vince sees more pain in the near term.

Readers may be interested in our discussion of the importance of time frames.  We also have written about how a single trade can have two winners -- those who have different time frames or investment objectives. It is not just a question of the immediate stock reaction.

Conclusion

In a difficult market it is important to have one's primary objective in mind.  Traders and investors can reach different conclusions.  One theme is the reaction of the individual investor -- scared out at market bottoms.

Is this the bottom?  Probably not, but that does not mean bailing out of one's retirement account.  We have a nice list of attractive stocks with good valuations.  When the Gong sounds, we will get more aggressive.

Employment Situation Preview: A Risky Outlook

Each month we report the results from our payroll employment model.  This is not really a forecast.  We take a number of different economic indicators from the middle of the preceding month and ask what change in payroll employment is consistent with the other data.  The variables are not causally related, but are all different measures of the economy.

We get a pretty good fit from the University of Michigan sentiment reading (yet another new low this month), the four-week average of initial claims (using the period ending in mid-month), and the ISM manufacturing report.

The data are all consistent with continuing weak economic growth and a loss of about 90,000 jobs.  The market is looking for a loss of 60,000.  Since the 90% confidence interval (sampling error only) is about +/- 100,000, we think that an actual gain is pretty unlikely, while a big loss is possible.

It is interesting that whatever is reported is overly hyped and interpreted as an official count, rather than as a statistical estimate.

Readers interested in learning more can do three things:

1. Take a look at last month's article, where we explored forecasting issues.
2. Check out Little Known Facts about the Payroll Employment Report.  They are still little known!
3. Try playing our Payroll Employment Game.  It shows you the range of different results the BLS might get even from a well-designed survey.  It is cheaper than trading the actual report.

Forecasting the Payroll Employment Number

Each month credit and equity market observers alike pounce on the Employment Situation Report from the Bureau of Labor Statistics.  Markets gyrate wildly when the actual result is 40,000 or 50,000 payroll jobs away from the economic forecasts.

This is a big mistake.  Here is the problem.  The level of change that the market believes to be significant, is not great enough to be measured by the BLS tools.  It is like measuring a hair with a grade-school ruler.

And that is just the measurement issue.  It is even more difficult to forecast the results.  We always enjoy reading Bob McTeer, whose blog is one of our featured sites.  His recent commentary notes that economists are not very good at forecasting.

Unfortunately for the reputation of economists with the general public is that they think the job of economists is forecasting.  That's unfortunate because economists can't forecast very well.  It just can't be done.  Economists know that, and most will admit it, but they are stuck with it, especially if they wish to be employed as an economist outside academia.

His main point is that these predictions are difficult.  People making forecasts should provide probability figures and ranges.  (For a similar argument, check out Barry Ritholtz on The Folly of Forecasting).  The problem is that when the forecaster is honest about the error range, the consumer becomes less interested.  Well, so be it!

McTeer has a number of excellent points on using models and some Greenspan stories, so enjoy reading the entire article.  After noting that most everyone is a "closet extrapolator" he writes as follows:

I've always been amazed at economists who presumed not only to see around a corner, but to see around more than one corner, as in "the economy will pick up through year-end, then slow for several months, and then take off again stronger than ever."  To me that sounds as silly as what they would say to the waiter after sniffing a wine cork.

It is something to think about the next time you hear some elaborate, multiple-turn economic or market forecast.

So with this in mind, here is our monthly update on the employment report, (also appearing in the Columnist Conversation on RealMoney).

The Employment Report

Each month we take a look at the likely change in payroll employment jobs, given other economic data. This is not really a forecast, since the data are all contemporaneous indicators of current economic conditions. We simply use the already reported data to ask what we expect from the payroll number.

Our model takes Michigan Sentiment (hitting new lows), the ISM index (hanging in there at just under 50), and the four-week moving average of initial jobless claims (weak, but not recessionary). Be careful not to consider today's jobless claims numbers. The non-farm payroll report is based on a survey where respondents are asked to tell about employment during the week including the 12th of the month. Revisions to the report come because many employers do not submit the information on time. Some never report, since for many there is no legal requirement to do so. The BLS tries hard to make it easy, and get a complete response.

