Portfolio Design – the basics

There are several factors that enter into the design of a portfolio. A couple of these are not normally talked about (the books on trading assume you will trade, the books for the average person assumes you will hold stocks for 10 years.) I’m going to start with a point that is very important, but usually ignored or just assumed.

What time period can you “work” stocks?

On a 10 year chart with weekly time scale increments you can reliably see the major movements of the stock within the longer term trend. Short term news and emotion is largely filtered out by the weekly average of data. You are seeing the impact of financial performance, of market schedules (like the expiration of options on the 3rd Friday of the month) and of external cycles like the announcement of employment data from the government. Things with time cycles of months or longer.

The same tools are used as for shorter time periods, but you get faster response on shorter time period charts since the data is averaged over shorter times. There are very tradable trends here and the tools are fairly reliable. You need to keep the longer term context in mind (bull vs bear market) but they tend to show through on this chart anyway. Shorter term events can still impact you, but these are easily handled with ‘trade rules”. (Things like having a stop loss order or buying when a stock touches a moving average rather than when it has run away to the upside in a parabolic spike – which is doomed to return to the moving average at some point [ by definition, the average and the thing it averages will be forced to approximate each other. No stock can grow to infinity, the planet is not infinite. Companies, and folks emotions, can only move so far so fast, and the price must catch up to the average, or the average to the price).

So the first question to ask is what time period can you “work”? If you want to day trade, using a 10 day hourly chart, you must watch that chart at least once per hour. Every Single Hour. Otherwise you are seeing what you ought to have done half an interval ago… Can you look at the chart once per day? Use a daily chart. Can you only look at the chart on Saturday? Then that 10 year weekly chart is your major tool. But can’t this work on any stock? Well, no.

Each stock and “ticker” has it’s own volatility. A flakey little South American or Chinese start up can go from rocket ride to zero in a couple of days. A gold stock can move 20% in a couple of days on international panic or a new gold strike. Earnings reports on a speciality retailer can whack it 20% in a day. If you are going to trade or invest in those kinds of stocks, you need to watch them much more closely, and much faster, than a 10 year weekly interval chart. Indexes and baskets of companies (ETFs or Exchange Traded Funds) work much better for the slow investor. Scaling in with buys over a few weeks rather than one trade-in at a point in time. So pick your time scale, and then choose your vehicles accordingly. Look at the 1 year daily chart of each target ticker. Does it have dramatic spikes and drops? Does it rocket up and plunge down? That’s day trader land. Does it move more slowly but predictably over months? That’s the ticker for the weekend warrior. There are plenty of each from which to choose.

Look at this GLD Gold chart and notice all the “gaps” up and down between one day and the next. It drives you nuts to try and trade this kind of jumpy ticker. Buy it as a long term hedge, but trade other things.

Coal, Oil, and biotech companies are almost as hard to trade. So start with things that are less “jumpy” (low “beta”) and work your way up to the harder stuff (or treat them as long term investment themes).

“Bet” Management

Trade 1/10th or less of your stake at one time in one “ticker”

It is better to trade one thing well than 10 things poorly. Focus on just one “ticker” to begin with for trading. You can add a few more over time.

Having less than 10 total positions is risky in that a 10% loss in 1/10 of your total is a liveable 1% loss. A 10% loss in your total is just too damaging. The only exception to this is your cash position. You can carry a lot of cash. It is also worth noting that things like SPY hold many stocks, so the risk of a CFO “Perp Walk” or a sector collapse are taken away. The only “single point of failure” is the company that makes the basket. I let myself carry larger positions in baskets (ETFs) from reputable companies.

Take your total account value. We’ll use $10,000 just because it’s an easy number to work with. You might have more or less. Divide it by 10. That’s $1000. This is the typical “size” of your typical “position”. If you buy more than $1000 of something, ANYTHING, you are taking excessive risk of catastrophic loss. If you lose 50% of a position (it’s now $500) you need to get a 100% gain to make it back. This law of mathematics has strong implications for the value of loss avoidance. Avoiding a loss is twice as valuable as getting a gain. You just can’t afford to take a loss on all your account. The magic bullet is cash. You can hold multiple positions worth of cash. There is some risk to this but for most folks that just isn’t very important. You pay your mortgage in your local currency and don’t care much about the other side of the planet. If you want to, you can diversify your cash across a couple of other currencies to guard against the risk of your cash taking a relative loss.

