Investing Process - or how to think about investing

[suomalainen]

On another thread, and I've run across this attitude in other places, I've been dismissed when asking about people's investing philosophy. "If you don't already know how to invest like a pro, like me, I'm not going to tell you; go get yourself educated, just not here" is the attitude. So, I thought I'd start a thread for those brave enough to open their kimonos about their investing process or how they think about investing.

THIS IS NOT A THREAD FOR STOCK TIPS, but if you find a trade you made illustrative of your process, feel free to share.

For example, are you:

1) a chartist? watching moving averages and double carrot-tops and whatever else? (I don't know the terminology; I'm clearly skeptical)
2) a fundamentalist? reading dozens of 10Ks per week?
3) an indicatorist? like ratios? which ones and how many?
4) other?

As for me, my process is:

1) I try to stay on top of market news. This is pretty easy in my job.
2) If there's an industry that is in a downturn, I pay special attention.
3) If there's a company (whether in a downturned industry or not) that is unfairly punished for bad news, I buy. VW and GLNCY are two examples.
4) I try to not sell, but I sold VW and GLNCY after nice gains. To be fair, after my 100% gain in GLNCY it doubled again. Woops.
5) If I don't have any ideas generated from the above, I pick industries I want to be in for the long-term and I try to find stocks that aren't too overvalued via various ratios, PE chief among them, and I like dividend stocks, just because I like to see the checks roll in. I also check debt to ebitda (my brokerage doesn't have an easy column for this) because I don't want to invest in a company that has a risk of bankruptcy liquidity. In theory, I would also read a 10K to check the debt maturity profile, but...well, I'm not into reading 10Ks for fun so this has only happened once or twice. This step 5 is really just dart-throwing, I'm not afraid to admit.
6) Sometimes an investor friend of mine, who is an actual professional, fundamentalist investor for his day job (not equities), will give me a "hot tip". I made money on Mexico (EWW), Turkey (TUR) and short government bonds (TBT), none of which I would have thought of myself.
7) I don't have the conviction to put a substantial percentage of my personal account in any single trade idea, so I end up pretty diversified just clipping coupons.

What is your process?

[IlliniDave]

I'm a quasi-boglehead. I don't trade. I don't pick stocks. I do use valuations, inflation, gdp growth, and dividend yield (typically off Shiller's site or multpl) to estimate medium-/long-term return on the SP500. But it is not for making investment "moves", it just provides an ongoing measure of how robustly my savings targets might achieve my goals. I do some playing on the margin in my small Roth IRA, and I use sort of a contrarian strategy, buying funds that cover market segments that have been battered/lagging. It has not panned out very well.

Very boring and probably not very useful, other than to someone out there reading who might be stressed about hearing opinions that elegant investment strategies and the like are something they are required to do to stay afloat. Success and simplicity are not mutually exclusive.

[George the original one]

If you trade, keep track of the trades so you know whether your ideas are working or not in a statistically meaningful way. One big score out of 5 trades with the other 4 trades all over the map is not meaningful, especially if that one trade used different criteria.

If investing for income, whether rental property or dividends or bond yield, make sure the income is growing faster than inflation.

When using leverage, make sure you're getting paid a premium for taking extra risk. In my leveraged income portfolio experiment, I found that the premium I was paid for the added risk was close to nothing, so choose not to use leverage now that I'm retired (not to mention that interest rates are now going the wrong way).

[jacob]

I worked in the algorithmic trading department (Note: I don't use any of these methods in my private account---the knowledge I developed professionally is almost entirely useless for retail investing) in a company that also had "human"-traders, that is, people who trade manually. At one point, the bosses decided that it might be a good idea if the algo-group got to interview/talk to some of the human star-traders (that is people whose existence under the random walk hypothesis would be astronomically impossible(*)(**)) about what their "process" was. The hope was that we could build a trading algorithm---which is obviously a very explicit and entirely context free computerized process---around it. Imagine a computerized version of an expert trader. Wouldn't that be nice?

(*) This is also why it's a bit frustrating/annoying to always read on the internet what "nobody" can do when I know many people who in fact do do this.

It quickly became clear that none of them had much of any kind of deliberate process. At least not one they could formulate in words or on a list form.

(**) You can verify this experience by reading the biographies in the Market Wizards series.

Instead they had developed hundreds and hundreds of experiential rules from which they could intuit when to put a position on and how much. The best amongst them were somewhat consciously aware of these rules and capable of changing their behavior and trying the opposite even if it didn't "feel" right to them. They were de facto capable of rewiring their own intuition. This level of mental self-awareness is impressive to say the least.