Our approach suggests a loss of 82,000 jobs, weaker than consensus guesses of -60K or so, and much weaker than the ADP estimate. It may surprise readers to learn that losses in this range are still consistent with modest GDP growth of 1% to 1.5%, as are the other economic numbers. I don't recommend any big bets, since the 90% confidence interval on the survey is +/- 100,000 jobs! And that is after all of the reports are in and revisions done. The excessive attention to this report creates a dangerous trading situation. You can make your own (risk-free!) guesses at our Payroll Employment Game site. This shows you that even if you know the answer in advance, the reported number can be far away. The "official" answers will eventually converge on the "truth" but there is plenty of error along the way. In addition to this resources, you can check out some other little-known facts in this prior article which has links to other references.

How to Win a Recession-Predicting Contest

For the last few months, and much longer for some, there has been a barrage of commentary about whether or not "we are already in a recession."  The financial media have continuously polled the public to see what they think, and asked everyone who appeared on financial television.  It did not matter whether the respondent had any forecasting credentials.  Apparently we all want to know what anyone thinks about this important question.

Maria Bartiromo, whose interviews other journalists might well study, draws a careful line.  She does not advance her own ideas and then ask the expert to agree.  Instead, she draws out information.  She does show enthusiasm and encourage her interview subjects.  It is amazing how often she asks exactly the question we are thinking of.  Bartiromo warns that the media fixation on recession could become a self-fulfilling prophecy.  On the Today Show, she made this thoughtful comment:

“[T]he truth is, [“Today” co-anchor] Meredith [Vieira], it doesn't matter if we’re in a recession,” Bartiromo said. “We can talk ourselves into a recession, and that seems to be what we’re doing right now and that certainly begets more weakness.”

Meanwhile, a majority of economists now agrees that "we are in a recession."

It Really Does not Matter

There is no light switch that makes things terrible if the economy is in a recession and acceptable if it is not.  Economic data show that we are experiencing a period of economic weakness.  This means lost jobs and lost profits.  It is a permanent loss and painful to many.  This is true whether or not the economic weakness attains status as an "official" recession.  It is a range of results, not a black or white question.

The Contest

There is persistent interest about the recession question.  It is a proven winner for pundit and media articles.  It is as if there was a contest going on.  Let us make it official.  How should we determine the winner?

There are some logical steps involved in winning a contest.  First, you need a definition of a recession.

Warren Buffett, esteemed by all and especially by us, says that this is a recession by any common sense definition.  Maybe so, but common sense is difficult to codify.

Some embrace the rule of thumb:  two quarters of negative GDP growth.  This is the definition at InTrade, where those living outside of the US can trade the prediction markets legally.  Since InTrade has to pay off to winners and losers, they need a specific rule with a timely resolution.  The punters have the odds at 70% or so.   Bespoke Investment Group and Greenback Consulting report these results.  We should note that the public has been wrong on this forecast and those selling short the 2007 InTrade recession contract collected.

The official definition comes from the National Bureau of Economic Research (NBER), the accepted recession-dating authority for decades.  Here are the NBER criteria, also known as the official rules for this contest by which contestants will be judged:

The committee places particular emphasis on two monthly measures of activity across the entire economy: (1) personal income less transfer payments, in real terms and (2) employment. In addition, we refer to two indicators with coverage primarily of manufacturing and goods: (3) industrial production and (4) the volume of sales of the manufacturing and wholesale-retail sectors adjusted for price changes. We also look at monthly estimates of real GDP such as those prepared by Macroeconomic Advisers (see http://www.macroadvisers.com). Although these indicators are the most important measures considered by the NBER in developing its business cycle chronology, there is no fixed rule about which other measures contribute information to the process.

The NBER also notes the following:

A recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales. A recession begins just after the economy reaches a peak of activity and ends as the economy reaches its trough. Between trough and peak, the economy is in an expansion. Expansion is the normal state of the economy; most recessions are brief and they have been rare in recent decades.

Implications for a Winning Entry

Many contest entrants have not studied the rules:

• The effects must be broad based, not emphasizing a specific sector;
• The definition is monthly, not quarterly;
• The official start of the recession will go back to the last peak of economic activity; and
• The result is not known for months -- maybe many months -- after the recession has occurred.

