Two Plus Two Newer Archives

Two Plus Two Newer Archives (http://archives1.twoplustwo.com/index.php)
-   Business, Finance, and Investing (http://archives1.twoplustwo.com/forumdisplay.php?f=32)
-   -   Improving On Buffett And Desert Cat (http://archives1.twoplustwo.com/showthread.php?t=548181)

David Sklansky 11-17-2007 04:56 AM

Improving On Buffett And Desert Cat
 
In another thread Desert Cat wrote:

"The real challenges of value investing are

1) Intrinsic Value is always an estimate or even an estimated range and its accuracy depends upon the predictibilty of the investment and your skill and experience.
2) You need to buy at a substantial discount to that IV estimate, your margin of safety, to protect against mistakes.

The biggest problem most attempted practitioners have is a lack of fortitude, patience and ignorance of there own limitations. If you think you can buy at a 10% discount because your IV estimates are always within 5%, you are guilty of hubris and will pay dearly. You need to have the patience to wait, even when awash in cash, for those 50% discounts. And when you find one, you have to have the fortitude to commit a big part of your portfolio to it.

Buffett demonstrates awareness of his own limitations when he refuses to look at a stocks price before he estimates its value, he doesnt want to influence his analysis. That is probably a weakness of mine, I sometimes find a set of assumptions that justify buying at todays price to see if I'm comfortable with them, i.e. XYZ is a buy if I believe its merger is at least 85% likely. Thats backwards, I should make my best estimate of the merger happening and the figure what price makes sense."


Pure thought tells me there is probably a better way. Better that is than merely searching for stocks whose price is half of what you think it is worth. Obviously I might be wrong since I am coming to these conclusions without research. More importantly is the problem that my method, nor Desert Cat's for that matter is of any use unless your expertise is at least sometimes better than the market in general.

Desert Cat didn't make clear in his post whether Buffett requires of himself a stock price fully one half of his own valuation before he buys. He also didn't say whether Buffett, after coming up with his valuation (without knowing the price) adjusts that valuation AFTER seeing the price. Surely he should, even though he would still probably profit without doing that, as long as he mainly stuck to big companies.

I mean as good as Buffett it, I'd bet anything that when his valuations differ from the price by a large amount the eventual results, on average do not come close to his valuations. If they did he would be a quadrillionaire. In other words, no matter how good you are, if your opinion differs markedly from the present price, the true valuation is almost certainly somewhere between those two prices. And probably closer to the market price.

In spite of all that, if you stick to big stocks that have little chance of unknowing shenanigans, and you are good, and you don't pull the trigger unless there is a large discrepancey, you ought to make money. But I think I have a better method for the real experts. (A method closely akin to something I wrote about in GBOI regarding horse racing.)
In theory, this method takes into account not only the stuff Desert Cat talks about, but also aspects of human nature, human stupidity, and human dishonesty, that technical analysis claims to deal with. To pull it off you have to know what Desert Cat knows plus more. But if my idea is right it makes for bigger profits. Not only because your picks should have a greater edge (given equal discrepencies between your opinion and the markets). But also because you don't need as big a discrepancy to buy or sell short.

The idea is to first do as Buffett does. Come up with what you think is the right price for the security without knowing that price. But then come up with a second number. Which is your guess as to what the price actually is. When these two estimates are close you should be very wary of making a play, regardless of the actual price. If you think a stock is worth 50 and you think the market will make it 50, don't buy it a 40 or short it at 60.

If, on the other hand your evaluation of a stock's worth is significantly different from your guess as to what the market price is, you have a play if you are about right about the market price. My gut feeling is that the best situation occurs when the actual price is shaded slightly toward you own valuation. In other words if you think a stock is worth 20 and you think the public will price it at 29, I would feel best about shorting it if it is about 27. Its an indication that some rich, smart people might be agreeing with me.

All this of course is related to my Fundamental Theorem Of Investing. Don't invest unless you can explain why people are taking the other side. In fact using that Theorem you can theoretically beat the market without having expertise on the other end of the spectrum. The experts here hate that technique since it results in trades that ignore their abilities. So forget that for now. Combine them both.

Of course combining them might not be that easy for MBA types. They are used to making good estimates as to what a stock is worth. They are not used to estimating what other people are estimating what it is worth. But if you have that first talent, I would be very surprised if learning that second talent doesn't signifacantly improve your results.

adios 11-17-2007 10:37 AM

Re: Improving On Buffett And Desert Cat
 
Interesting post. If a stock has a lot of analyst coverage (we can and probably should have a debate on what constitutes alot of analyst coverage) I'm going to be very reluctant to get involved FWIW. What I've observed about your recommendations over the years (they've been very good) is that seem to be based on:

1) You disagree more or less with how the market is interpreting the news/information on a particular company.

2) You have an insight into into a particular company that others genearlly don't have.

On criteria 1) I can think of your recommendation on the Martha Stewart company options years ago and the recommendation of TASR.

On 2) SHFL comes to mind.

A couple things about Buffet, he tends to buy companies outright. I realize he makes other investments but I'm fairly certain that the major share of his success is acquiring companies and including them in the Berkshire-Hathway conglomerate. Could be wrong about that.

spider 11-17-2007 10:38 AM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
Don't invest unless you can explain why people are taking the other side.

[/ QUOTE ]

I think you are underestimating the scope of this. The "other side" is a very large number of people with all kinds of ideas, motives, fears, etc. I mean, can you even predict what your wife/GF is going to say or do on any given day? But you are going to predict the collective actions of millions of other investors? Granted, there are a relatively small number of institutional investors who matter much more than anybody else, but predicting their actions systematically is still close to impossible.

Were you able to predict the Nasdaq would peak in 2000 and US residential housing would peak in 2006? Lots of people saw those for bubbles (and lots said they weren't bubbles) but getting the timing right is really, really difficult. Also, it seems like maybe you are assuming the existence of systematic mispricings over time, which I would dispute. Although US residential housing was and is overpriced, I don't think this is something you can expect to guide you over time. Maybe in 10 years it will be fairly priced or underpriced. If you know, please share with us!

BTW, you really should pick up some books about Buffett if you care to keep using him as an example. With all due respect you are not really characterizing his methods all that accurately. And in particular, the structure and size of Berkshire Hathaway gives him certain advantages and disadvantages that make him a poor comparison case for ordinary investors.

DesertCat 11-17-2007 11:31 AM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]

...

Desert Cat didn't make clear in his post whether Buffett requires of himself a stock price fully one half of his own valuation before he buys. He also didn't say whether Buffett, after coming up with his valuation (without knowing the price) adjusts that valuation AFTER seeing the price. Surely he should, even though he would still probably profit without doing that, as long as he mainly stuck to big companies.

I mean as good as Buffett it, I'd bet anything that when his valuations differ from the price by a large amount the eventual results, on average do not come close to his valuations. If they did he would be a quadrillionaire. In other words, no matter how good you are, if your opinion differs markedly from the present price, the true valuation is almost certainly somewhere between those two prices. And probably closer to the market price.

In spite of all that, if you stick to big stocks that have little chance of unknowing shenanigans, and you are good, and you don't pull the trigger unless there is a large discrepancey, you ought to make money. But I think I have a better method for the real experts. (A method closely akin to something I wrote about in GBOI regarding horse racing.)
In theory, this method takes into account not only the stuff Desert Cat talks about, but also aspects of human nature, human stupidity, and human dishonesty, that technical analysis claims to deal with. To pull it off you have to know what Desert Cat knows plus more. But if my idea is right it makes for bigger profits. Not only because your picks should have a greater edge (given equal discrepencies between your opinion and the markets). But also because you don't need as big a discrepancy to buy or sell short.