Needless to say---but I'll say it anyway---we couldn't "algorithmitize" that intuition... because that would be the equivalent of solving the hard AI problem. Intuition is required to know which rules are the important ones in this particular case and which are not.

Another analogy might be the process of catching a random ball thrown at you with your hand. It's in a sense very easy to do so... but describing the process of exactly how you do it in a context-free manner is very hard. Consider how tricky it is to get a robot to do that. Same with crossing the street. You look at the cars and intuit whether it's safe. If you had to describe it as a process, you'd need differential equations set up in some min-max system. But of course none of us think of crossing a road in those terms. We just do it after lots of experiantial practice of gauging distances, moving our legs, etc.

While you don't know what specific cars will be there in the (future) and when (timing) they will be there ... you do know that the rules for crossing streets will be the same in the future. IOW, you don't cross the street by putting on a blindfold and following a process of "if sound of car is not detected, walk 30 steps forwards".

It's the same with investing. To some degree, we look at something, and then we just do it. And we do it correctly, because we've practiced a lot.

Now trading is very different from investing, investing-expertise works along the same lines. There are hundreds of experiential rules that are connected together to form some kind of intuition. I keep stressing the importance of reading finance101/202 and investment101/202 books in order that the "fundamental science" is correct. In other words, so that I don't come to believe the financial equivalent of a flat earth just because it has always "looked" that way to me in my experience. It's been my experience that many are highly resistant to the idea of developing the basic knowledge required to understand complex fields.

I suppose I mostly look at a combination of indicators but that's mostly to quantity what I'm otherwise reading and thinking about. In general, I think strategically (not tactically). This means I have an idea of where "the armies of sentiments and indicators (stock or reservoir)" looks like. That is via the process of reading and looking at metrics. I also have an idea of what "the landscape of sentiment-changes and indicators (flows, etc.)" is going. I then try to arrange myself (my portfolio) so that it "wins battles" not based on tactics but based on being in the right place (like not a swamp). I emphasize position more than timing.

IOW, it's a web-of-goals mindset but for investing. Like with the standard WOG, I usually need more than one reason to buy or sell something ... and positions NEVER exist in isolation. Positions are always in relation to other positions. IOW, the whole is greater than the sum of the parts. Obviously diversification is not something I pursue as a goal. Obviously, my individual positions therefore don't translate into buffet-like stock tips. I don't do my main analysis of the company itself but rather of how it relates to other aspects/companies of/in the economy.

I also know a hedgie who's a fundamental analyst. He "retired" after finding ERE and now runs his own portfolio. As far as I understand he does pretty well for himself. What he told me is that he does read 10Ks but that he goes somewhat beyond Dupont analysis and other simple stuff you might find in the 202 books. Instead he has dozens of different mental models; one for each type of company. This is similar to Munger's latticework except his is business-specific. He then perceives the company with the framework of that mental model. This makes it a lot easier to see what the deviations are and if a company is doing something really well or really bad. He then does standard value investing informed by that. He owns some weird stuff (companies I've never heard about).

So in all cases, the "process" is something like.

1. Study a lot to internalize the fundamentals; whether that's a latticework of business models, sector models, or Musashi-like self-awareness.
2. Pay attention to the market and do some standard stuff to translate it into familiar variables: ratios, price points, support levels, news frequency, whatever...
3. Insert the magic ingredient. (That can not be explained in words.)
4. Profit.

So, while there are [always] some "mechanical" ingredients in there, they're never the whole thing.

[suomalainen]

they had developed hundreds and hundreds of experiential rules from which they could intuit when to put a position on and how much. The best amongst them were somewhat consciously aware of these rules and capable of changing their behavior and trying the opposite even if it didn't "feel" right to them...[my hedgie friend] has dozens of different mental models; one for each type of company. This is similar to Munger's latticework except his is business-specific. He then perceives the company with the framework of that mental model. This makes it a lot easier to see what the deviations are and if a company is doing something really well or really bad. He then does standard value investing informed by that.

Thanks, this is helpful. Some of my clients/colleagues have described their "process" as well in similar terms when I've picked their brains about investing. One reaction and question though: I feel like I have a perfectly sound understanding of the 101-202 level stuff, from a mixture of classes, self-study and talking to my clients/colleagues. Where I run up against the wall is that I think the time and effort it takes to develop the "hundreds and hundreds of experiential rules", the feel, the intuition, is in the "consistent effort for at least several years" range. In other words, you have to read and read and read market reports, industry reports, company reports to be able to establish those frameworks so you can get a feel for the parameters of an industry and who the best players are within that industry, etc. Is that right? Is there another way to ratchet up the learning curve?

