When I first got in the business, I remember hearing

Leon Cooperman

-- a really smart guy who was then head of the research department at Goldman Sachs -- talking about a "healthy correction."

(Now I'm dating myself. This was about 15 years ago when I was a salesperson at Goldman. Cooperman was -- and is -- a dominant figure. These days, when most people think they could slay

Liberty Valance

in a heartbeat, there are a couple of real Shanes out there, guys who know what the heck they are doing. Guys you listen to, by dint of how much money they have made. Guys who have made killings. Cooperman is one of those guys and I am proud to consider him a friend.)

I remember thinking to myself: Who is this double-talking joker? Healthy correction? How about "terrific form of cancer"? How about "cheerful plague"? Why not "needed California earthquake"?

(Like many newcomers, I could only imagine that going up was good and going down was bad. But sometimes going up when something shouldn't go up can also be bad. Markets that divorce themselves from the fundamentals can be dreadful for everybody. If you don't believe me, think about Japan. Looking back, what they wouldn't have done for a healthy correction. Straight up markets breed massive speculation; massive speculation breeds losses. Losses sap everybody. We want a market that doesn't skyrocket unnaturally even when the fundamental backdrop stinks.)

The Dow

was locked at about 1400 at the time. It had just put on 400 points in a straight line and Lee thought it was due for a pullback. It happened. We shook out the margined players and the quick-change artists and we socked in some stocks at good prices. The correction ran its course and the market came back stronger than ever.

(This is a process that makes for a healthy level of skepticism. We have been so spoiled by this market that it is hard to believe that markets can get absolutely hammered, stay down for quarters if not years, and yet still come back. In the time I am talking about the market got too hot -- it was another one of those periods where the IPO griddle was seething -- and speculation was rampant. But business was just okay and companies' stocks were doing much better than the companies themselves.)

This morning I got out of bed, feeling older, and I said, "Hmmm, looks like we are having a healthy correction." I'm sure, as you read this, many of you are having the same reaction to that seemingly oxymoronic phrase as I did when I was a fully fro'd 28-year-old who knew everything at Goldman.Sachs.

(Yes, when I got to New York after law school, I still had hair. What is interesting about this paragraph is that corrections have become so scarce that it is a big deal to even say we are having one, let alone predict one. In other words, to me, it is a statement of fact. Let's use sports. Your team has just lost six in a row. To me you are in a slump. So if your markets just went down for five days and lots of points were lost, call it a correction. To not do so is to deny the obvious and to me seem very false.)

Man, I didn't know jack.

In the midst of a correction I always imagine it feels like some hopelessly retreating Army somewhere. I am sure that some of you feel like you are

Lee's

troops at Gettysburg, and you have gone as far north as you will ever go.

Others perhaps view it as the Battle of the Bulge, where overextended green U.S. troops went a bit too far and almost got crushed by an army that should have been finished a year before.

The chaos of it is frightening. The rally that started "too early" Tuesday gaffed a ton of people and the selloff engulfed the rest. It reinforced the difficulty of retreat.

(These end of the day machinations have gotten completely out of control. We now have this nagging problem of funds coming in at the end of the day to accentuate whatever direction the market is taking. This happens because the

NDX 100

, is really just a couple of big cap stocks all of which are quite susceptible to being raided down. We had three days this week where the close was hammered up or down by this kind of buying and selling. The last half hours of Tuesday and Friday were merciless if you were a bull, and Thursday was nirvana. But take it from me, in the end, this kind of thing is very bad for the market. Stocks always fall much faster from concentrated selling than rise from concentrated buying because everybody is scared to death when stocks get hit. Remember, when a stock is rising, you might have a millions reasons why you might want to sell. But when a stock is falling, it always seems like it is falling because something is wrong. By the time you contact every broker and analyst that matters to be sure nothing is wrong, the market is closed. That's why you get these vacuums down on no volume. Nobody is willing to stand there and buy until they can be sure that everything is okay. And you can't be sure until the next day. So the NDX sellers will continue to create havoc until they change the way they do business. As long as there is this rogue trading going on, it pays to keep cash on the sidelines. )

And yet, it is a healthy correction. We have come so far so fast from our October lows that we have to take some pain here. If the pain is too great, we sell. If we can handle the pain, we hold. If we have cash, we wait until the prices get so ludicrous that we can profit from them. (The war analogy works here, because we all know that the Battle of the Bulge allowed the U.S. forces to crush the Nazis much faster in the end as they got hopelessly extended without fuel or supplies.)