The last point is especially noteworthy.  The study of economics involves the use of accurate data.  The official data get revised as more information becomes available.  This means that no one really knows the key factors for many months.  It is not that economists are stupid  or incompetent.  They are attempting to develop data that are useful in econometric models.  Recession dating is inevitably retrospective.

The dating procedure has important implications for those entering the contest.

1. Those who started forecasting the recession too far in advance should be disqualified.  There has to be some statute of limitations.
2. If the current period gets defined as a recession, those "calling it" at the time of the last peak will be the winners.  The official date will go back to that time -- perhaps October or November.
3. There could be a recession, as defined by the NBER, even if there is no single quarter of negative GDP growth.  Careful readers will observe that all criteria could be met without an actual decline in GDP.

Handicapping the Field

The Perma-Bears. Readers can submit their nominations, but quite a number of forecasters have been on the recession theme for more than a year.  They are disqualified, unless they specified a starting point.

October-November predictors.  These entrants might win, if the NBER conditions are met.  These are the pundits who have said for months that "we are already in a recession."  A better formulation of their comments might be that these conditions, if sustained for a long enough period, will eventually be judged as a recession by the NBER.

The Deniers.  There are several respected pundits who do not believe that current experience will constitute a recession.  These include Rich Karlgaard, Vince Farrell, Gary D. Smith, Dick Green, Mark Perry, and David Malpass.  These observers, all of whom we respect, note that current data are not actually at the levels of past recessions.  We note that the NBER might deem this a recession if the pullback from the peak is great enough, broad enough, and long enough.

The Probability Analysts.  Some of our most thoughtful and respected sources consider the data and make forecasts in a careful fashion.  These forecasts weigh probabilities, mostly the question of whether the NBER criteria will be met.  They ask the question of whether the current economic weakness will eventually prove to be long enough and broad enough to meet the definition.  This group includes two of our favorite sources, including the ECRI (recently calling the recession) and Econbrowser (with an ongoing evaluation).

The Winner?

Winning the recession forecasting contest is a bit like wining one's NCAA March Madness pool.  The most thoughtful handicapping may lose to those taking an eccentric but winning position.

We expect the ultimate winner to be a pundit from two polar opposite groups:  one calling the start in October or November or one predicting that no recession will occur.

Our reason?  If this time period is ultimately deemed to be a recession, the dating will go back to the last peak.  Those are the rules.

Corrections?  We invite anyone whom we have missed to check in with an entry.  Readers, too.  Also, please let us know if we mis-represented anyone's position.

Winners will be announced in a year or so, when the NBER analyzes all of the revised data!

Media Competition: The Erin Burnett Effect

Successful investing -- or trading -- requires a method and discipline.  This often means knowing how to interpret information and reacting to changing prices.

For many months the key issue has been the prospect for a recession.  Many individual investors and hedge fund managers expect the worst and look at estimates of forward earnings with skepticism.   Friday's market decline, which will be extended tomorrow at the opening according to overnight futures trading, is mainly a reflection of this concern.

Two Approaches to Forecasting Recession

The bearish camp points mostly to problems in housing.  These  pundits argue that housing problems will affect employment directly, other businesses indirectly, and consumers as well since they view consumer spending as fueled by borrowing.  These observers predict with certainty.  They profess to know what will happen.

The consensus of economists continues to see recession chances as about 35%.  This is somewhat above the normal odds of 20% -- since an economic shock can create a recession at any time.  In a time when the Fed has intentionally attempted to create below trend economic growth, fighting inflation expectations, the recession odds will be somewhat higher.

Consumers of Recession Forecasts

At "A Dash" we are a consumer of recession forecasting since we do not have  our own model.  With the exception of a few professional economists, our readers are also consumers of forecasts.  Most consumers make a serious mistake.

The bearish argument is easy to understand in plain English.  It has a logical appeal, since all of the pieces are stories that the media highlights on a daily basis.

The economic argument cannot be understood by an observer without economic training.

This disparity is exacerbated by a general tendency among pundits and bloggers to disparage the results from economic forecasting.  We have tried to explain the error of this -- certainty versus edge -- but it is a difficult concept to grasp.