The idea is to first do as Buffett does. Come up with what you think is the right price for the security without knowing that price. But then come up with a second number. Which is your guess as to what the price actually is. When these two estimates are close you should be very wary of making a play, regardless of the actual price. If you think a stock is worth 50 and you think the market will make it 50, don't buy it a 40 or short it at 60.

[/ QUOTE ]

Buffett never lets the market price influence his estimate of value.To better understand Buffett's philosophy, you must first read this.

This interview with Liz Clamen he explains how he did the PetroChina investment.

[ QUOTE ]

“I sit there in my office and read an annual report, which fortunately, was in English, and it described a very good company. It told about the oil reserves, the refineries and everything else and I sat there and read it and said to myself this company’s worth about $100bn. I don’t look at the price first. I look at the business first and try to figure out what it’s worth because if I look at the price first I’ll get influenced by that

[/ QUOTE ]

He made a killing on PetroChina, but since he sold it, the market went crazy.

[ QUOTE ]

On PetroChina:

“We sold it [all stakes], but we sold it based on price [not politics]. At the time of the annual meeting it was around $100 and since then it has more than doubled so, unfortunately, I sold it a little too soon. It was a decision based 100% on valuation.”

“If it went down a lot I’d buy it back.”

“We made about $3.5bn on a $500m investment but we still sold it too soon. I left a lot of money on the table.”

[/ QUOTE ]

Buffett didn't really make a mistake, but he often uses that term self-deprecatingly about investments like that. There is no way he thinks PetroChina is worth $500B or whatever it's trading for today, so there is no way for him to hold hoping the market gives him a price well over value. If you read his writings back into the 50s, it's a theme that pops up regularly, he sold too soon or bought too early, had he omniscience about market prices we (his investors) would have done much better.

Despite that, Buffett has doubled his net worth on average every three years for fifty years. It was faster when he started, and slower now due to portfolio sizes, but that's still very fast. He left the Graham Partnership in the mid-50s worth $100k, and turned that into what would be worth around $65B (if he hadn't shipped a bunch to the Gate's charity recently) now. So not a quadrillionaire, but he's done better than any other long term investor in history. If he can be the audited, certified best without being able to predict price action or market psychology, how likely is that to be a big leak? Maybe, what you propose is way too hard to offer a significant edge in practice?

It's interesting that you did this "pure thought" experiment, because it's similar to what Buffett would do (I've often thought you and he share many characteristics, besides high IQ's, being intellectual leaders in your chosen fields, maintaining an intellectual curiousity throughout your lives, and I'd bet you have a photographic memory similar to his). You should read more of his writings.

But you should also give him some credit. He's been thinking and rethinking this stuff for over 50 years and he's never espoused anything similar to what you propose. I can almost certainly say, that it's because he concedes his inability to estimate what other people will do with any measure of accuracy, especially a mass of strangers with differing financial objectives and philosophies.

[ QUOTE ]
Its an indication that some rich, smart people might be agreeing with me.

[/ QUOTE ]

To me as an investor it is a mistake to make a decision because some smart people agree with you. Smart people make mistakes too. You can only make correct decisions if your facts, not your heroes, are in agreement with the action.

[ QUOTE ]

All this of course is related to my Fundamental Theorem Of Investing. Don't invest unless you can explain why people are taking the other side. In fact using that Theorem you can theoretically beat the market without having expertise on the other end of the spectrum. The experts here hate that technique since it results in trades that ignore their abilities. So forget that for now. Combine them both.


[/ QUOTE ]

When you decide to make an investment, you should always have considered the risks that others have weighed so heavily to give you this bargain. In the case of my 1/3 book value investment, it is a smaller, less liquid stock, that has no analyst coverage that I'm aware of. And this little illiquid "market" thinks the company will not be able to monetize this large asset without hugely punitive taxes, and is concerned about how long it will take. They are probably also worried about it's being majority owned by a rather colorful billionaire with a "interesting" and litigious past. I see those same risks, but based on the facts it's pretty easy to see that I'm getting a great price for the potential return.

Oh, and the 50% discount is something I should look for because I run a small very liquid portfolio. Buffett has had to adjust his filter because his portfolio has grown so enormous that his investment choices have become much more limited, and the markets he can buy in are much more liquid and hence efficient. My guess is he'd be really happy with a 30% discount right now, and probably is forced to take some 25% discounts. Discount of course isn't the exact right way to think about it, it's really expected rate of return which depends on holding periods, but discount is a nice short cut rule of thumb. I buy lots of stuff at 20% discounts if I think my holding period will be 3 or 4 months.

And lastly, It's DesertCat, not Desert Cat. I don't go around calling you Sk lanksy, do I [img]/images/graemlins/smirk.gif[/img]?

DesertCat 11-17-2007 12:04 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
Interesting post. If a stock has a lot of analyst coverage (we can and probably should have a debate on what constitutes alot of analyst coverage) I'm going to be very reluctant to get involved FWIW.

[/ QUOTE ]

For anyone running less than $100M, this is always the correct strategy, as long as you are good at estimating risk. Better bargains are always found in less traveled markets.

[ QUOTE ]

A couple things about Buffet, he tends to buy companies outright. I realize he makes other investments but I'm fairly certain that the major share of his success is acquiring companies and including them in the Berkshire-Hathway conglomerate. Could be wrong about that.

[/ QUOTE ]

Buffett didn't really start buying lots of companies until his portfolio got enormous. He did buy BRK in the early 60s, Sees Candies in the early 70s, but until the late 90s (his first 40 years) he was predominantly a stock market investor. I think it was his purchase of Gen Re in 2000 that finally shifted the mass of the company to wholly owned subsidiaries.

The problem with having a $30B portfolio as a value investor is ideally you only want to invest in your best ideas, five or so at a time is ideal. That would be $6B each. There are a very limited supply of companies that are liquid enough to allow you to buy $6B in stock in a reasonable time frame without driving the price sky high.

So he tends more and more to just buy companies outright. It's the same skill/decision as purchasing stock in a public company, finding the right price on a good business based on the cash flow that will accrue to him as a shareholder. There just isn't an end game since he'll never sell subsidiaries.

A recent example is his $4B purchase of the Israeli company Iscar. But he'll also buy smaller companies, right now the bar is a minimum of $75M in net operating income per year. He actually has an "ad" in every annual report that lists his criteria, and mentions he'd like to do a $20B acquisition.

One of the most interesting things about Buffett is how much he's changed from just an investor who sits alone in an office reading thousands of annual reports a year (with his PC gathering dust), to a CEO of a company with 73 subsidiaries. The change is, none. When he buys a company he leaves it in the owners hands to run, and just expects a monthly report from them on results. Buffett still spends his time reading annual reports.

Berkshire's 73 subsidiaries have 217,000 employees and nearly $100B in annual revenue. Headquarters headcount is 19 people, almost entirely to produce Berkshire's 9,000 page federal tax return. Buffett's management techniques are counter to everything you would learn in an MBA program or from the GE school of hard driving CEOs.

But so far it's worked very, very well. Not by accident, but by careful design. Buffett may not be able to predict the behavior of crowds in the stock market, but he's an expert on individual human nature. He is really good at buying good companies at good prices where the owners are honest, trustworthy, enjoy what they do, and he is able to motivate them to continue to work hard even after he buys them out. All while he's still plugging away reading annual reports and finding gems like Petrochina.

DesertCat 11-17-2007 12:51 PM

Re: Improving On Buffett And Desert Cat
 
I always recommend people read Buffett's annual reports. This year contained the following ode to Walter Schloss. Walter's track record is around 17-20% annualized for 47 years, or nearly double market returns. He's an even more extreme example than Buffett, no acquisitions, no access to any special information or any care about prices, and holds stocks for long periods until they reach intrinsic value. He's also unique among value investors, as he's not concentrated, he holds as many as a hundred stocks at time. He can manage so many positions because he buys solely based on simple fundamental valuations gleaned from reading financial reports. No need to visit management, build complicated models, or understand what the market thinks.