(I think it is a correction in a bull market. I keep cash on the sidelines. Tech is getting harder and harder, but other areas are getting more interesting and the playing field between tech and non-tech is getting much more leveled. )

How far have we gone? Let's take the new whipping boy,

Cisco

(CSCO) - Get Report

. Here's a stock that had a triple disappointment in the super bulls' eyes. It did not announce a split. It did not talk up numbers. And it did not come to Goldman Sachs' tech conference. Well, excuse me,

John Chambers

!

(This was a totally facetious paragraph but buried within it are a ton of worries that I personally feel about this great stock, and I am long it. I don't give a hoot about a split but this market got split-happy. As I write this I am going passed that Tunnel Diner with the griddle as hot as an incinerator and I think, that's what the market felt like the day before Cisco reported, when the stock boards -- and I do check these -- were filled with jokers talking about a split. As there are a tremendous number of players in the market these days, talk of a split got this stock going among some of the more active ones, and when Cisco didn't split people felt that something had to be wrong. That's crazy. But when Cisco didn't show up at the Goldman conference, the boards were flooded with worries. That's just plain insane. Cisco wasn't supposed to be at the conference. It comes to the networking conference. Different conference for heaven's sake!!!)

But that's not why it is going down. It is going down because it was priced to perfection. It was priced for everything to go right and then some. It was priced as a Netizen, not a tech company. The longs had simply decided the price-to-earnings multiple, long the staple of the game, didn't matter any more. If it did, what the heck was Cisco trading at 70 times earnings for?

(This is the real reason why things are so choppy. All week the bulls and the bears were battling it out about whether January was a weaker than expected month or not. Everything traded in synch with whoever spoke last. On Tuesday and Wednesday the consensus was that January was weak. On Thursday people had heard enough from the Goldman conference to say things were strong. But on Friday, the whole psychology shifted and powerful forces decided that January was terrible! Yes, the market can be that schizophrenic. Technicians and true believers will be mystified by this whole discussion, but, BELIEVE IT OR NOT, these stocks trade on how strong earnings might be and if January was weak, these stocks will go down because they were priced as if January was blow-out. Hence the rotation into the quieter types of stocks, a rotation that got derailed Friday because the bond market was bad, and these foods and drugs trade off the bond market's action.)

Now we are in the unwinding phase. I want to hold on to my Cisco. My discipline says a stock, down 20% from its high, with good fundamentals, should be bought aggressively. But my gut says: Hold it. What if that benchmark that I am using, the 117 level where it traded, was hopelesslyinflated? What does a 20% decline from ridiculous levels mean?

(Meaty stuff here. Classically, if you think we are still in the bullish phase, you should wait until stocks comedown 15 to 20% after a big run-up and buy them. But what if stocks were accentuated by a mania. What if Cisco got too ahead of itself because of the .com hilarity? That would mean that maybe a 20% pullback is not enough. Some might just say leave Cisco entirely because it is too expensive. Me, I think Cisco is always expensive except right not it is more expensive so keep some on the sheets but be prepared to buy it lower. How low? Who knows? In the meantime I will have to keep checking to be sure that Cisco is still doing well. Could Cisco ever do badly? Sure, this is always a seasonally tough quarter for them. Do I think Cisco is doing badly? No, or I would not be long it. I don't want to sell this stock because of valuation reasons alone.)