So, divide your cash into 10 “positions” worth. MAYBE diversify some of it into other currencies. Then only trade one “position” worth of cash on one ticker. Never put in more. If that position rises to a double (so you have 2 positions worth of money tied up in that trade) sell 1/2. Yes, that’s a simple trade rule, but a good one. If you’ve let a position run to a double, it’s just nuts to leave it on the table. Take off 1/2 and let the rest run.

I will sometimes allow for 2 positions to be counted as one. So, for example, I’ll buy both RCL and CCL as cruise liners. This is the opposite of a ‘pairs trade’ where you buy one and short the other (capturing the outperformance of one company over the other while neutralizing the sector and broad market moves). Buying both is a bet on the sector with less company specific risk. I think of it as a personally constructed “Cruise Lines” basket and trade it as a single thing.

There is a “Magic Number” for positions to hold. That number is 20. If you have less than 20, you are increasing your concentration in any one position and taking more risk that it can move against you (from the CFO being arrested or The Ministry of Stupidity Speaking against your industry or …) Clearly you would not bet everything on “Red 7” at roulette. One loss and you are out of the game. Similarly, you could put a chip on every square and win a tiny bit most of the time, only losing on “green 0” (and “green 00” on American wheels where excess greed and naive Americans let them double the odds against you with another “all lose” square…) The difference with stocks is that, on average, the game has about 10% odds in your favour instead of against you. That is the basic reason that “indexing” works. Buy the whole market and you get 10%, on average, over time. Your job is to beat that index. The premier index is the S&P 500. It has the 500 largest stocks in America in it. For companies, size matters, so this screen for size automatically catches growing companies as they become interesting and drops old industries as they shrink. It is a hard strategy to beat. Yes, it’s that good.

Now what are the odds that you can find a better 500 stocks than those that make up the S&P 500 ? Not good. As you drop the count, you have a better chance of finding a better set. You could look at 100 stocks every few days. You could look at 30 stocks even more often. The end game here is that if you have less than 10 positions, you have too much concentration and the risk that comes with it. If you have more than 30 or 40 it becomes very hard to track them all and ever harder to beat the baseline index. “20” is magic number in betting on blackjack with card counting (bet 1/20 the stake) and the same number works for stock positions. You can vary from 20, but only by increasing risk as you drop to 10 (so you must watch more often – like a day trader) or by decreasing performance over the index (as you approach 30 or 40 you have a harder time finding consistent over performers and it takes ever more time to monitor them).

So put that number firmly in mind. 20 positions +/- 10 with the 10 side being for rapid trades or close watching and the 30-40 side being for longer term investors. Only violate this rule if you know exactly why you are doing it and exactly what added risk you are taking.

The Value of Cash

Be very reluctant to ever commit your last 1/10 of cash to the market. It’s very valuable to have some money in reserve.

Why? There are times that a great deal presents. If you must sell something to get the new entry, you must wait for the cash to “settle” (be delivered by the other guy) before you can safely trade into a new ticker. This is presently a 3 day wait. Your entry point may be long gone in 3 days. There will be times that a sudden market reversal hits you. You need to protect your account FAST. You may have bought some position with money that was not settled (so now you are married to it for 3 days, or until you present “settled money” to cover that stock) or you may not want to sell a particular stock becuase of a pending dividend or for tax reasons. You can quickly “sterilize” your risk by purchasing several types of protective instruments. You need cash to do this. I’ll just mention what these things are here, we’ll cover them in greater depth later. You could buy an exchange traded fund that takes the negative side of the market (sells short) such as the QID that is the inverse of the QQQQ nasdaq 100 stocks. You could buy “puts” that are a form of protective stock option. Or you could buy a stock that typically moves in opposition to the one you own (If you owned an oil company, you could buy a trucker. One profits on rising oil, the other on falling oil). ALL of these strategies take ready cash to do them. No ready cash? You are stuck.