[ QUOTE ]

Let me end this section by telling you about one of the good guys of Wall Street, my long-time friend Walter Schloss, who last year turned 90. From 1956 to 2002, Walter managed a remarkably successful investment partnership, from which he took not a dime unless his investors made money. My admiration for Walter, it should be noted, is not based on hindsight. A full fifty years ago, Walter was my sole recommendation to a St. Louis family who wanted an honest and able investment manager.

Walter did not go to business school, or for that matter, college. His office contained one file cabinet in 1956; the number mushroomed to four by 2002. Walter worked without a secretary, clerk or bookkeeper, his only associate being his son, Edwin, a graduate of the North Carolina School of the Arts.

Walter and Edwin never came within a mile of inside information. Indeed, they used “outside” information only sparingly, generally selecting securities by certain simple statistical methods Walter learned while working for Ben Graham. When Walter and Edwin were asked in 1989 by Outstanding Investors Digest, “How would you summarize your approach?” Edwin replied, “We try to buy stocks cheap.” So much for Modern Portfolio Theory, technical analysis, macroeconomic thoughts and complex algorithms.

Following a strategy that involved no real risk – defined as permanent loss of capital – Walter produced results over his 47 partnership years that dramatically surpassed those of the S&P 500. It’s particularly noteworthy that he built this record by investing in about 1,000 securities, mostly of a
lackluster type. A few big winners did not account for his success. It’s safe to say that had millions of investment managers made trades by a) drawing stock names from a hat; b) purchasing these stocks in comparable amounts when Walter made a purchase; and then c) selling when Walter sold his pick, the luckiest of them would not have come close to equaling his record. There is simply no possibility that what Walter achieved over 47 years was due to chance.

I first publicly discussed Walter’s remarkable record in 1984. At that time “efficient market theory” (EMT) was the centerpiece of investment instruction at most major business schools. This theory, as then most commonly taught, held that the price of any stock at any moment is not demonstrably mispriced, which means that no investor can be expected to overperform the stock market averages using
only publicly-available information (though some will do so by luck). When I talked about Walter 23 years ago, his record forcefully contradicted this dogma.

And what did members of the academic community do when they were exposed to this new and important evidence? Unfortunately, they reacted in all-too-human fashion: Rather than opening their minds, they closed their eyes. To my knowledge no business school teaching EMT made any attempt to study Walter’s performance and what it meant for the school’s cherished theory. Instead, the faculties of the schools went merrily on their way presenting EMT as having the certainty of scripture. Typically, a finance instructor who had the nerve to question EMT had about as much chance of major promotion as Galileo had of being named Pope.

Tens of thousands of students were therefore sent out into life believing that on every day the price of every stock was “right” (or, more accurately, not demonstrably wrong) and that attempts to evaluate businesses – that is, stocks – were useless. Walter meanwhile went on overperforming, his job made easier by the misguided instructions that had been given to those young minds. After all, if you are in the shipping business, it’s helpful to have all of your potential competitors be taught that the earth is flat.

Maybe it was a good thing for his investors that Walter didn’t go to college.

[/ QUOTE ]

PRE 11-17-2007 12:52 PM

Re: Improving On Buffett And Desert Cat
 
Desert,

I'm currently around extremely smart/successful investors who in their careers have produced above-average returns over long stretches of time. Whenever someone like Buffett or Lynch is mentioned, it seems like one of them will go out of their way to make the point that both (especially Lynch, but I'd like to hear your take more on Buffett since you've follow him closely) know a lot more than they make it out to be. Would you agree or disagree with this point?

DesertCat 11-17-2007 01:13 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
Desert,

I'm currently around extremely smart/successful investors who in their careers have produced above-average returns over long stretches of time. Whenever someone like Buffett or Lynch is mentioned, it seems like one of them will go out of their way to make the point that both (especially Lynch, but I'd like to hear your take more on Buffett since you've follow him closely) know a lot more than they make it out to be. Would you agree or disagree with this point?

[/ QUOTE ]

I would disagree. Lynch wrote a series of books, Buffett has his annual reports, gives interviews, lectures. Both seem to genuinely want to teach others what has worked for them.

Of course, since I admire both, it's easy to wonder whether I'm allowing my biases to influence my opinion here. So it's important to look at their track records rationally (esp. Buffett whom I'm much more familiar) what they have done matches up well with what they say to do.

David Sklansky 11-17-2007 06:53 PM

Re: Improving On Buffett And Desert Cat
 
"Buffett never lets the market price influence his estimate of value"

I'm sure you are wrong. You are misreading his words. He is only saying that he makes an estimate of value without letting the price (which he would prefer to be unknown to him while he is making this estimate) influence him.

But if the price is signifantly different than his estimate he would be an idiot not to revise it. No expert in any field would do otherwise. In other words lets say he thought a stock was worth 100 and the market made it 60. a He has to bet whether or not it will over or below 90 in three years. Disregarding interest rate considerations, I'm sure he would bet under. He might bet over 70, or even 80. But he would have to believe that his estimate is probably too high once he saw the price. You basically admitted this yourself. Otherwise why give yourself this large margin of error?

Jimbo 11-17-2007 06:55 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
"Buffett never lets the market price influence his estimate of value"

I'm sure you are wrong. You are misreading his words. He is only saying that he makes an estimate of value without letting the price (which he would prefer to be unknown to him while he is making this estimate) influence him.

But if the price is signifantly different than his estimate he would be an idiot not to revise it. No expert in any field would do otherwise. In other words lets say he thought a stock was worth 100 and the market made it 60. a He has to bet whether or not it will over or below 90 in three years. Disregarding interest rate considerations, I'm sure he would bet under. He might bet over 70, or even 80. But he would have to believe that his estimate is probably too high once he saw the price. You basically admitted this yourself. Otherwise why give yourself this large margin of error?

[/ QUOTE ]

Errr, what does the current market price of a given stock have to do with an estimation of it's "true" value?

Jimbo

David Sklansky 11-17-2007 08:05 PM

Re: Improving On Buffett And Desert Cat
 
Plenty. I am 99% sure.

Billy Walters and his team make their line befor they see the real line. They are supposedly the best in the world. How do you think I would do if I could bet into their line, laying 11-10 armed ONLY with the information as to what the actual line is.

I'm not talking about hedging. Just betting one side. Furthermore I won't force them to put up a line on every game. Just the two or three every week where they are most sure of their opinion. If they let me do that (my strategy would be to bet the public side when they disagreed by four points or more, getting HIS LINE don't forget) I would be the one owning golf courses pretty quickly.

Jimbo 11-17-2007 08:52 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
Plenty. I am 99% sure.

Billy Walters and his team make their line befor they see the real line. They are supposedly the best in the world. How do you think I would do if I could bet into their line, laying 11-10 armed ONLY with the information as to what the actual line is.

I'm not talking about hedging. Just betting one side. Furthermore I won't force them to put up a line on every game. Just the two or three every week where they are most sure of their opinion. If they let me do that (my strategy would be to bet the public side when they disagreed by four points or more, getting HIS LINE don't forget) I would be the one owning golf courses pretty quickly.

[/ QUOTE ]

If only determining the value of a corporation were as easy as this.

Jimbo

DesertCat 11-17-2007 09:20 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
"Buffett never lets the market price influence his estimate of value"

I'm sure you are wrong. You are misreading his words. He is only saying that he makes an estimate of value without letting the price (which he would prefer to be unknown to him while he is making this estimate) influence him.