It's almost like investing is a full-time job! And since I already have two, I can't find the time for a third.

[jacob]

It's not a fulltime job but it is an experiential education (as opposed to an academic education .. or no education at all). That education significantly cuts down on the time one needs to spend on an ongoing basis. It's not like lessons have to be continually learned or maintained. As far as I'm concerned there are no short-cuts to education other than the normal ones like reading textbooks instead of blog posts, etc. I'm not sure reading various reports does much to help one develop an investment thesis, but I don't know. It's not something I spend my time on. Some do though. They don't write those reports just to toss them in trash.

Keep in mind that the efficient market hypothesis (regardless of how strong or weak you think it is) only applies to information. People will inevitably reach different opinions based on the same information because people have different kinds of knowledge---and the EMH does not apply to knowledge. The difference is not in having more or better information but in having more or better opinions(*). This comes about from thinking. Short-cuts kinda imply to me that there's a desire to spend less time thinking if possible... That's not good.

(*) An opinion here being defined as either long, short, or out.

[suomalainen]

@jacob, no I get it what you're saying, and I'm not looking for a short cut. My short cut is just not doing it! But the "it" that I'm not doing is doing massive amounts of reading to get a "feel for the market". Maybe I'm not saying this right. For example, I'm a lawyer right? I have expertise in a certain area. Pick me up and plop me in a totally new area and I can still do the job...but only at about 90% effectiveness compared to someone who has expertise in that area and knows the parameters of that particular market. I can do M&A deals, but I don't have a good sense of "the market" for this unique-to-M&A-deals-provision or that unique-to-M&A-deals-provision, because I don't do a lot of those deals. I have to rely on my M&A lawyer and ask "Is this a fair provision? Is this common in the market?"

So, when I think about "experiential education", I think what you mean is this idea of what's "market". For example, certain industries are asset heavy (mining) and others are asset light (accounting). I shouldn't think about mining firms the same way I think about accounting firms, but how can I know how to think about any firms without doing a lot of reading about mining firms or accounting firms (i.e., "market reports, industry reports, company reports")? So that's how I hear what you're saying, but it may not be what you're saying.

So, can you clarify what you mean by "experiential education"? Thx.

[frommi]

Thanks jacob.The guys at Renaissance Technologies have proven that purely "quant" models work very well, so i wouldn`t say that finding algos that work is impossible. But in the end everybody has to find a way that works for them, i don`t know anybody whose investment process looks exactly the same, because it is something that has to fit the personality of the investor/trader. If you haven`t developed your investment process yourself it is very hard to stick to it when shit hits the fan.

[jacob]

@suomalainen - If I understand you correctly (I'm a little confused here), a field (or subfield) are governed by certain principles. So if you're plopped into an adjacent field where you're 90% efficient, you only have 10% to go. [Would also remark that in order to "beat the average" ... you're already there]. As you spread out to the next area, you have already have two sets of "education" working, so maybe you'll be 92% efficient. That's how the renaissance person strategy works. The first one is hard to learn, but it gets progressively easier the more you add. The reason is you can reuse what you already have.

Once you have the model, you don't have to read the entire report to build up the model. You just have to skim through to see what's different/unexpected, etc. For example, I rarely read the FOMC statements. However, for years I've been saying [correctly] that the Fed is not going to raise rates as fast as they say they are. I've been saying this because in my model [of interest rates, debt levels, and trade deficits] this would be "unpossible". Now, if most of the market believe the Fed at face value, then interest sensitive securities would be undervalued. Therefore I had ONE reason to buy them. I will generally require more than one reason. For this, I tend to use a simple kind of majority voting system. That is, if I have 3 reasons FOR and 2 reasons AGAINST, I might dip my toes. 5-0 would be a larger position, etc.

The trick, if there is a trick, is to grok "what's really going on" (Plato's Cave and all) ... and then head back and compare reality to the shadows. If the shadows are different from reality but bound to move in the direction of reality, it's time to lead that.

The experiential part simply means experience-based---as opposed to academic. Especially in investing, there's a big divide between academic research and "practitioner" research almost to the level where people pay scant attention to academic papers (because they're often decades behind due to lack of data and manpower). For example, before there was an options model, people would happily trade options and not too bad either. It turns out that ATM deltas were traded so they value was about 0.5. This is also where mathematical models say they should be. So there are two ways to learn how options should be priced. One is to do the mathematical derivations of Black-Scholes. That's the academic way. The other is to trade them a lot and develop a feel for how their prices react when the underlying moves. That's the experiential way. It might be interesting to note that BS is now so popular that what traders actually think about in terms of trading is the implied volatility. According to BS, this should be constant over all prices. It is not. Traders buy and sell based on how ivol differs from where they think it should be. BS itself is now seen as a purely mechanical translation. IOW, the abstraction level is +1 higher than it used to be. One needs to be at the abstraction level where the current trading is actually happening. Otherwise, one will just be steamrolled.