The Media Background

Some months ago, CNBC lost Liz Claman.  She had to wait out a non-compete agreement, but her plan was pretty clear.  Liz Claman has good ties to Warren Buffett.  She earned that relationship through respect ("Mr. Buffett"), charm, and an engaging style.  She wrote a book based upon her work.

When the Fox Business Network was announced there was aggressive competition for interviews during the launch week.  Buffett was a prime target, and CNBC lost, seething at the result, according to one report.

Friday's Trading

The market opened down about 0.65% on Friday, and Caterpillar opened down less than three points before rebounding.  CAT revenues were in line.  Earnings were about 2% light since they (once again) did not push to raise prices and the forecast for 2008 was reduced about ten cents -- something that was already widely expected.  [Full disclosure:  We are long-term holders of CAT and big winners.  If they are growing revenues and earnings at a 10%+ pace in negative times, we see big things and a higher multiple.]

The Erin Burnett Effect

CNBC viewers were treated to an in-person interview with Elaine Garzarelli.  She was on the program because of her prescient forecast of a market crash twenty years ago, before Black Monday.  Over the next few years, she seemed to lose prestige, although her quantitative method is quite strong.  She explained why she was very bullish on the market based upon fundamentals, with sentiment as the only "neutral" factor.

At this point  Erin revealed that she had some bad news.  That was the introduction to the first segment of her interview with the CNBC response to Buffett, Julian Robertson.  The key point of her interview seemed to be that "Julian" was expecting a "doozy of a recession."  CNBC viewers got to see this story in repeated segments all day, with reminders of the crash of 1987.  Burnett also interspersed her own reading of the Caterpillar report, emphasizing weakness in non-residential construction.  Others noted that this was an unusual foray for her, so that probably explains the lack of context -- a company that has been conservative in forecasts and aggressive in calling for Fed action.

Does This Matter?

For the "buy and hold" investor, this is all meaningless noise.  Those who watched the Warren Buffett interviews caught his quip that he hopes to live through a couple of more recessions.  He buys good businesses (as do we) and does not worry about economic fluctuation.

Unfortunately, many investors take the wrong cues from information.

Who Knows What?

The average investor (and many rookie hedge-fund mangers) think that someone with the title of "legendary investor" knows everything.  Do you know Julian Robertson's record at recession forecasting?  We do not, so we are not impressed.

The doom-sayers on the economy and the market have a multi-year record going:  all wrong.  One of our best sources of wisdom, Scott Rothbort, compared Garzarelli and Robertson on TheStreet.com's RealMoney Silver site (the really good stuff for those who pay the most.  Worth it for active traders).

CNBC is pitting the bullish Elaine Garzarelli against Julian Robertson. Garzarelli is a bull. Robertson is a bear. While you have to respect both, Robertson peaked in the early 1990s, missed the entire tech boom and made some poor investments in currencies and airlines. Garzarelli's track record since her prediction of calling the crash has been excellent.

When I made my first "Kudlow & Cramer" appearance several years ago, to my surprise and delight, Garzarelli was the other guest. To borrow a phrase from Oprah, "You go, girl!"

Conclusion

This article is not intended as a criticism of Erin Burnett.  She is doing her job, and doing it well.  The question for investors is how to interpret the media imperative.  CNBC seemed to be on a mission:  Their "legendary investor" interview had to be as important as Claman's Buffett interview. 

This is a trap for investors--and an opportunity.  The bearish story is more exciting, more newsworthy, and more understandable.  There is a chance that the Fox Business Network launch will be a race to the most dramatic interviews.  Discussing the fundamentals of interest rates and forward earnings is boring but profitable.

Avoiding Confirmation Bias

Sometimes an answer is easier to see if one steps away from the immediate problem, instead looking at an analogous situation.  This is a common teaching method, and one that we use frequently at "A Dash."

Background

Getting away from the immediate question helps to avoid what behavioral psychologists call the confirmation bias.  This tendency to see evidence as supporting one's preconceptions is very powerful.  We believe that even the leading market pundits, who are well aware of the phenomenon, fall victim to its power.

