As the late-'90s surge in Internet stocks has sent time-honored notions of value investing, long-term horizons and the "efficient market" up in smoke, nobody has looked on with more horror and amusement than Christopher Byron.

In his regular columns for MSNBC, the New York Observer and Playboy, Byron is a master of mordancy. His column headlines stand out for their gleeful -- if occasionally monotonous -- vituperation: "A Tale of Two Dicks: How Bernstein, Snyder Wrecked Golden Books," "Internet Hysteria Brings More Fools to Unstable Market," "Wilbur Ross Bought a Turkey and It's Still a Sick Bird."

But Byron's dissections of what he calls "anything with a dot com in the title" have gone beyond both old-fashioned skepticism and "Internet-is-crazy" knee-jerking. Unlike the many commentators who greet skyrocketing stocks with either uncritical cheering or gassy skepticism, Byron presents a coherent view of what is making the market explode -- and how the influx of new day traders has combined with a raft of thinly traded stocks to inflate the tech stock sector.

It's a view to which more analysts are being converted of late, as the market shivers and regulatory agencies show signs they might clamp down on momentum trading. If they do, Byron will be the first to say he told you so.

Salon recently talked to Byron by phone from his Connecticut home.

For some time, you've argued that day trading and momentum trading are responsible for the recent bouts of Internet stock mania. But isn't day trading, in terms of total trading volume, still pretty small potatoes?

They're not small anymore. Not at all. The day trading phenomenon didn't begin on the World Wide Web, and until a few months ago it wasn't principally concentrated on the Web. I'll give you one example. In the months of August and September, one relatively small day trading firm, called Broadway Trading and Consulting, was responsible for roughly 8 to 10 percent of the total monthly volume in Amazon. That's an unbelievable amount of trading. Another day trading firm -- Schoenfeld Securities -- has a total trading volume equal to 5 to 10 percent of the daily volume of the New York Stock Exchange.

You're talking about what started out as these companies with rooms where people would go in and trade?

Exactly. Up until 18 months ago, these traders were chasing other volatile stocks, because their whole thing is to chase volatility. But there hadn't been the kind of momentum available that you have in Internet stocks. When this opportunity in Internet stocks developed, the day traders started coming into this market, and they have tremendous buying power. And if you add up all the volume of these types of operations, and concentrate them all in one relatively small sector of the market, you get these explosive moves that can double or triple a stock's price in an afternoon. There's really nothing more to it than that.

You've also argued that relatively small floats -- the percentage of shares outstanding that are actually held by the public -- among new stocks contribute to that.

The small float is one of the things that makes these stocks appealing -- because the money here is made in the volatility of the stock, not in the ability of the company's management to manage its affairs. The whole thing is focused on creating a wild imbalance between supply and demand, and from the point of view of the day traders, it really doesn't matter which way the imbalance flows.

So it's not that the small float vs. shares outstanding is contributing to this. What's happening is that investors, or speculators, in this corner of the market actively seek out very small, thinly traded stocks. This was happening on Feb. 8 with a stock called Micros-to-Mainframe, just to give one small example. This stock traded nearly 8.8 million shares and went up 40 percent in one day. That was all run off the chat rooms, because the company had put out a press release saying they had opened a Web site. None of that reflected any calculation whatsoever of the value of that business. That's irrelevant. What mattered was that a bunch of chat rooms piled onto this stock -- and because one went into it, the others followed.

Can you explain a bit about float vs. unregistered stock, and how you think these numbers are out of whack in so many recent initial public offerings (IPOs)? How can a relatively small float make for an inflated company value?

Float is one of the most misleading statistics published in the financial world. People rely on official float numbers at their peril.

When a company goes public, usually it sells a limited number of shares to the public, and a large number of shares are held by individual officers, directors and so on. Typically speaking, those shares are not registered for sale to the public. They become what's known as "lockup shares." Securities and Exchange Commission rules say unregistered shares normally can't be sold for 180 days after the first shares sold to the public go to market. The combination of the total number of shares sold to the public plus shares in the lockup is the total number of shares outstanding. The "float" refers only to the portion of total shares outstanding that are sold to the public.

