In theory, the volatility over time in a stock should be proportional to the immediate spread.


The asymmetry in information and opinions should be similarly scattered.


So when you say "thin" you are only talking in relative terms.


If you have two stocks which you expect to rise 10 points from the offer, and one has twice the spread of the other, is it possible you should always buy the one with the narrower spread? Because you expect it to move more? Or are your ideas simply twice as outrageous?


All I'm trying to say is that even though the margins relative to price stay fixed - so that you have to put up the same money for the big-spread stock and the small-spread stock both trading at 10 - you're actually getting more bang for your buck in the big-spread stock. And spread is relative to bang.


Meaning, you don't compare the 1/4 spread in one stock to the 1/2 spread in another stock, you compare a stock only to itself. If one stock has a 1-cent spread, that should not distract you from buying the offer in a totally different stock you know something about. Not unless you're choosing between competing opportunities. But you shouldn't try to pry an extra half out of one stock just because another stock has a narrower spread, and it isn't "fair."


Moreover, if you move the stock a 1/2 point just to fill your thousand, you only have to have one guy with 100,000 shares "discover" the stock, and you're off to the races. Market impact is huge because not that many people disagree. You should be encouraged by how few people disagree with you, to chew through, if you are right.


And of course I don't got for that it-can't-be-a-dollar-on-the-sidewalk bit.


And I gave quite a bit of second thought to my second "proof" and still stand by it. The thing that bothered me is that you had asymmetric information, to the extent you knew you were a buyer at 21, and the market maker didn't know this or he might have bid for 1,000 at 20 1/2 to sell to you. But, in reality, you're not a buyer at 21 if you're not buying at 21, you think it is a bad price. Or something.


Plus, let's pretend you somehow knew with certainty the stock would go to 40, only you weren't sure exactly when, wouldn't you grab the offer then? My point is that if you are choosing your price based on some combination of the immediate, public bid and ask, and your own unique opinion, that you might as well throw out the symmetric-information part. Plus there are some other ways of looking at it which I forgot.


So far as you making money on limit orders at Brown - and people using market orders apparently "losing" at your expense - my only point was that any "expected" profit was not created by limit orders, but rather by asymmetric friction. A lot of people made fortunes day-trading like that, but most of the ones I know lost most of it - though it's picked up again a little since 9/11.


I think that the assumption a market order is losing and your limit order is winning is usually very suspect, even if the spread hasn't moved. It may be skewed. You may have "learned" a strategy which works, but inevitably takes equally long to unlearn when it stops working.

At most, you are both winning, unless you have happened upon some very fleeting or lucky situation - which plenty of people have, and of course they instantly throw infinite money at it (like the proverbial broken slot machine at Stardust that gives you 2 credits for every quarter you drop in).


Maybe I should have said that, yes, some traders have made money using limit orders. But you should be aware when you are investing and trying to trade at the same time. People should realize that, if even they are making money investing, they are often losing money trading at both ends. There was a great deal of this in 1999, people who could use limit orders and still win, which is why ECN-types mase so much money. The only way to be safe, as an investor, is to use a market order.


When you use a market order, in most systems and microstructures around the world, you will get a fair price given everything that competing market-makers know. When you use a limit order, it's usually just a matter of the guy with the lowest friction winning the battle to screw you.


As an investor, if you can't beat the spread and make money, you probably shouldn't be buying. If the spread swallows your edge, that is just saying "I could sell at 40, that is the right price - if only there were one other guy on Earth who knew that to buy from me." There are no fair or correct prices, only individual buyers and sellers. You only attempt to buy mid-spread if you know of a seller. Anf if you do know of a seller, maybe you should be selling short.


This post has gotten too long, and I haven't even made my point yet. Oh, well, I forgot what it was...


I guess it was that, if Ray knows of some buyers and sellers, he should try to match them off. But if somebody thinks there should be a seller at 20 1/2 to mirror your buy, just because the market-makers at 20 and 21 are mirrored, that is silly. I'm lost...


leroy