There is another important advantage:  Clarifying our objective.  So many prominent market commentaries claim not to make specific forecasts.  Frequently the authors criticize those who make specific and quantifiable predictions, pouncing on their errors.

We find this position quite remarkable.  If a market pundit is not trying to make some prediction, what is the value to readers?  Those claiming to have a "variant view" are making predictions.  The "variant view" idea involves even more complex predictions.  The author must make his own prediction, and then also prove that the market has not already discounted his widely-publicized idea.

The critical reader should ask whether these predictions are specific, quantifiable, and falsifiable.  If not, the argument is not useful for trading.

Bill Rempel's strongly-stated article on this subject deserves a wide readership:

There are some who would have it both ways. These people actively manage money! Perhaps for client accounts, where they buy or sell based on their technical models. Perhaps in mutual funds, where they decide whether to hedge with index puts, and how much hedging to put on. However, they try to have it both ways by saying “it’s not a prediction” or authoring articles about the fallacies of making predictions.

Please read Bill's entire analysis.  As we attempt to guide readers to the best sources on the Internet -- especially informing the explosion of new readers who have not joined us yet -- Bill's criteria should have a prominent place.

A Test of Confirmation Bias

Let us pretend that we live in Chicago -- something that means living and dying with "Da Bears."  [Giant fans and others can substitute their own team.]  We are interested in whether the Bears will win each game, each week.  We have many sources of information about Bears players, injuries, strategies, coaching, and most importantly, whether "good Rex" or "bad Rex" will be at the helm this week.  There are also many predictions from a community of experts.  Since we are fans, we have a lot of information and plenty of opinions.

This information can be our undoing.  We think that we know more and can parse information better than the real experts.  Everything that we see on TV, hear on talk radio, or read in the paper feeds our confirmation bias.  This is reflected not just in our interpretation of what we read, but also in what we choose to read.

An Alternative

Now let us suppose that we wished to predict the weekly result of the  Phoenix Cardinals, a team in which we have little interest.  Our  search for information  would be much more objective, including  a more open mind about expert predictions, sources, and data.

[It would be an interesting experiment for those like Scott Rothbort and Brett Steenbarger who have a ready audience, the appropriate intellectual interest, and the skill to conduct such tests.]

The Market Application

Let us imagine that a group of potential investors, abandoning the business of condo-flipping, decided to look at stocks with a long-term view.  These investors did not have any preconceived notions.  They had no market theory.  Their only question was whether to invest in stocks, and whether this was the right time.

In doing their research, they discovered that there was a community of experts.  These commentators had no allegiance to a particular viewpoint.  Their earnings were strictly based upon their results.  Poor performance, poor revenue.

The group of new investors might discover Mark Hulbert, who monitors the long-term performance of market advisory letters.  He identified the best and worst performers over the last ten years.  These groups of experts, unlike bloggers or pundits, cannot "fake it."  Revenues flow from performance.

The rookie investors discover the following from Hulbert (Barron's subscription required) -- a truly remarkable result:

The bottom line? None of these nine top timers are bearish. The average equity allocation among all nine is 92%. This is higher than where this average stood a year ago, as well as where it was in early May.

This 92% average is good news for the stock market in its own right, of course. But it's particularly bullish relative to the average forecast of the 10 stock-market timing newsletters with the very worst risk-adjusted performances over the last decade. The average recommended equity exposure among these worst performers right now is 0%.

In other words, the worst market timers are quite bearish right now, while the best timers are quite bullish. Rarely are we presented with a contrast this stark.

There are no guarantees. But to bet on a new bear market right now, you have to bet against the timers with the best long-term records and with those whose records have been awful.

Conclusion

There are many issues surrounding the prospects for stocks.  One needs to understand the worries, the probabilities, and how much current prices already reflect these concerns.

The intelligent investor reads a lot of information and has opinions on everything.  It is an easy question:

Do you think you are smarter and better informed than the consensus of all of the investment newsletter writers?

Or might  you be falling victim to the confirmation bias, spending a little too much time at your favorite bearish blog?

The Outcome Bias

Readers of "A Dash" know that we are not attempting to describe or explain every zig and zag of the market.  It is not that we lack opinions; it is just not our purpose.  By way of contrast, each day we write a market commentary for our clients on our private blog.  For our investors, we discuss what is currently happening with special attention to the positions we own.