Is this

U.S. Surgical

at 95, down from 130, to use a benchmark some of us remember as a killer place to average down? I don't know. (Oops, talk about three words you aren't allowed to use in this business. They are going to take my guru card away for sure!!)

(U.S. Surgical was the classic loved story, and it doubled and doubled and doubled again -- and then one day it hit a wall as competition came into the marketplace. As it climbed, it got more and more adherents, and they reiterated buy all the way down when the stock cratered. Once it peaked, there was no way to average down. You just had to leave. If you think Cisco has "had it" I don't know why you would be long it. But very few people knew U.S. Surgical had had it when it had!)

But when I don't know I do things small. That's why I use the word "nibble." And if Cisco trades up a tad, maybe I let some go that I bought lower, so I have room to buy it back again. That's my "trading around a core position" concept that I do constantly. It keeps me from getting too long at one level so I eventually have to blow it out because I am out of capital or out of position. In a correction it is vital to keep some dry powder around for when things have gone too far.

(Here is the essence of something I do at my fund. I might have 50,000 shares of Cisco that I keep as a core position, which at the end of the day I want to have on because I have a low basis and because I think the stock is a long term winner. Periodically, the stock will get hammered and I will pick up more Cisco. But then it will reassert itself and I will sell that particular lot of stock that I bought on weakness. Or, as I mentioned when Cisco ran up before the quarter, I might sell some stock short against my common stock, and then bring in that stock after Cisco falls back to normal. Sometimes, when I think Cisco is really undervalued, I might take my core position up to 100,000 or even 200,000. But other times, like now, I will keep a minimum of Cisco on and wait for an opportunity, caused by the mindless end of the day NDX selling to pick up more. This method generates a lot of little gains and keeps me in the game. I love taking advantage of dislocation and selling strength. It is what I do. Trading around a core position has been very profitable for me for the last 15 years, so I keep doing it. )

So, when does the correction run its course? Guru-rejection time again: Who knows? I prefer to see a crescendo, where everybody pukes everything, a la Oct. 8. We don't have that yet. Remember, that was the time when the whole world was bearish, except

Ron Insana

, who came on and boldly presented guests who anticipated the turn.

(We had a big rally Thursday and then a disappointing session Friday. That's all part of this churning and rotation that I am not that fond of. Some groups got overextended-the techs. Others got cheap-the banks. You are seeing continued selling in techs followed by buying. I don't think we are through this process which is why I keep cash on hand. I cannot stress that enough. By the way, I will be Ron's guest on a special CNBC show from 12 to 1 on Monday)

How do we know that this time it is not different? That this time it's all over? That it's unhealthy? For that, we have to rely on the fundamentals. There is little to no inflation. Growth is good. Business is strong. About the only thing that got out of kilter were the valuations.

(There are always problems. Personal computer sales slow and get faster. Spending slows and speeds up. Sales get better and get softer. But as long as the fundamental backdrop-low inflation, good growth-stays intact, the price we pay is that stocks will get a little too robust-and we will sell some of them-- and then come back to earth and let us back in again. We want stocks to be in synch with the fundamentals. Our biggest fear is Japan circa 1989 where the fundamentals peaked but the stocks kept going up. I don't see that happening here, by the multiple to earnings right now is TOO high and I, like virtually everybody else except those who only think stocks should go up forever, sincerely hope that stocks settle back to prices where there is less risk and more reward. )

And that's exactly what is correcting. Healthily.

(If we get valuations returning from nosebleed levels, we will see a lot more money come in to the market. A market that doubles from October is too hot. We are cooling. Remember the griddle! If we were at 1000 degrees, its been cut to 700. Still too hot to fry eggs on. But going in the right direction, at least.)

James J. Cramer is manager of a hedge fund and co-founder of TheStreet.com. At the time of publication, the fund was long Cisco, although positions can change at any time. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. Cramer's writings provide insights into the dynamics of money management and are not a solicitation for transactions. While he cannot provide investment advice or recommendations, he invites you to comment on his column by sending an email to letters@thestreet.com.