@fromni - Oh, I didn't say that quant methods don't work. They do. What I meant was that it is nearly impossible to build an algorithmic version of a human trader (and vice versa). Analogously, if you want to build something that throws fist sized objects, it's better to do it with a potato gun than by building a robotic model of a baseball player's arm. Algos can do things humans are bad at and vice versa. They all belong in the eco-system.

[daylen]

I do not have a whole lot of investing experience myself, but I have read some investment literature.

Successful investing requires an understanding of how the price and value of assets, businesses, and industries change over time. Each business/industry is a set of tools and each tool can be analyzed to estimate their individual value and price. The fragility of each tool in relation to each other also seems to be useful in determining potential cascades.

Different investors operate on different time scales extending into the future based on their age and other subjective circumstances. Do you want to explore the macro-cycles of history and project them into the future? .. and/or do you want to focus on the micro-cycles of the news/reports to figure out what will happen in a month? I seems to me that both are required for successful investing; the macro-cycles increase awareness of value, and the micro-cycles increase awareness of price. To become more systematized you could document the set of circumstances (or triggers) in which you buy or sell an asset (investment journal). This at least provides you with some data about your past decisions that you can learn from.

[jacob]

@soumalainen - I should also note that there's a general conception that there are experts out there will to vacuum up every inefficiency and therefore inefficiencies do not exist. It's like that joke about the $20 bill on the street which the EMH guy claims is not there because if it was, somebody would have removed it already, QED. In reality, you find money on the street all the time. It's usually pennies and nickels though; rarely $20 bills.

Similarly, if the market has a $70,000 inefficiency lying around, a team of professionals with a total compensation of say $500,000/year + 100% overhead (I have no idea what the order of the overhead is) is not going to spend six months trying to come up with a way to eliminate that inefficiency. They're just going to leave it on the proverbial ground. They will only stoop down and try to pick it up if it's worth $1M or more.

If you think meta, that's a business decision similar to deciding whether or not you personally should spend time learning how to find inefficiencies or whether you're better off working overtime in your dayjob. Somebody is similarly deciding what those professionals they hired should spend their time on.

So for these smaller five or six-figure inefficiencies, you're mainly competing with other amateurs or small-timers.

Add: This also explains why one can't just open up a fund and let others in on it. The inefficiencies are finite in size. They don't scale.

[frommi]

+1. As a small guy the best thing to do is look at the tiny stocks, this is where the competition is small. Its just that people rarely do this because its deemed to be risky. But risk is mainly in not knowing what one is doing.

[suomalainen]

@jacob I get it. Dunno if I have the skill for it, but I get it.

@frommi (and maybe also @jacob?) I think you mentioned Buffett doing this in the 60s in the other thread. Aren't those the "cigarette butts" he'd pick up off the ground for a one or two puffs and he said something like Munger was the one who "cured" him of this approach? But you're saying that if you stay small, you're content with one or two puffs? And Munger's "cure" was really only useful in the sense that they went "upmarket" with greater AUM?

EDIT: Also, can I just ask (and I wonder if this is contentious), but is there a difference between "trading" and "investing"? The $70k inefficiencies/cigarette butt thing seems more like a "trading strategy" or methodology rather than an "investment strategy" or methodology. Not trying to be contentious or prove a point...but in my mind there is this dichotomy and I'm curious about your takes on it. Maybe it's a nomenclature distinction without a difference.

[ThisDinosaur]

I don't believe in strong EMH or "nobody can beat the market."
I do believe that the market portfolio represents a weighted average of investor opinion. And that that is worth knowing.
I do believe that there will be diminishing returns from effort : alpha for retail investors.
I use index funds and ETFs for equity diversification.
I don't try to convince traders not to trade, or stock pickers not to pick. But I like that no single company represents >1% of my NW.
I believe in diversification of asset classes as well as strategies.
I believe essentially any occurrence in financial markets has some precedent in history. So I try to learn the history.
I believe the number of things I don't know will always outweigh the things I do know. So I try to hedge for that.
I believe that my strategy may change as I learn new information. I try to hedge for that, too.

Amen.

For all I've learned since I started investing (+/-6 y ago), if I were starting from scratch right now, I'd probably just do permanent portfolio plus some heavy international exposure.

If anyone wants to pick apart any part of that, I'm all ears.