Background:  Explaining the Daily Market

This daily exercise puts us in the position of the journalist, looking at the blank page and needing to write something intelligent to describe market action.  The journalist has no choice.  Something must be written.  Tomorrow's readers need an definitive explanation.

We have a choice.  We frequently observe that market moves were basically random noise.  Often we disagree with what we know will be the headline story in the next day's papers.  [For today -- Oil prices went up, despite OPEC supply increases.  There was a lot of futures buying driving a low-volume rally.  Some big player(s) wanted more exposure.  There is no real explanation.  Others will cite a "rethinking" of Fed moves.  Yada, yada.]

The Outcome Bias

One of the many useful contributions of the study of cognitive biases is the outcome bias.  When one starts with knowledge of the result, it is easy to find explanations and easier to conclude that one's own decisions would have been perfect.  There are strong psychological studies of this effect.

My non-trader friends often observe, after a volatile market day, how wonderful it must have been for traders.  It shows how little they understand.  The volatile day provides potential for a wide variation in results.  It all depends on how one was positioned going in.  For those with each position there will be successes and failures.

Long Premium.  This means that you own options which gain deltas (your favorable position gets bigger) as the underlying stocks move higher and lose deltas as the stocks move lower.  The trader takes "scalps" by selling short stock on the rally and buying it on the decline.  Even this trader may kick himself for "selling too soon" or not guessing the trading range correctly.  The actual traders with this position will do many different things, some getting the optimum result by selling at the top.  Others might wind up losers if they do not trade at all, expecting a big run rather than a trading range.

Long the Underlying.  Your stock (or index) rallies.  You get a big upward move.  Did you sell anything?  If you did, you can buy back lower.  If not, you have some explaining to do, since experienced traders always sell something into rallies.

Short the Underlying.  Your stock (or index) rallies in your face.  Your losses are mounting.  Do you throw in the towel?  Do you add to positions?  Either could be correct, but today, only one decision was right.  As in the other cases, some of those with this position make each decision.  There are many different stories, including those selling at the top.  Every trade has two sides.

Short the Premium. This is the toughest.  Suppose that a trader decided after last Friday's employment report to sell some index calls, expecting them to expire worthless.  A big rally like today's can cause calls to explode in value, making the position much larger than the  original planned size.  Should the trader bail out?  We know from experience that every trader has a price where he gives up.  If the trader does not have this price, his backer or clearing firm does.  Brett Steenbarger has a great discussion of trader fear, and this is one of the causes.  [searching for Adam Warner's recent great article on this topic.  UPDATE: Link added.]  Our experience with traders is that original position size is often too large, based upon what the trader hopes to gain, rather than the risk of loss.

An Experiment

In tournament bridge circles (a group including many leading traders and investors) we often give a problem "on a napkin."  It is called this because it involves card play, and is written down at dinner on a handy piece of paper.  Sometimes there is an obvious way to play the hand -- clearly best via expert analysis.  The person posing the problem hopes to get confirmation for his (losing) decision or admiration for his brilliancy from his fellow experts.  The problem is that some of those getting the problem may try (consciously or subconsciously) to gain acclaim from dinner companions by finding the winning answer on the particular deal.  That respondent may choose an anti-percentage action, just because of the problem setting.

This would be an interesting trader experiment.  We are not going to summarize the factors leading to today's trading, since the information is readily available.  Instead, let us imagine an experiment.  Perhaps Brett Steenbarger, who works daily with traders, or Scott Rothbort, who uses innovative methods in his classes, will give this experiment some thought and find an implementation.

Take a day like today, and some other big market moves.  Provide whatever information might seem to be relevant -- charts, economic fundamentals, earnings stories, breaking news -- to everyone in the test panel.

Tell them the news --  in advance!

Each participant gets to predict the market outcome knowing the news in advance.  The experiment could include a variety of situational examples with different facts.

The key point is one of our recurring themes -- the difficulty in predicting unlikely events.

We would expect a range of outcomes with very few coming close to maximizing the result.  We suspect that those with the "wrong" positions would do the worst, reacting to fear.  Those with the "right" positions would not come close to optimizing the result.  This expectation is based upon experience.  We have been there.