But if the price is signifantly different than his estimate he would be an idiot not to revise it. No expert in any field would do otherwise. In other words lets say he thought a stock was worth 100 and the market made it 60. a He has to bet whether or not it will over or below 90 in three years. Disregarding interest rate considerations, I'm sure he would bet under. He might bet over 70, or even 80. But he would have to believe that his estimate is probably too high once he saw the price. You basically admitted this yourself. Otherwise why give yourself this large margin of error?

[/ QUOTE ]

I guess you didn't read the "Mr. Market Parable" I linked to. Read it and decide whether it is I or you who are misreading his words.

And a large margin of "safety" is because IV is an estimate. But it's an estimate that doesn't become more accurate by taking a vote from a disparate group of people with different investment philosophies, goals, and abilities, i.e. "the market". In the short run the market is a voting machine, in the long run it's a weighing machine (Ben Graham).

When Buffett was done reading PetroChina's annual report for the first time in 2003, and before he knew their stock price, he said anyone could who read that report could see that PetroChina was worth more than $100B. When he found out it's market cap was $37B, he didn't revise his estimate, he bought with both hands as fast as he could. In fact, he didn't sell until PTR reached a $200B market cap, demonstrating what he feels PetroChina is really worth.

David Sklansky 11-17-2007 11:06 PM

Re: Improving On Buffett And Desert Cat
 
Note: I added a second post immediately after this that explains my stance in greater detail. Please read both posts before replying

"When Buffett was done reading PetroChina's annual report for the first time in 2003, and before he knew their stock price, he said anyone could who read that report could see that PetroChina was worth more than $100B. When he found out it's market cap was $37B, he didn't revise his estimate, he bought with both hands as fast as he could."

He did revise his estimate. Or perhaps this was an very rare exception and he didn't. But the fact that "he bought with both hands" says nothing about whether he revised his estimate. And I would be extremely shocked if Buffet wouldn't fully agree with my stance. The logic is almost irrefutable. It works like this:

Mr Market may be a manic depressive and you are not. But your estimates are admittedly far from perfect. And sometimes Mr. Market is feeling both sane and more expert about a stock (for legal or illegal reasons) than you are.
When that happens you are taking the worst of it when you invest. One way to insure an overall profit is to restrict yourself to trades where it APPEARS your edge is huge. Your edge won't be as huge as it appears because you will occasionally run into a sane, smarter (or more dishonest) than you Mr. Market. But that won't happen often enough to make up for the times you are up against crazy Mr. Market.

So the above method will work, though not to the degree it would appear. But it is leaving money on the table. Because it forces you to eschew what appears to be smaller edges, since you are not sure you are fighting smart Mr. Market. To get some of that money on the table requires an ability to know which Mr. Market has shown up today. If you are reasonably accurate you can start betting more of what appears to be 20% edges and sometimes pass on what looks like bigger edges.

David Sklansky 11-17-2007 11:40 PM

Re: Improving On Buffett And Desert Cat
 
"And a large margin of "safety" is because IV is an estimate. But it's an estimate that doesn't become more accurate by taking a vote from a disparate group of people with different investment philosophies, goals, and abilities, i.e. "the market"."

Here I can say with 100% certainty that you are wrong. But before going any further I want to say to you what I say to my private poker students who get upset when they realize that were wrong about a long cherished belief. "Don't be upset. Be happy. Because you are now going to get richer."

Forgetting interest rates and the inherent upward bias of stocks a randomly picked stock will show a break even EV if you buy or short it. Known fact.

If there exists a perfect stock evaluator then buying or selling a stock at the price HE quotes will show break even EV.

If the perfect stock evaluator thinks the right price is ten dollars less than the market price, then shorting the stock will give you an EV of ten dollars. And you will average 17 cents profit if there is a 17 cent discrepancy.

If the expert stock evaluator is less than perfect, buying or selling at HIS price will STILL show a break even EV but only if you buy or sell RANDOMLY. But if you buy or sell into HIS prices with INFORMATION you will beat him. As long as that information isn't totally random. One example of non random information is Jimbo's opinion of a stock. If you invest at Desert Cat's price opinions randomly, except when Jimbo disagree's, you MUST show an eventual profit.

YOU MUST. DO YOU UNDERSTAND? Because you would break even even if Jimbo merely threw darts.

If instead you used the Market's opinion rathe than Jimbo's you would also show a profit. Since the market's opinion of a stock is also obviously better than random.

If the last paragraph is true, which it is, that means that the "true" expected value of a stock must be somewhere between the non perfect expert's opinion and the market price. In other words if you are an expert who has estimated a price before seeing the market price, you MUST, on average, revise your estimate TOWARD the market price after you see it. Period.

Jimbo 11-17-2007 11:54 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
If instead you used the Market's opinion rathe than Jimbo's you would also show a profit. Since the market's opinion of a stock is also obviously better than random.


[/ QUOTE ]

I found this portion of your post very interesting. How do you reconcile the fact that one second in time the market says a stock is worth say, $109.10 per share and one second later the market says the value of the stock is $109.45? What changed in the underlying value of that stock from one second to the next?

Keep in mind there may be 100 million shares outstanding at that moment in time. Did the company's checking account balance just increase by 35 million dollars?

Jimbo

David Sklansky 11-18-2007 12:04 AM

Re: Improving On Buffett And Desert Cat
 
The non randomness may not apply to pennies. In fact I speculated in a different thread that this may be part of DE Shaws strategies. But I am not sure you are getting the concept. The market's opinion is obviously closer to correct than a random opinion. Any non random opinion shows a profit against Desert cat's line if there is no juice. I used Jimbo's in my example but I could have just as easily used Aunt Hattie's. I probably should have.

DesertCat 11-18-2007 01:39 AM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
"And a large margin of "safety" is because IV is an estimate. But it's an estimate that doesn't become more accurate by taking a vote from a disparate group of people with different investment philosophies, goals, and abilities, i.e. "the market"."

Here I can say with 100% certainty that you are wrong. But before going any further I want to say to you what I say to my private poker students who get upset when they realize that were wrong about a long cherished belief. "Don't be upset. Be happy. Because you are now going to get richer."

[/ QUOTE ]

You know David, I'm not offended by this. But I probably should be. I'm a professional investor with annualized returns north of 30% per year. You are an amateur investor who has a fraction of my investing experience. Don't get me wrong. I am always open to learning more. I study my craft intensely and don't feel I know everything. But it's a little disconcerting to hear that you are going to make me better at something I'm already great at, when you don't even have a track record in this area.

[ QUOTE ]

Forgetting interest rates and the inherent upward bias of stocks a randomly picked stock will show a break even EV if you buy or short it. Known fact.

[/ QUOTE ]

A specific randomly selected stock will show an EV based on it's IV vs. the price you paid for it. I think you mean that a group of randomly selected stocks will likely approximate the markets average return, with variance that depends on the size of the group in relation to the size of the market group. That's because the mispricing that occasionally occurs is "evened out" in big groups. But for individual stocks it can be egregious.

[ QUOTE ]

If there exists a perfect stock evaluator then buying or selling a stock at the price HE quotes will show break even EV.

If the perfect stock evaluator thinks the right price is ten dollars less than the market price, then shorting the stock will give you an EV of ten dollars. And you will average 17 cents profit if there is a 17 cent discrepancy.

If the expert stock evaluator is less than perfect, buying or selling at HIS price will STILL show a break even EV but only if you buy or sell RANDOMLY. But if you buy or sell into HIS prices with INFORMATION you will beat him. As long as that information isn't totally random. One example of non random information is Jimbo's opinion of a stock. If you invest at Desert Cat's price opinions randomly, except when Jimbo disagree's, you MUST show an eventual profit.

YOU MUST. DO YOU UNDERSTAND? Because you would break even even if Jimbo merely threw darts.