Now, when you have a new company that has just gone public, figuring out the size of the float is extremely easy, because the float consists of nothing but the shares sold in the IPO. After the thing is publicly traded for a little while, it gets more difficult to figure out what the float is, because some of those unregistered shares might get sold to the public. Or the company might issue new shares out of its own treasury and buy another company, as Yahoo has done and Amazon has done.

In most cases, after a while the float is bigger than it was when the company originally went public. The problem is that most of the data providers don't want to take the trouble to go back and look up every single S-4 registration statement, and find out how many shares are actually sold to the public.

Do you do that?

I do that constantly. So I know my numbers are right. It's a pain in the ass and it's really anal and I don't like doing it, but it's the only way you can get the right number. You've got to go back and sit there like some moron, hour after hour, adding up every damn unregistered stock filing that's been done by the company.

What's the biggest gap you've found between actual and perceived float?

I did it with Amazon, and found that the numbers being put out by Market Guide were nuts. They were crazy. They were saying there were 20 million shares in the float. We did a story [last July] demonstrating that there could not have been more than 7 [million] or 8 million shares at the most. And the reason this is important is that it explains why Amazon got so pricey. I really feel strongly about this, because there was a large amount of misleading commentary about what this price rise meant. And a lot of people were saying that investors had come to the conclusion that the value locked up in the Amazon franchise was almost limitless. But the reality of it was that there were momentum traders chasing this stock that they couldn't find. There wasn't enough stock around to justify the demand, and the price went up. And when I look at what's happened to Amazon in the last three weeks, I feel very good about that story, because it's given up half its value.

Do you take any credit for that?

No, I don't take any credit at all. I didn't make that happen. I just know why it happened.

Now, if I can just stay with this for a second: I'm looking at the Market Guide numbers for Amazon right now, and they have Amazon's float listed at 61.8 million shares as of Feb. 5. There is no way that that's true. That stock split twice [since the July article estimating Amazon at 7 million or 8 million shares], which would take it down to 15 million. A 15 million-share float would mean that the float has doubled [since the July count] just from officers selling their own stock. There's no way that's possible. The officers of the company have filed to sell only about a million and a half shares since we did our story last July. If you add that to 7 or 8 million, that doesn't come anywhere near the size of the float that the public thinks is out there.

So where would a number like that come from?

They're just calculating an arbitrary thing, taking the total number of shares outstanding, and subtracting the number of shares held by officers and directors, and calling everything else the float. The correct way to do it is to go and count how many shares have been sold to the public.

That explains why prices swing so wildly. But since day traders are chasing volatility anyway, who's being hurt?

Stupid people. The cleverest people on the planet Earth work on Wall Street. In a bull market everybody thinks he's just as clever. But they're not. It's really tragic when people think that because the stock market has gone up 12 straight years, this is the new condition of life.

Come on -- nobody is saying, "I'm buying Ebay at $200 as a retirement investment."

I think most of them don't know what game they're playing. I disagree with the idea that these are sophisticated investors. There have been a few cases where day traders not only didn't know anything about the company they were trading, they didn't even know what company they were trading, but reacted when some company with a similar ticker symbol sent out a press release.

Why should that make a difference if all you're doing is trying to make a profit on momentum? You think most traders don't understand the gamble they're taking?

Yeah. Just go look at the chat rooms. Or at the trading rooms in particular. In most cases, these people think that if you buy 1,000 shares of a stock at 20, you'll be able to get out at 22. But because of the way the Nasdaq quotation system works, you're likely to have to sell off 100 shares at a time -- and by the time you get out, that stock could be at 12.

Why would you have to sell in lots of 100?

It all depends on how much of a float there is in that stock, and what the market maker is willing to buy or sell. If he's up there for 100 shares, he might say, "If you want to sell that thing at 20, I'll buy 100 shares at 20; my next buy will be at 19.50. And after that I'll buy another 100 shares from you at 19." If you've got 1,000 shares, the stock could be at 17 before you can sell any of it. That's just how it works; it's a really, really complicated card game that goes on very fast.