Individual investors who understand the advantage of staying with the normal odds -- and not trying to predict extreme outcomes -- can gain a significant advantage.  For the individual investor it comes down to understanding the difference in time frames and not being frightened by volatility.

Interpreting Comments

Bond and equity markets both reacted negatively to comments from Countrywide CEO Angelo Mozilo.  In a CNBC interview he spoke about the problems in housing and predicted a recession would result.

Should investors heed this warning?

Background

At "A Dash" we have tried to emphasize two themes that can have a great payoff for longer-term investors.  The themes may also benefit traders, although the timing is trickier.

• Identifying the expertise of the source.  The Countrywide CEO obviously has great information and insight concerning his own business.  He is providing additional information about something of current concern, and that is useful.  He is not an expert at predicting recessions.  There is no evidence that he has experience or added value in considering how the housing situation will affect individual or business spending, countervailing factors, or a quantitative forecast of the effects.  He knows about housing.

Try this comparison.  Suppose a CEO was making an extremely bullish forecast about his own company.  We might view this with some skepticism.  Applying the same attitude, one might conclude that the CEO of a business that was struggling would want to make the problem seem as great as possible.  Compare the Countrywide comment with the Bear Stearns CFO who responded to questions by saying it was the worst credit market he had seen in his 22-year career.  He was trying to explain what happened to some leveraged hedge funds, among other things.  These sources are experts at what happened in their own companies, but might be seen as biased when projecting impacts on the overall economy.

• What has already been "priced in" to the current market?  The sources cited have absolutely no expertise on the stock prices of other companies, normalizing earnings, or other factors that investors should consider.

Where to Look

Investors should look to those who are experts at taking all economic factors, viewing the impacts with quantitative impacts, and making forecasts that are tested by prior results.  In particular, it is important to understand that recession forecasters are predicting something that typically is an unlikely event.  They have few past cases to work from, particularly in the modern era.  Any economist recognizes the impact of the housing problem in GDP forecasts.  They differ in the size of the impact.  At "A Dash" we have tried to highlight sources that have avoided "false positives" in the past, including the ECRI.  Many recession forecasters place no time frame on their predictions, moving the date forward each year as they are proven wrong.  Sidelined investors have missed an opportunity by listening to these predictions.

One should also consider how much earnings forecasts have already been reduced to reflect recession chances.  This is the only method to figure out what is "baked in."

Where Not to Look

We suggest avoiding pundits who cite each statement from a housing source as important fresh information.  Many of these pundits seem to think that they are the only ones with any vision.  They repeatedly claim that others "do not get it."

In fact, everyone "gets it" in that they are aware of some GDP loss.  The difference is that some forecasters attempt to quantify the effect, while others (mostly non-economists) cite each statement as proof without sound analysis on quantifying the effect.

It is a curious fact that pundits who are not economists strongly believe that they have greater insight than those who studied for many years to learn theory, data analysis, and forecasting.  They operate from a stereotype of homeowners that reflects only a small segment of the population.

Summary

The many market pundits who take a bearish view of the economy can find plenty of specific statements and data to support the housing effect.  The difference between the real experts and the pretenders is the ability to quantify those effects.

Having arrived at an economic forecast, the investor must still ask how much the market already reflects a negative perspective.  There is an advantage to buying when fear has been priced into the market.

We rarely have the opportunity to gain a contrarian trading advantage -- buying fear -- by relying upon the real experts.  Now is such a time.

Forecasting Unlikely Events

At "A Dash" we have tried to develop two themes that are important both to traders and to individual investors:

1. Experts add value in forecasting.  The notion that everyone's opinion is equal, while popular on the web, can be costly to one's portfolio.
2. Forecasting should be judged with respect to the difficulty of the task.  Some problems have a high error rate because of their inherent characteristics.  Improving the forecast is still valuable.

Getting a Fresh Perspective

Sometimes the best way to understand a problem is to step away from the specific issue and look at a comparable situation where everyone does not already have an opinion.  When that exercise is finished, we can all consider whether it is a true analogy and what lessons might apply.

If readers will join in this exercise with an open mind, we expect some interesting results and discussion.