[/ QUOTE ]

Are you trying to say the market price is the sum of all skilled participants opinions? Because if you are, you are only correct most of the time. Because the market itself, is only efficient most of the time. And I'm not a market participant when the market is efficient. I think you might believe in the efficient market theory a little too strongly.

Do you realize that many market participants are unaware of earnings of a company they buy, that most participants never do any sort of valuation estimates at all? That many practise TA concepts with no FA at all? That a large participant with an urgent need for cash can quickly drive a stock's valuation down to unreasonable levels? I.e. a single participant may have a long lasting influence on price far greater than thousands of other participants, even if they are more rational and better evaluators than he.

[ QUOTE ]

If instead you used the Market's opinion rathe than Jimbo's you would also show a profit. Since the market's opinion of a stock is also obviously better than random.

If the last paragraph is true, which it is, that means that the "true" expected value of a stock must be somewhere between the non perfect expert's opinion and the market price. In other words if you are an expert who has estimated a price before seeing the market price, you MUST, on average, revise your estimate TOWARD the market price after you see it. Period.

[/ QUOTE ]

The value of the stock is the value of the stock. It's not affected by a participants opinion, my opinion, or the stock price. If you come across an obviously mispriced stock, you should adjust your opinion based on others who are clearly wrong?

I'll give you credit in one way. The reason to have a margin of safety is to overcome uncertainty. Part of that uncertainty is that you misunderstand, or mis-estimate company risks. By having a large margin of safety, if it turns out that your risk estimates were too low, and the markets were too high, and the real valuation is in between, you can still make a profitable investment. But that doesn't imply that the real value is always between your IV estimate and the market's. My experience is that is seldom true.

So I don't adjust my estimates for market opinions. I adjust them for facts. I use my interpretation and my judgment to make the best estimates, and I stick with them until new facts require me to change.

Real World Example: Preview Systems is a microcap with over $3.50 per share in net cash, easily ascertainable from their financial filings. They have sold their business and have few remaining liabilities and employees. They announce that they will liquidate and distribute $3.25 to shareholders immediately upon completion of a shareholder vote at a meeting to be held in one month. Afterwards they expect to pay out another 25-50 cents over the next year as they complete the liquidation process. I know that management in these cases is highly incented to be conservative in their estimates for some obvious reasons. So I estimate PRVW's value to be around $3.50 per share, given the likelihood of a $3.75+ payout, discounted a little for time. For the week after the release of this information the stock price trades at $2.93.

Should I adjust my estimate because of this low price? No, think about who is setting the price. The existing shareholders don't know anything about liquidations. They bought PRVW thinking it was going to be the leader in internet ecommerce and be a $400 stock someday. Now they are just going to get a tiny fraction of that in cash. Some want out of their "dead" internet company so they can trade into a live one so they can make "high returns"? Some are frightened this means bankruptcy, or hate mgmt for failing and don't trust a word they say and just want out before they lose it all. And a 20-30% return seems small to those who were waiting for a big payday, they don't realize that over one month makes it equivalent to an annualized 1600% return. No one from Wall Street is covering them anymore, it's a sub $50M market cap. So why should I adjust a clearly rational IV estimate for a bunch of irrational actors who are selling for differing reasons?

I'll give you the river card on this one. I stuck to my estimate, bought with both fists, and the eventual payout was $3.90 total, so I made about a 30% return for one months use of my money.

DesertCat 11-18-2007 02:00 AM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]

"When Buffett was done reading PetroChina's annual report for the first time in 2003, and before he knew their stock price, he said anyone could who read that report could see that PetroChina was worth more than $100B. When he found out it's market cap was $37B, he didn't revise his estimate, he bought with both hands as fast as he could."

He did revise his estimate. Or perhaps this was an very rare exception and he didn't. But the fact that "he bought with both hands" says nothing about whether he revised his estimate. And I would be extremely shocked if Buffet wouldn't fully agree with my stance.

[/ QUOTE ]

David, Buffett disagrees with you. It doesn't make you wrong, but clearly he doesn't think the same way you do. I've read almost everything he's written or said, and he's very clear about his philosophy. He doesn't incorporate market prices into his valuation estimates. He is looking for situations where the market is wrong, so why would he listen to it's opinion? He's actually said once he's established his positions he would not care if the market was open only once a year, or even every five years. He buys many private businesses, his approach has no need for public stock quotes.

In the PetroChina case he described clearly what he does. Never did he mention using the price to validate his estimate. For you to be right, you must assume that Buffett lies about his methods. The problems with that is

a) Buffett shares his methods freely, even though he isn't required to. If he were lying about his methods (to throw off the competition?) he could just tell people he uses astrology.
b) Even if Buffett was lying about his techniques, they've been documented by his teacher, Ben Graham, in several books that are taught in CFA courses as FA bibles much as TOP is a poker bible. They are responsible for dozens of other top value investors of outstanding accomplishments who credit those accomplishments to following the same rules and wisdom in interview after interview. Walter Schloss is a perfect example.
c) Buffett a life long reputation for directness and honesty.
d) If he used the market price to validate his IV estimate in the PetroChina example, wouldn't he be forced to dramatically lower it, say to $60B to better match the price? So why did he wait until $200B to sell his shares? He always sells before stocks reach IV. It's hard to imagine he nearly quadrupled his IV estimate when earnings only doubled during that span.

So while it's always possible someone is lying, all evidence suggests it's extremely unlikely for Buffett to be lying about using market prices to validate his opinions. The burden is on you to show some evidence he ever has.

[ QUOTE ]
The logic is almost irrefutable. It works like this:

Mr Market may be a manic depressive and you are not. But your estimates are admittedly far from perfect. And sometimes Mr. Market is feeling both sane and more expert about a stock (for legal or illegal reasons) than you are.
When that happens you are taking the worst of it when you invest. One way to insure an overall profit is to restrict yourself to trades where it APPEARS your edge is huge. Your edge won't be as huge as it appears because you will occasionally run into a sane, smarter (or more dishonest) than you Mr. Market. But that won't happen often enough to make up for the times you are up against crazy Mr. Market.

So the above method will work, though not to the degree it would appear. But it is leaving money on the table. Because it forces you to eschew what appears to be smaller edges, since you are not sure you are fighting smart Mr. Market. To get some of that money on the table requires an ability to know which Mr. Market has shown up today. If you are reasonably accurate you can start betting more of what appears to be 20% edges and sometimes pass on what looks like bigger edges.

[/ QUOTE ]

I think your error is that you think there is a huge amount of hidden information incorporated into most stock prices. Buffett and I would say not. Many business can be valued simply through audited financial statements. And that value usually doesn't change much depending upon the news from one quarter or who is running the company.

The market is generally not inefficient because of fraud or insider trading. It's because there are skills to valuing companies (FA) & making investing decisions (patience to wait for big margins of safety, rationality to focus on businesses where IV can be accurately estimated, resistance to Mr. Market and other negative influences,etc.) and there aren't that many skilled practitioners. And the smaller the market cap, the fewer skilled participants exist.

And missing small edges is probably +EV. If you can eventually expect to find big edges, pursuing smaller edges ties up time and capital on lower return investments just when it might be needed for the next PetroChina.

David Sklansky 11-18-2007 02:32 AM

Re: Improving On Buffett And Desert Cat
 
One last try. The Bellagio makes a line on football games. They are the best linemakers in the world. Their line is almost always perfect. So even a professional bettor who beats his bookie 56% of the time only picks 50.1% when he faces the Bellagio.

Joe Schmo is barely break even in his opinions. But he is better than the dart board thrower. (This is an important point.)

