Darn red-hot griddle just about near fried everything that landed on it Wednesday morning. And while early on the selling was concentrated in

IBM

(

IBM

:NYSE) and

Chevron

(

CHV

:NYSE), by the end of the day -- instead of ignoring those two ne'er-do-wells -- all the other stocks seemed to join them.

(This piece stirred a ton of interest and I was glad to get another chance to have a go at it. The premise of the piece is that very often the Nasdaq market gets over-heated at the opening and then seems to give up the ghost. This pattern always amazes people, yet it makes a tremendous amount of sense if you think about stocks as merchandise and market-makers as retailers of that merchandise, trying to have and find inventory to sell to people. I have been using this griddle analogy because it references the way things get cooked; if the grill is too hot you can fry an egg in a flash. That's fine if you know what you are doing; but if you don't you burn the butter and it tastes awful.)

These bouts of selling have become the predictable flip side to the craziness of the rah-rah opening, where everything is roaring. So let's spend some time working through what happens to a market maker, say, in

Cisco

(

CSCO

:Nasdaq) when the stock is in great demand as it was at the opening.

(I picked Cisco because it is kind of the quintessential tech stock these days. Got good Net news? Cisco goes higher. Got good telco news? Cisco goes higher. Got good personal computer news? Cisco goes higher. It's the stock for all seasons. The term "market maker" references the people who buy and sell Nasdaq stocks at major brokerage houses, the Goldmans and the Merrills. The selling bouts seem invariably to follow these "looking up big" openings telegraphed by the CNBC Bartiromos.)

Cisco is a stock that is "made" by market makers. That is, groups of sell-side traders commit their capital to compete for the business of buying and selling Cisco from large buy-side institutions.

(The term "commit capital" is the operative one in this sentence. Think about it: When someone wants to sell 25,000 shares of Cisco, the "market" buys it. But who is the "market?" It is the trader who "makes" Cisco at one of these firms. He then places it with customers or holds on to it. It is not his job to hold on to it. He is not trying to "play" the market himself. He just wants to facilitate and make the "spread," the price between the bid and the offer, which he splits with the salesman who covers the account.)

When everybody wants to buy it, that's like everyone wanting to take the

Broncos

in the

Super Bowl

. The market makers are like great bookies. They aren't going to sell you the Broncos even up. They are going to tack on some points to entice you to buy the Falcons, or subtract some of the points that you hope to win with the Broncos.

(This is THE crucial point about market-making. These guys don't want to bet with you or against you. Heck, they don't even want to bet. As much as you think they may "know" which way an oblong ball is going to bounce or Cisco is going to land, they don't know or care. They just want to make markets. That's the first thing you learn when you sit on an OTC desk. Don't get caught with the merch. Never. Price it to move, preferably somewhere where nobody gets hurt.)

Same in stocks. Cisco was on a tear Wednesday along with a bunch of other Nasdaq stocks. Buyers overwhelmed sellers. At the end of the day there was very little supply available "in line" for Cisco.

(This was a mistake in my copy that I made. I meant to say "Tuesday." Here I was discussing what happened the day before, because it is crucial to understanding these griddle openings. In the last hour the market roared, and everybody wanted Cisco. So if there was any Cisco kicking around on the desks of these major firms, it got priced and was taken up high -- so high that some firms probably figured that Cisco was due for a rest or a pullback the next morning. Hardly. It kept ramping. That would be as if no matter how often the bookies changed the Super Bowl line to reflect more odds in favor of the Broncos, people mindlessly kept betting on the Broncos.)

Wednesday, when the market shot up, the market makers collectively but independently took the market higher to a level where supply comes out.

(This is the line change to try to bring the market into balance. Remember, these firms have no idle inventory. They are not like retailers who always have a lot of inventory and if you don't see one sweater you like you can grab another. These are more like the pushcart people when it is raining. They appear with umbrellas, which they can charge virtually anything for, but once they are out of umbrellas, all they can do is hope to buy umbrellas from someone else and mark them up to the next guy. Not a good business.)

(Another way to look at it is to think of it like ticket scalping, a pure market if there ever were one. I used to scalp tickets at the Flyers when I was a teenager. I would start with two tickets and try to swap to get other tickets, which I would progressively mark up until about a half-hour before game time, when I would take the prices sky high. Sure enough there would be a frenzy, but at a certain moment, like two minutes after the game started, either was going to have a whole lot of seats for the Rangers-Flyers or I was offering them for below the price paid, or, even worse, below the posted price on the ticket!)

Let me explain how this works. I come in long 75,000 Cisco. I have no desire to sell some Cisco at $106, where it went out the night before. But

National Gift Wrap and Brokerage

caught a buyer who wants to pay "up" for the stock.

(I use National Gift Wrap and Brokerage because whoever I single out here will be none too happy. Nobody likes to get beat, and these market-makers now get beat all of the time. There is a perception among the more vocal of the online revolutionaries that these OTC guys make out like bandits every day of their lives, but that's nonsense. I am sure that a lot of these guys, if they have kids wanting to go into the business, are urging them to go into anything but Nasdaq because the profit margins are so crummy and the stress so great. I use me, because, heck, that's exactly what I was doing.)

(I am carrying around my 75,000 Cisco because I love Cisco and that is my core position. I have no desire to sell Cisco. But I also hate to be greedy. Someone wants to pay up a couple to get some Cisco in, I will let some go, with the idea that I can buy it back lower later, which is exactly what I did. I was playing market-maker myself in a way here, thinking that these prices are too good to be true and will quickly drop, which they did.)

National doesn't want to short the stock if it can avoid it. Maybe he knows that the National Gift client will just keep buying and buying if the price stays down here. So he takes all the offerings or "lifts" them rapidly up to $108 -- there may be only a few thousand shares available between -- and he simultaneously goes out as a "size buyer of Cisco" for his customer.

