Yes, and according to "statistical probabilities", the 1987 crash was also "impossible", but it did happen. I have not seen the interview, but my guess is he is talking about large funds with broad exposure to the market.
I was watching the same thing. My immediate impression was "idiot". The reason I was watching was because while on vacation, after the family went to sleep. the CNN stuff was the only way I could keep in touch. Then I looked at Fox too. I don't watch TV during the trading day, so now I know who Martha and Maria are. And now I know O'Reilly. And I know more than I care to about Cramer from Kudlow and Cramer. And that Brian news anchor guy. Sheesh, everything is devastating with him. And full of misquotes. I can't help but think that Kudlow wants to end every remark with "that b_itch!" Although, for the week, I found I shared his views. Way off topic, sorry guys.
I wouldn't go as far as to say that. In fact, one of the explanations for the well-known and barely explained equity premium puzzle is the embedded probability of something really bad happening that people are afraid of. So in fact, just by looking at the historical valuation of stocks it appears that investors perceive there being a relatively large chance of a significant drop.
You description of the possible self-fulfilling prophecy effects of large scale momentum investing is correct. However, I don't quite see how it's related to index investing. Index funds are not trying to beat the index by following some strategy (e.g. the momentum strategy you are referring to). They are merely trying to replicate/track the performance of the index. If some stocks in the Nasdaq100 went up in value, I don't see why the QQQ fund would load up more on those, and vice versa. They are just trying to stay as close to the index as possible.
A lot of people confuse mathematics with absolutes per situaiton. I think statistics and probabilities plays off deeper than this. To suggest something is "mathematically impossible" is absurd. Take for instance a multi-state lotto. It is mathematically "highly improbable" that you will win. However, at some point, someone does win -- does this mean that this person is an expert lotto player? Of course not. I have noticed in my short time observing ES movements that every time you take a trade, you are already in a losing position. The spread puts you at a disadvantage right from the start. If you do manage to buy on bid or sell on ask, you sometimes eliminate this spread, but generally if you do get filled on these positions, the market will trade right through you and you will end up with the -.25 loss for awhile (microseconds maybe -- but it is still there). However, it comes down to style and ability to be nimble. A retail investor / trader has a much smaller account size than a 100 billion dollar mutual fund. I've been arguing lately with a friend who is a wizard in mathematics and he says that the long-term ability to make consistent money in a zero-sum game is questionable. How in the hell should I know if that means it is questionable for me -- or questionable "on average." Everyone trades for the same reason -- to take money from other traders.
aphie, i don't see why zero sum has anything to do with this issue. zero sum just means there is a fixed amount of money in the market. if one person gains it, another loses it. what does that have to do with it being possible/impossible to consistently make money long term in the markets? i think these 2 issues are completely irrelevant. i will prove my point right here. take any game where there is never a tie. let's say you and michael jordan play one on one basketball and agree to do tie breakers in the case of a tie. you could say this game is a zero sum game because for jordan to win, you have to lose. or for you to win, he would have to lose. now, i would bet michael jordan is a better player than you and could beat you consistently over a long period of time in this zero sum game. so how is the "long-term ability to make consistent money in a zero-sum game" questionable? in the markets, if someone is a better player/loser than someone else in a zero sum game, why wouldn't they be able to over time take the money from the bad players/losers? p.s. yes, in the markets, we're negative from the start because of commissions and spreads, but that doesn't change my argument.
Damn, all my systems with positive expectancy for the past 20 years of testing sure must be "mathematically impossible". Well, I think he's not mentioning about traders. I think he mentions about Randomly picking a stock to Buy and Hold, it's for "average" investors. There's a difference. Still, does that mean I'm defying God's Law of Mathematics? Do I make 1+1=0? Wooot!!!!
in the interview, Solin did indeed mention daytraders. i don't believe he would tell you what he said applys only to buy and hold people. i remember him specifically mentioning people that manage their own accounts, trade online, and daytrade.
my experience and relationships tell me otherwise. this is all i know, and unless the friends who are most consistent and profitable (even in these lean times) are about to swirl into a 10 year tailspin, then dan solin and o'reilly could be wrong..
You must not forget that he is (according to you) a math wizard. Being a mathematician myself, I can assure you he is right: Since I can ask the question "Can anyone make consistent money in a zero-sum game?" it is obvious that the ability to make money in a zero-sum game is QUESTIONable (by definition). Game theory also shows that there are cases in which one player can win consistently in a zero-sum game due to superior strategy. Therefore we not only know that the aforementioned ability is questionable, but also that certain players in certain zero-sum games possess that ability.