And why is an individual day trader at a natural disadvantage?

The thing to remember is that these day trading rooms are trying to re-create the trading process that goes on in a trading room at Bear Stearns or Citicorp or Salomon Smith Barney. The people in such a room have no information beyond what's now available to any single person trading on the Web. But they have enormous experience. Some of them have been in there for 10 years trading eighths of a point on a stock. They understand what happens when you take down 10,000 shares of a stock, and they know how to read the intentions of market makers when they start quoting and changing their quotes on a stock.

The people who are getting into this thing for the first time don't have the slightest idea what this is all about. All they know is that this stock has gone from 2 [points] to 2 1/2 to 2 3/4, and they think the damn thing is going up. For all the logic that goes into their investment strategy, it's just a completely random roll of the dice. And the very next second the whole thing could fall apart on them.

But you don't think that the average investor here has some idea that these traps are out there and can respond accordingly?

Absolutely not. The numbers are self-evident. There's no mystery that people got slaughtered in that case of Micros-to-Mainframe. Now every single person who got hurt in that stock got hurt for a different reason. But the common denominator for all of them was that they didn't know what the hell they were doing. Nobody bought that stock believing that once you got into it you couldn't get out of it, and that it was going to fall in half-point increments from 12 bucks a share down to about eight. And while that was going on, the market makers were only willing to buy in lots of 100 shares each, so if you had 1,000 shares of that stock you couldn't sell it to anybody.

But at the same time, some of the saddest stories on the message boards come not from traders following a rising stock but from people who tried to short these stocks.

These are very difficult stocks to short. If you have a very limited float and you try to short the stock, the underwriter can corner it, and you'll get killed. To take one example, we followed the stock of a company called Shopping.com over a six-month period. As it turned out, most of the investors in that stock were clients of the underwriter of the original IPO. They were able to force a high price on the stock, and the short sellers had to cover that.

How closely do you monitor the activities of market makers taking positions like that?

Well, that's my job. I've got three computers on my desk. One's got a Nasdaq Level II screen, one is watching news all the time, and one is watching a Level 1 real-time quote system that has an easier display, so you can see when stocks start to fall out of their historical patterns. And when that happens, we try to figure out why. Sometimes you can tell whether one or another market maker is dominating the pricing in a stock. If you've got a price that's rising, and you've got one market maker continually selling into it, you know he's got some agenda on that stock.

If a significant player were continually selling, wouldn't that tend to drive the price down?

You don't know. He might be trying to stabilize the price. He might have a short position and think it's all gonna fall apart. All you know is that there's an enormous number of potential market makers for each stock, and if one of them is always there on the inside, there's gotta be a reason for it.

Look, I'm just a journalist here. I don't trade in the market or anything like it, but I try to figure out what's going on so I can write about it. Market makers go through this stuff in their sleep. They understand this stuff infinitely better than a journalist would.

But still, you're one of the only journalists to consistently follow the impact of momentum players.

Most people don't do it. It's more fun to hype some stock all the way to the moon, isn't it? People forget that Newtonian physics always prevails.

Do you get a lot of hate mail?

Yeah, you want to hear some?

Sure.

I'm getting more lately because the market is starting to turn. Nobody wants to be reminded of this stuff when things are going against them. Here's one: "You have a lot of opinions. Why are you trying to scare the hell out of everybody? Just shut up!" Here's another one: "You must be a Democrat. You can't handle the success of others. You're part of the growing conspiracy to stifle the market. It's none of your concern if fortunes are won or lost. It's my money. Go home."

As you mentioned, the market has been pretty nasty lately, and the regulatory agencies are starting to put the brakes on some of this volatility. Whatever happened to caveat emptor?

I don't think the market should be policed to prevent this kind of stuff. But I do think there should be a constant public education process in the media. Instead of becoming a cheering section for bull markets, the media would do a better job if we were constantly trying to take the punch bowl away.