The Bellagio says the Yankees will win 70% of the time today. Joe Schmo thinks they will win 60%. In other words the Bellagio says that anyone holding a $100 even money bet has an equity worth $40. Joe says it is worth $20. Its TRUE worth is a tad below $40. Maybe $39.50. Change Joe Schmo to Mr Market. (And I would still love to bet that Buffett both agrees with me and doesn't think his words dispute my words)

Meanwhile your comment about being a 30% a year winner and me being an amateur is not only flawed, but ironically in a similar way that your analysis is flawed. Because I don't claim to be able to top your 30%. I only claim that I could top it if I had a DesertCat clone working for me.

Allinlife 11-18-2007 03:03 AM

Re: Improving On Buffett And Desert Cat
 
i only read parts of this thread but I got the feeling that I can learn a lot reading this debate.

thanks mr.sklansky/ desert cat

crunchi 11-18-2007 03:15 AM

Re: Improving On Buffett And Desert Cat
 
Am i the only one that doesn't understand whats going on in this thread?

madnak 11-18-2007 08:54 AM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
One last try. The Bellagio makes a line on football games. They are the best linemakers in the world. Their line is almost always perfect. So even a professional bettor who beats his bookie 56% of the time only picks 50.1% when he faces the Bellagio.

Joe Schmo is barely break even in his opinions. But he is better than the dart board thrower. (This is an important point.)

The Bellagio says the Yankees will win 70% of the time today. Joe Schmo thinks they will win 60%. In other words the Bellagio says that anyone holding a $100 even money bet has an equity worth $40. Joe says it is worth $20. Its TRUE worth is a tad below $40. Maybe $39.50. Change Joe Schmo to Mr Market. (And I would still love to bet that Buffett both agrees with me and doesn't think his words dispute my words)

[/ QUOTE ]

The information Joe Schmoe uses to form his opinion is presumably available to the Bellagio. Therefore, the Bellagio and Joe use the same information to reach their conclusions. The Bellagio tends to reach much better conclusions based on the information available. Thus, there is no reason to pay attention to Joe. Any difference between Joe's estimate and that of the Bellagio is likely the result of Joe's error.

Where Joe is correct, the Bellagio will also be correct. Joe's opinion is worthless - unless you're betting against Joe, in which case a large discrepancy between Joe and the Bellagio will indicate that you should put it all in there!

I don't know much about investing, or when to buy and sell, but I think your example here is flawed, David. It's no more likely that the actual value is $39.50 than that it's $40.50.

Very nice thread, makes me want to learn about financial analysis.

madnak 11-18-2007 09:08 AM

Re: Improving On Buffett And Desert Cat
 
Another example that I'd like to hear your thoughts on, David.

There are 31 students taking an exam. One of the students is a genius, the rest are of average intelligence. The genius gets an answer of 70 on a certain problem. The mean answer of the other students is 60.

Based only on this information, you must choose an answer. What do you choose?

PairTheBoard 11-18-2007 10:30 AM

Re: Improving On Buffett And Desert Cat
 
I don't see that Sklansky's idea, even if it has merit, really brings much to the table. Suppose DC's strategy is to do his DC evaluation of Intrinsic Value, DC(IV), without looking at price. Then buy if the price is 50% of DC(IV) and hold till the price reaches DC(IV). Suppose DessertCat has determined that following this strategy maximizes his total returns.

Now Sklansky claims his DC-clone can improve on DC's strategy. How? The DC-clone will evaluate intrinsic value via DC's evaluation plus market price considerations, DC+MP. Now, will a DC-clone strategy of buying when the market price is 50% of DC+MP(IV) and holding till it reaches DC+MP(IV) improve on the original DC strategy? No. DessertCat has already determined that his strategy optimizes his total returns. The DC-clone strategy will be missing out on some of the opportunities that the DC strategy benefits from. The DC-clone strategy is essentially the DC strategy altered to something like, buy when market price is 45% of DC(IV). But DessertCat has already determined that the 50% figure optimizes for his DC evaluation of IV.

An interesting wrinkle is how the DC-clone strategy works on the sell side. If the DC-clone is holding until the market price hits the DC+MP(IV) he will be looking at a moving target for his sell price. As the market price increases and approaches the DC(IV) the Sklansky adjustment to IV via market price becomes smaller and the DC-clone's DC+MP(IV) approaches the DC(IV).

I think there is some merit to David's idea here. Certainly if DessertCat evaluated a stock and determined it's IV was $100/sh then looked and saw the stock was trading at $1/sh I imagine he would dig a little deeper to see if there was something he was missing. But generally I don't think DC has to look too far to make a decent guess for why the market is mispricing a stock in relation to the DC(IV). You see that while he disagrees with Sklansky he did have a good idea why investors were mispricing the stock going into liquidation with DC(IV)=$3.50 and market price of $2.95. And while there may be special circumstances as in that case, just a climate of overdone fear or greed covers a lot of ground in general.

PairTheBoard

DesertCat 11-18-2007 11:33 AM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
One last try. The Bellagio makes a line on football games. They are the best linemakers in the world. Their line is almost always perfect. So even a professional bettor who beats his bookie 56% of the time only picks 50.1% when he faces the Bellagio.

Joe Schmo is barely break even in his opinions. But he is better than the dart board thrower. (This is an important point.)

The Bellagio says the Yankees will win 70% of the time today. Joe Schmo thinks they will win 60%. In other words the Bellagio says that anyone holding a $100 even money bet has an equity worth $40. Joe says it is worth $20. Its TRUE worth is a tad below $40. Maybe $39.50. Change Joe Schmo to Mr Market. (And I would still love to bet that Buffett both agrees with me and doesn't think his words dispute my words)

[/ QUOTE ]

I may have a great counter example, and I think the problem is that your sports book is a poor model for the market. Give a day to digest it and see if I can come up with a succient post.

[ QUOTE ]
Meanwhile your comment about being a 30% a year winner and me being an amateur is not only flawed, but ironically in a similar way that your analysis is flawed. Because I don't claim to be able to top your 30%. I only claim that I could top it if I had a DesertCat clone working for me.

[/ QUOTE ]

I should not have said that. Your ability to synthesis ideas has nothing to do with your investing track record. That would be like a mathematician telling Andy Beal he could not contribute to mathematics because he had no track record;)

ahnuld 11-18-2007 12:12 PM

Re: Improving On Buffett And Desert Cat
 
The bellagio example is wrong because it is set by experts who are the best in the world. Thats like if Buffett got to pick the trading price of every company every day and then trying to beat this new supermarket. Obviously undoable after vig. But its not bellagio experts setting the price in the market, its the general poorly skilled public. even the bookies opinion from your example is going to be more highly skilled than the markets because he has some talent in this regard or he wouldnt be in the business. He is similar to mutual fund managers. Yet the market is worse because while there are semi competent mutual fund managers, there are also alot of irrational joe schmo buyers who drive prices or lines way past what is neutral EV setting up +EV bets.


edit: blah, I misunderstood, nm.

West 11-18-2007 12:31 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
The bellagio example is wrong because it is set by experts who are the best in the world. Thats like if Buffett got to pick the trading price of every company every day and then trying to beat this new supermarket. Obviously undoable after vig. But its not bellagio experts setting the price in the market, its the general poorly skilled public. even the bookies opinion from your example is going to be more highly skilled than the markets because he has some talent in this regard or he wouldnt be in the business. He is similar to mutual fund managers. Yet the market is worse because while there are semi competent mutual fund managers, there are also alot of irrational joe schmo buyers who drive prices or lines way past what is neutral EV setting up +EV bets.

[/ QUOTE ]

He's saying Buffett is the Bellagio and Joe Schmo is the market.