(This is the process by which a stock moves up or down. This is the "market" in its purest form, raising the price of something until supply comes in. If this were cotton, you would have to wait a whole season to see how much cotton got planted at the new prices. If it were newsprint, there might be a quarterly negotiation between newspapers and paper companies. In this world, it is instant repricing based on demand. It is something that, in the crowd I travel in, I am considered very good at. That does not make me really smart or smarter than you, or Diogenes. It is just something that I can do rapidly and usually correctly.)

(It is totally an acquired skill, in my opinion. When the market maker lifts the offerings, he literally goes out "into the Street," or his opposing sell-side firms, and lifts their offerings to get supply in. It is almost a blitzkreig to get it up there, and it is reflected in extremely fast moving prices on your terminal.)

Guys like me, we know nothing's going on today at Cisco and we think that everything is too hot in Nasdaq.

(This is crucial at the level I work at. When Cisco is up two points I have to figure out whether that's because Morgan went out positive or Pru upgraded or DLJ pushed it, or whether Merrill says they just won a new contract. If it is none of the above, and just genuine market frothiness, I boot some of what I think of as trading stock, with the expressed interest in buying it back at a lower level -- or at a higher level if I am wrong. If I am right more than I am wrong, then I am going to make good money, but I guess that's kind of self-explanatory.)

At the 108 level, National has at last found that seller. We at

Cramer Berkowitz

let go of 25,000.

(At our offices, what would happen is I would turn to my trader, Mark Kantor, and I would say, "Hey, I don't think this is sustainable, sell the buyer 25 right in here. Try to get $108.")

We aren't the only ones. Others hear about and see this dramatic price move and they go to liquidate their positions, too, as they know that it is only market fury and not something intrinsic to Cisco that is doing the moving here.

(In other words, many other people see what we see, and for that split second they want to take profits in Cisco, too!)

At that point, when enough Cramer Berkowitz accounts want to sell, market makers know they are safe to start shorting themselves. The stock has reached a level where supply can be recovered.

(Here are two very key points, sale and recovery. The market makers, like me scalping tix at that Flyers game, we don't know at what point the market is going to turn and I am going to be long the very tickets that were so prized a few dollars or seconds ago. The market makers start seeing offerings, which is street talk for sellers, and they immediately have the same instincts I do. They know that is a good time to sell merchandise, any merchandise, even merchandise they DON'T HAVE, or short stock.)

(If they can sell it at $108, they will probably want to buy it back slightly below there or recover it, because, remember, they don't want to make any bets. They know less about Cisco than you think and they don't want to have to explain to their bosses that they like the brand new Cisco router and they stayed long Cisco down three points because of it. They will get fired for that.)

Soon, on days like Wednesday, when the market ramps breathlessly, there are too many sellers at that level, and the price starts retreating.

(All of a sudden, the market makers are deluged with Cisco offerings, the price has gotten too high too fast and human nature has taken over; guys like me start talking found money, and the next thing you know, there are sellers everywhere.)

At this point, the frenzied buyers of the open are no longer so frenzied. They have hit a wall of supply and it's not budging.

(Some of these takers were just doing it out of emotion. Others were doing it simultaneously with spillover buying from SPX and NDX futures. Still others figured, heck, Cisco could be ramping through $110, so why not take some now and sell it higher. And still others might think there is something going on and want to be long so when the news comes out they can sell it.)

So then they pull their buy orders altogether. They've been picked off by their own frenzied buying. You then get the vacuum down. Sometimes the vacuum takes us right through where the stock closed the day before! At that point buyers get very discouraged, particularly the ones who paid up. They bolt, causing a further meltdown.

(So, there was nothing going on at Cisco, huh? Well I don't need this Cisco then. That's kind of the thinking. Worse, there are some people who had been brought out as sellers at $108 who now think the stock has something wrong with it and want out pronto. Still others realize that the market was too hot and they want to sell before the stock goes negative for the day.)

Next thing you know you have a rout on your hands. Which is why you will never hear me speak positively of the tape when we have one of these ride-'em-cowboy openings. There is too much firepower out there trying to contain the market while inventory is found.

("Contain" is my word. No one firm can really contain a market. But these market makers for that moment do want the market lower because THEY ARE OUT OF POSITION. THEY DON'T HAVE ENOUGH INVENTORY TO DO THEIR JOBS WELL.)

As soon as it is found, it immediately saps the ammo of the buyers who brought us up so high. That's the hot griddle problem; remember it and you won't get burned.

(Now the game has begun, and the tickets are worth much less. Or a player has gotten hurt and everyone wants to lay off the bet or go the other way. Whatever analogy you need, you must understand that these declines will seem justified by the media as based on earnings fears or rates fears or profit fears or Brazil fears. What they really are is the market repricing merchandise rapidly, so rapidly that too many people at once are enticed to ring the register and the market fails.)

(That's why I like orderly openings; they have none of that artificial stimulus that clouds judgments and make you do the wrong things, like buying up two or three or four points. Unless it is the Internet. There, as I have shown over and over again, the rules do not apply.)

Random Musings

Too bad IBM can't split everyday. Then it wouldn't go down seven points as it did Wednesday. Heck, it wouldn't ever go down. Yep, the lesson of IBM is that after the split hoopla subsides, you are left with the same crummy quarter as you had before.

Lately we've heard a ton of information about how stock splits signal management's confidence in the ongoing business. But the reality is that a stock split gets done because a stock is high. IBM's stock was high in part because the Street thought the company would blow out the quarter. When it didn't, sellers materialized. They took a powder during all of the split news, but they were back with a vengeance Wednesday.

James J. Cramer is manager of a hedge fund and co-founder of TheStreet.com. At the time of publication, his 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.