A Hypothetical Problem

Here is a problem that few have considered.  Let us suppose that we have assembled a panel of experts on baseball.  Let us also suppose that we have assembled a panel of baseball fans from a particular city.  The team in that city is playing a home game tomorrow.  We ask all of them the question:

Will the home team lose by more than three runs?

The Statistical History

Since baseball is such a wonderful game for statistics, we have some good information about this question.  The chance of a home team loss by more than three runs is about 20%.  (That is also the chance of a home victory by that margin.  The home field advantage is offset by the fact that the home team does not bat in the bottom of the ninth if already winning, so their scoring is reduced by 1/9).

Our panel of experts, if challenged to predict a three-run loss, would all vote in the negative.  They know the data.  They would be wrong 20% of the time and correct 80% of the time.  No panelist would vote in favor of the proposition.  If the panelist votes were publicized in the newspaper, such a prediction would  look foolish.

How Did the Experts Do?

If we did a statistical analysis asking only about the success of spotting three-run losses, we would conclude that the panel was terrible.  They never made such a prediction, missing every case, even though such losses happened 20% of the time.

(Of course they also did not predict a lot of "false positives", but that was not the question).

Posing the Question Differently

Let us next suppose that we asked the panel to give us a probability estimate that the home team would lose by three runs.  The base estimate of our experts would be 20%, but there could now be some variation.  If an average team (my White Sox, sad to say) was playing a strong team like Boston or New York, the chances would be higher.  If the opponents had an ace pitcher on the mound and the ChiSox were using someone recently called up from Charlotte, the odds might get close to 50-50.  Even in extreme cases, however,  the odds of a three-run victory do not get above 50%.

Let us suppose that an expert put the chances of the big loss at 40%, but the home team actually won the game!  Was the expert wrong?  Not necessarily.  We cannot tell from a single game.  The odds might well have been 40%.  It would take many games of similar circumstances for us to judge the accuracy of the prediction.

Briefly put, our experts would never predict a three-run loss in a specific game, although their probability estimates would reflect the specific circumstances.

The Fan Panel

It is easy to imagine that the range of fan predictions would be different from that of the experts.  When fans get down on their team or a specific pitcher, many of them expect the worst.  The result is that some fans might predict a three-run loss on a given day.  They are going by "feel" and momentum, and other factors.

They also make many loose forecasts that defy measurement.  The "team is not hitting" and "fielding is sloppy".  Performance will be uneven -- maybe poor.  The bullpen has been erratic.  The team is hard to predict.

The result is that the Fan Panel predicts more three-run losses than do the experts.  They are better than the experts at predicting the big losses!  (They also have many more "false positives.")

Conclusion

The average statistical expectation in any baseball game is a small margin of victory.  The modal (most common) difference is one run.  Three-run losses come from "random shocks" which sportswriters explain in detail the next day.

Predicting a very unlikely result like a three-run loss, even under the worst circumstances, is a losing proposition.  This means that no expert handicapper would make such a prediction.  As a result, the entire expert panel appears inept at predicting big losses.  The fan panel would always have a few random people who made winning choices, so they would seem better than the experts on this problem.

There are plenty of implications for this important principle, including many which are important for investors.  Some astute readers will have already made key inferences.  We shall elaborate in future articles.

Payroll Employment Forecast

Regular readers know that we have developed a good regression model linking various economic data with the monthly payroll employment report.  Our prediction this month is for a gain of 154,000 jobs, more than the 135,000 consensus and more than the 120,000 needed to maintain the current unemployment rate of 4.5%

There are two main problems with forecasts of the job change:

1. Our model (and those of everyone else) uses the final data series from the BLS.  This is after all businesses finally report in the survey and benchmark revisions have been applied.  The benchmark changes come a year or so later, and can be signficant.  This is all part of the non-sampling error.
2. The sampling error for the report has a 90% confidence interval of almost 100,000 jobs!  That means that even if you knew the truth, the BLS might miss in its estimate.

You can check this out by playing our Payroll Employment Game.  It is cheaper and more fun than betting on the actual report.  The PEG site has a lot of information about the BLS method and surveys, as well as other charts and data.

Our prior work on this subject can be found here.