This may be a side point, but I'm not getting why Joe Schmo is necessarily better than a dart in his opinion. Obviously we should assume the market as a whole is, but why any one individual?

ahnuld 11-18-2007 12:40 PM

Re: Improving On Buffett And Desert Cat
 
reading comprehension ftw [img]/images/graemlins/tongue.gif[/img]

David Sklansky 11-18-2007 03:25 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
[ QUOTE ]
One last try. The Bellagio makes a line on football games. They are the best linemakers in the world. Their line is almost always perfect. So even a professional bettor who beats his bookie 56% of the time only picks 50.1% when he faces the Bellagio.

Joe Schmo is barely break even in his opinions. But he is better than the dart board thrower. (This is an important point.)

The Bellagio says the Yankees will win 70% of the time today. Joe Schmo thinks they will win 60%. In other words the Bellagio says that anyone holding a $100 even money bet has an equity worth $40. Joe says it is worth $20. Its TRUE worth is a tad below $40. Maybe $39.50. Change Joe Schmo to Mr Market. (And I would still love to bet that Buffett both agrees with me and doesn't think his words dispute my words)

[/ QUOTE ]

The information Joe Schmoe uses to form his opinion is presumably available to the Bellagio. Therefore, the Bellagio and Joe use the same information to reach their conclusions. The Bellagio tends to reach much better conclusions based on the information available. Thus, there is no reason to pay attention to Joe. Any difference between Joe's estimate and that of the Bellagio is likely the result of Joe's error.

Where Joe is correct, the Bellagio will also be correct. Joe's opinion is worthless - unless you're betting against Joe, in which case a large discrepancy between Joe and the Bellagio will indicate that you should put it all in there!

I don't know much about investing, or when to buy and sell, but I think your example here is flawed, David. It's no more likely that the actual value is $39.50 than that it's $40.50.

Very nice thread, makes me want to learn about financial analysis.

[/ QUOTE ]

First of all what I am saying is unquestionably correct in the real world. I will get to that in a minute. In the theoretical world I am saying that that if Jack is right about something 90% of the time and Jill is right even as little as 51%, then when they disagree, Jack's probability has gone down to 89.9% or whatever. If they are totally independent chances it easy to figure out the exact answer. You are saying Jack's probability doesn't change because they are not independent. He will get every question right that Jill will, plus more. If that was the case Jill's disagreement means nothing.

But that total lack of independence is obviously not the case in the stock market or sports betting. And once that is true, my contention MUST be true. It isn't even a contention. It is just an irrefutable math problem. If there are two differing opinions the true answer lies somewhere in between. On average. But closer to the guy who gets things right more often. As long as the other guy is better than random.

As to the real world, how can it not be obvious that I am right? Experts who use the Buffett-Graham-DesertCat technique make their play when their figures show they have some required big edge. When the smoke clears they are ahead, But obviously not to the extent that they thought they should be. Otherwise they all would be trillionaires. So the true price lays somewhere between their's and the market's.

Now why that should be the case is not clear. It is obviously sometimes because someone is illegally trading on inside information. But that is too rare to fully account for the syndrome.

When technical analysts say that the market will tell you where a stock is heading they are probably morons. But it is not moronic to say that the market is one of the best experts in telling you what a stock should BE. There are a few people who are even a little better than that. But if they are disagreeing with Mr. Market they should be very aware that the disagreement could signify that they have made at least a partial mistake.

BUT, their discomfort and trepidation should reduce if they can PINPOINT the reason why Mr. Market is disagreeing and refute his reason. When that happens they don't need to give themselves as large a margin of error.

I can't believe I'm going through all this again like I did thirty years ago in poker. You guys all need to just shut up, do what I say, and make more money.

stephenNUTS 11-18-2007 05:44 PM

Re: Improving On Buffett And Desert Cat
 
David,

I am really trying my best to grasp/enjoy/and even understand your posts for the most part in BFI.

But in my opinion,BFI deals with real life money issues and decisions for the most part....and your theoretical/on paper/+/-EV type replies,with War and Peace length transparent responses.....dont exactly accomplish much except confuse or ignite a debate to unnecessary extremes

Does every one of "YOUR" REAL LIFE decisions have to be put through some sort of POKERSTOVE-like program as well?

I have NEVER ever before seen anyone make something basically simple...and turn it into something so complicated than it really has to be?

Unless you are just kidding here also[img]/images/graemlins/smirk.gif[/img]

Stephen [img]/images/graemlins/cool.gif[/img]

David Sklansky 11-18-2007 05:55 PM

Re: Improving On Buffett And Desert Cat
 
I only delved into the theoretical to disprove a statement made by DesertCat. In actuality the point I am making is highly real life. Although it can only be used by people who have insight into why the public is making the price of a security different than you are.

Mark1808 11-18-2007 06:04 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
One last try. The Bellagio makes a line on football games. They are the best linemakers in the world. Their line is almost always perfect. So even a professional bettor who beats his bookie 56% of the time only picks 50.1% when he faces the Bellagio.

Joe Schmo is barely break even in his opinions. But he is better than the dart board thrower. (This is an important point.)

The Bellagio says the Yankees will win 70% of the time today. Joe Schmo thinks they will win 60%. In other words the Bellagio says that anyone holding a $100 even money bet has an equity worth $40. Joe says it is worth $20. Its TRUE worth is a tad below $40. Maybe $39.50. Change Joe Schmo to Mr Market. (And I would still love to bet that Buffett both agrees with me and doesn't think his words dispute my words)

[/ QUOTE ]

The information Joe Schmoe uses to form his opinion is presumably available to the Bellagio. Therefore, the Bellagio and Joe use the same information to reach their conclusions. The Bellagio tends to reach much better conclusions based on the information available. Thus, there is no reason to pay attention to Joe. Any difference between Joe's estimate and that of the Bellagio is likely the result of Joe's error.

Where Joe is correct, the Bellagio will also be correct. Joe's opinion is worthless - unless you're betting against Joe, in which case a large discrepancy between Joe and the Bellagio will indicate that you should put it all in there!

I don't know much about investing, or when to buy and sell, but I think your example here is flawed, David. It's no more likely that the actual value is $39.50 than that it's $40.50.

Very nice thread, makes me want to learn about financial analysis.

[/ QUOTE ]

First of all what I am saying is unquestionably correct in the real world. I will get to that in a minute. In the theoretical world I am saying that that if Jack is right about something 90% of the time and Jill is right even as little as 51%, then when they disagree, Jack's probability has gone down to 89.9% or whatever. If they are totally independent chances it easy to figure out the exact answer. You are saying Jack's probability doesn't change because they are not independent. He will get every question right that Jill will, plus more. If that was the case Jill's disagreement means nothing.

But that total lack of independence is obviously not the case in the stock market or sports betting. And once that is true, my contention MUST be true. It isn't even a contention. It is just an irrefutable math problem. If there are two differing opinions the true answer lies somewhere in between. On average. But closer to the guy who gets things right more often. As long as the other guy is better than random.

As to the real world, how can it not be obvious that I am right? Experts who use the Buffett-Graham-DesertCat technique make their play when their figures show they have some required big edge. When the smoke clears they are ahead, But obviously not to the extent that they thought they should be. Otherwise they all would be trillionaires. So the true price lays somewhere between their's and the market's.

Now why that should be the case is not clear. It is obviously sometimes because someone is illegally trading on inside information. But that is too rare to fully account for the syndrome.

When technical analysts say that the market will tell you where a stock is heading they are probably morons. But it is not moronic to say that the market is one of the best experts in telling you what a stock should BE. There are a few people who are even a little better than that. But if they are disagreeing with Mr. Market they should be very aware that the disagreement could signify that they have made at least a partial mistake.

BUT, their discomfort and trepidation should reduce if they can PINPOINT the reason why Mr. Market is disagreeing and refute his reason. When that happens they don't need to give themselves as large a margin of error.

I can't believe I'm going through all this again like I did thirty years ago in poker. You guys all need to just shut up, do what I say, and make more money.