[Stahlmann]

I write this post, because browser bookmarks are too mainstream nowadays. Google doesn't forget compared to my limited mind

Hoping to see a solid chunk of knowledge.


My plan:
To get indexes described in "Are there any better indexes than SP500?". It's not so easy in EU. Yep, you can read my knowledge with this (actually, there was the topic where pizzas described your taste in the investing).

EDIT:
Is it allowed to create "Investing/(whatever): best threads on forum" article on the wiki (only with links)?
I mean google uses its algorithms/AI for bigger sites... but I must say it's pain in the ass to find such valuable threads on ERE with search engine.

[IlliniDave]

EDIT: Also, can I just ask (and I wonder if this is contentious), but is there a difference between "trading" and "investing"? The $70k inefficiencies/cigarette butt thing seems more like a "trading strategy" or methodology rather than an "investment strategy" or methodology. Not trying to be contentious or prove a point...but in my mind there is this dichotomy and I'm curious about your takes on it. Maybe it's a nomenclature distinction without a difference.

To me they are different. Trading is buying something for the purpose of (hopefully) selling it at a higher price. Typically it is a short-to-medium term proposition (sometimes only milli- or nano-seconds apparently). Investing (when talking about stocks) is buying for the purpose of gaining a "share" of the revenue stream and (hopefully) its future growth, and is typically a medium-to-long term proposition.

Two caveats: these are iDave takes and academic types may disagree, and the Venn diagram between them can in some cases have a lot of overlap.

[jacob]

EDIT: Also, can I just ask (and I wonder if this is contentious), but is there a difference between "trading" and "investing"? [...] Maybe it's a nomenclature distinction without a difference.

Just to be clear, microcaps is not the only game in town for small potato inefficiencies.

I don't really see much point in making a distinction wrt nomenclature. I think most of the terms (investing, trading, speculating, saving) tend to be used in loaded ways to make the speaker feel good about themselves. Many will distinguish between trading and investing in terms of the time horizon (investing is long term, trading is short term). The other distinction is "investing" for the high-falutin whereas "saving" is low-brow. Also "investing" gives the air of being safely scientific whereas "speculating" implies that one is gambling.

In the end, it doesn't matter: They all deal with each other.

If I had to make the distinction, I'd say investors are concerned with changes in underlying security or economy, e.g. is the company increasing earnings; will interest rates rise; ... whereas traders are concerned with changes in the market pricing system itself, e.g. supply/demand; sentiment; psychology; ... Under this definition, investors wouldn't mind sending their orders by homing pigeon and only checking prices once a month whereas traders would be out of a job. Conversely, traders would trade anything at anytime often without knowing what it is they're trading responding only to the market psychology of other traders. There's a famous possibly apocryphal example with Joe Siegel successfully trading "green lumber" believing it was wood painted green. That kind of trader-attitude is not that unusual. So traders don't really care if what they're trading represents some underlying reality(*) or whether it's just baseball cards or whatever.

(*) Which for stocks is the interest rate for equity financing.

[IlliniDave]

I'm surprised to see no discussion of what some believe is the number one contributor to success versus failure in investing: behavior. In my case I've little doubt I have the intellect to swim with the big fish, but I don't have supreme confidence that my reptile brain won't betray me at the worst possible moment. Recognizing that "contradiction" seems to help me avoid RBSUs, but even with a ~30 year track record, I don't know if there's enough data for statistical significance.

[jennypenny]

When I use the term 'trader', I'm talking about trading as something completely different than investing. Investors see value in owning a product. Traders see value in selling a product. Obviously, that's my own personal take on it.

re: experience -- You can gain a lot of experience without actually putting your own skin in the game. It's very easy to set up different portfolios and watch how they move in different market situations. It also gives you a chance to see how you react to the movement. I started with a dummy trading account and made some really big mistakes before venturing into trading with a live account. I'm assessing several asset allocations for our retirement needs and I have them all set up and running right now (mostly pulled from Portfolio Charts ... thx Tyler ). I set them up like I would for our retirement and I'll 'pull' money from them quarterly to mimic our future behavior. It's the only way to know how I'll react to each and which fits 1) our needs and 2) my temperament.

[IlliniDave]

jennypenny, I'm impressed that you can do that. Whenever I've tried to do the same my reptile brain sniffs out that it's a simulation making it too easy to be dispassionate and make it all like a mathematical homework problem. On Monday I "lost" enough money to buy a fishing boat, a trailer, and a high mileage used pickup to haul them around with. I didn't panic (actually didn't even know until after markets were closed), but all I could think about the rest of the night was "It's easy to be calm now while I have a paycheck coming in ..."