[/ QUOTE ]

The problem with the example as I see it is the market price is not better than average at determing true value, because market prices on average are average. Now if you take a group of people on the other side of the trade they should tend to represent an average opinion and not a group with market beating knowledge. Where your example has some validity is if Buffett was on the opposite end of the trade of another market beating Graham disciple; in these types of trades you would not expect Buffett to do as well over a large sample size.

Table selection is very important in poker and I often observe where a respected player I know is sitting. His prescense at the table doesn't help me but I know he is very good at table selection and by going to his table I know I will be at a better table than I could pick on my own. In the same vein since company insiders have tended to outperform the market long term in their buying and selling I would like to own companies that are undervalued that are also being purchased in the open market by their own managemnet team and shy away from companies that I think are undervalued but where their corporate officers are heavy sellers. In this way I am not making trades opposite of those with market beating knowledge.

SlowHabit 11-18-2007 06:27 PM

Re: Improving On Buffett And Desert Cat
 
DesertCat's realized gains > Sklansky's Theory Buxs all day.

David Sklansky 11-18-2007 06:36 PM

Re: Improving On Buffett And Desert Cat
 
Firstly there is a difference between an average opinion and a random opinion. Secondly the market is not necessarily an average opinion. Do you think that the favorite on the tote board is always the horse that most people think will win?

And when you talk about avoiding acting on your opinions only if insiders are on the same side, you are actually agreeing with one aspect of my point.

Subfallen 11-18-2007 07:29 PM

Re: Improving On Buffett And Desert Cat
 
Isn't the "true" value of a stock only an idealized notion useful for analysis? All that really matters is predicting Mr. Market's valuation. (Which we imagine as "going to" the true value over time.)

But if the "true" value of a stock is Mr. Market's valuation at the limit, so to speak...then clearly DS is correct, as long as Mr. Market's valuation changes in a discernibly continuous fashion.

This is probably of little practical value, but theoretically it seems obvious.

Mark1808 11-18-2007 07:34 PM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]
Firstly there is a difference between an average opinion and a random opinion. Secondly the market is not necessarily an average opinion. Do you think that the favorite on the tote board is always the horse that most people think will win?

And when you talk about avoiding acting on your opinions only if insiders are on the same side, you are actually agreeing with one aspect of my point.

[/ QUOTE ]

I definately agree that if those on the opposite side of your trade are better than average then your expected return, even if the best, will be less.


I believe the favorite on the tote board has the most money on him to win, his price, or odds, is the collective wisdom of bettors who on average should do no better or worse in predicting than the odds given. Now if I was the best handicapper in the business and I saw a trainer with a great betting record taking another horse than mine, I might still have the best pick but the EV would not be as great.

I am just having a tough time seeing how the market price of a stock is a better predictor of its value than the average of all investors who in fact set that market price through their actions.

But I will say knowing the motivations and record of those on the opposite side of your trade can only enhance your EV.

ahnuld 11-18-2007 07:38 PM

Re: Improving On Buffett And Desert Cat
 
subfallen, it doesnt really matter if mr market gets it right or wrong ever, as we own a part of the company and are entitled to dividens/a share of the profit. So as long as we can forecast accurate cash flows and we pay a low price for those cash flows we dont care what the market does after it has given us the opportunity to buy those cash flows cheaply.

DesertCat 11-19-2007 12:48 AM

Re: Improving On Buffett And Desert Cat
 
[ QUOTE ]


First of all what I am saying is unquestionably correct in the real world. I will get to that in a minute. In the theoretical world I am saying that that if Jack is right about something 90% of the time and Jill is right even as little as 51%, then when they disagree, Jack's probability has gone down to 89.9% or whatever. If they are totally independent chances it easy to figure out the exact answer. You are saying Jack's probability doesn't change because they are not independent. He will get every question right that Jill will, plus more. If that was the case Jill's disagreement means nothing.

But that total lack of independence is obviously not the case in the stock market or sports betting. And once that is true, my contention MUST be true. It isn't even a contention. It is just an irrefutable math problem. If there are two differing opinions the true answer lies somewhere in between. On average. But closer to the guy who gets things right more often. As long as the other guy is better than random.

[/ QUOTE ]

I think I understand what you are saying, and it's mathematically sound. If Jack can guess right 90% of the time, and Jill can guess right 51%, you make perfect sense. If they were guessing a random event, like a coin flip. Assume Jack has a sixth sense that gives him 90% accuracy, Jill has a weaker version that only makes her 75% accurate, and their predictions are independant, i.e. when she is right he is sometimes wrong. If he is getting paid 9-8 on every right guess he should first ask Jill what her guess is, and if it's different than his, pass on the bet knowing his edge has disappeared.

But FA isn't guess-work, and I don't think your example applies. It's a logical, rational, framework for valuing an investment. The goal of a value investor is to find market inefficiencies, in a market that is usually efficient. In this case, Mr. Market isn't even being 50% accurate. So using an inaccurate "guess" can't make your rational evaluation more accurate.

This morning I thought you got here because sports betting involves lots of non-public information, and I think that makes a big difference. In sports betting it's important to take the line into effect when you do your own handicapping. You can never know if there is an unreported injury, or even a fix, that other betters are aware of, and averaging your results with the current line helps protect you from being cheated.

But in the stock market, there are many valuations you can do that don't involve any non-public information. And non-public information is much less prevalent and valuable than in sports betting. Public companies have disclosure rules, insider selling is controlled and reported, and the stock market is legal in every state with fraud is investigated by the SEC, state attorney generals and private lawyers. A company like Coke can be valued pretty confidently because it's value has little to do with any individual quarterly or annual report, it's value is based primarily on it's long term international and domestic growth rates. Value investors avoid companies where non-public information can radically change company valuations. Value investors seek out predictable, easy to value, stocks.

Imclone (the martha stewart) is one of the most well known examples of insider trading of recent years. But the CEO sold all his stock basically in a single day. If you valued ImClone fairly, and bought it at any time in the months before at big discounts to that valuation, you weren't cheated at all, there was no non-public information affecting the stock's valuation. You were only cheated if you bought on the same day the CEO sold. So even in this case it was a rarity.

And ImClone wasn't a stock a value investor would buy. It's value had no predictability, because it was all linked to a specific study and FDA authorization that was high risk. Even if you could handicap it well, it incorporated secondary risks such as fraud do to the importance of the test results. This is the exact reason why value investors try to avoid "fulcrum stocks" whose value changes dramatically based on a single event.

Your technique actually may be valuable for fulcrum stocks, but that means it can only be used rarely, esp. by value investors.

David Sklansky 11-19-2007 04:02 AM

Re: Improving On Buffett And Desert Cat
 
I will try to rephrase things differently. If there are stocks where my theory isn't true and my technique is unnecessary, then it is also unnecesary to have anywhere near the 50% cushion you require, when investing in these stocks, as long as you have any sort of reasonable value assigning skills.

The above is pure logic. You don't even need to know what a stock IS for it to be true.

Obviously fulcrum stocks are the most suceptible to my techniques. But the other stocks are not immune to them and when you say that you need a 50% cushion you have implicitly agreed. Your error is the PURE LOGIC error of not seeing how your insistence on conservatism MEANS you agree with me.

But enough with the theory. Let's get back to Buffett. Forget one example. He has presumably made thousands of trades where he rated the eventual worth of a security a lot different than the market did at the time of his trade. Assuming he honestly told you his original ratings then it could be checked five or ten years later. Your contention is that if that they were done one would find that the prices averaged out to his ratings. My contention is that that isn't even close. Yes they moved toward his ratings and away from the markets. But I bet they didn't even move halfway toward them. In fact if he used anywhere near the same criteria you use (think the real value is double before investing) the proof of that is one line.


All times are GMT -4. The time now is 04:39 PM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.