Getting Started

Every investor, at the time they buy or sell an investment, is a trader. You place an order and it executes at a point in time. While you can soften this a bit by “scaling in” (buying 1/4 or 1/8 of a position at a time over several days) that is really just making 4 or 8 trades. Knowing how to trade improves your performance as an investor. You don’t have to become a day trader, but you can benefit greatly from making your entry and exit trades better.

OK, so take your account. Divide it into 10/10ths or 20/20ths. Put one aside as reserve cash. Put one aside as your first trade position. What to do with the other 8? (or 18?) You could put one or two in other currencies. You could buy a couple of very long term investments (say the SPY is in a long term bull run and Brazil EWZ is also in a bull run, you could invest in the general market for the long term while you learn to trade). That would leave 5 ( or 15 )not committed and sitting in cash for now.

Realize that you are planning to leave that SPY and EWZ as INVESTMENTS for a YEAR OR MORE. If you don’t have that firmly in mind, you are just entering them as short term trades and deceiving yourself. You will be putting on 3 trades at once (your selected trade, plus EWZ and SPY) and not paying attention to 2 of them while taking a random time to enter them. Not Good.) Only do this if you are certain that those 2 (SPY and EWZ) are in fact long term investments and the 10 year chart says they are in a bull (rising) market phase.

I have an investment position in BRKA Berkshire Hathaway company that I’ve owned since 1987. Yes, I bought it at the bottom of the market crash that year. I’ve never sold it It has risen from $2800 to $150,000 and fallen back to $75,000, but is $90,000 or so as I write this. At times, it’s taken a cut by about 50% in it’s trade value at any one time (the week before I bought it, it was trading at $3600 …). That is an investment, not a trade. Years or Decades! Buying SPY today and planning to trade that position in a few months when you get good at trading is just entering a trend trade with an unplanned entry, no exit or protection strategy, and a random exit. That’s disaster in the making. If in doubt, leave it in cash.

Never have more than 20 total positions – both trades and investments – unless you are willing to put in serious time commitments. You just can’t do a reasonable job of tracking more than 20 things at once. Even if 15 of them are long term investments. It also becomes ever harder to accurately pick 20 things that are going to outperform the S&P 500 average. You might as well just buy the S&P500 (SPY) and sit on it as a long term investment. Maybe someday as a professional investor with a staff to support you with research you can break this rule and have 100 positions, but not before. Work your way up to the 20. If you find yourself with time left over at 20, then think about more. If you find yourself “forgetting” about a position at 17, then drop back to 15 positions.

I typically think in terms of a full position, half position, quarter position, and occasionally a “tiny” – typically 1/8 to 1/10 of a position. I reserve a “tiny” (which I pronounce TEE-NEE, not TIE-NEE) for those times that I just can’t contain my emotional need to own a stock. If you just can’t keep from thinking that this is just a stock you’ve Got To OWN!!!, buy a tiny. At least you’ve limited the damage if it goes against you and bought time to work the stock properly. I’ve bought amounts as small as 1/100th of a position for things that I knew were wrong, but just couldn’t sleep if I didn’t own a chunk. If you have better emotional control, skip the tiny positions. If you just CAN’T STAND IT to not own that hot stock, limit your total dollar volume in all your tiny positions to something like 1/2 a position (i.e. $500 for a $10,000 account or $5000 in a $100,000 account) and preferably less. At least it will let you sleep and get back to working on those full positions in things that matter. I don’t count my tiny and toy positions against my total count of positions. They are for entertainment and a loss of 1/4% of my stake is not significant.

When something is moving against you a little bit, but you can’t bring yourself to sell out a whole position, sell 1/2 a position. Do some more homework. Put a tight stop loss behind the remaining 1/2. Later you’ll be able to sell that final 1/2 and if the market DOES move up, well, you have 1/2 a position participating. I have a general rule: If I don’t know what to do, sell 1/2. When you look at something and it isn’t working out and you just don’t know what to do; sell 1/2. Think in cash.

I use 1/4 positions for new tickers that I’m not familiar with or for risky trades. My first buy will be a 1/4 position. As I get comfortable, I’ll move to trading 1/2 positions in that ticker. Eventually I’ll move to trading full positions. I’m not fond of “trading in” a partial position then adding more if it moves in my direction. This hasn’t worked well for me. It just dilutes my position, that had a good entry, with some bought after the good entry has gone and I’m closer to an exit… When I first started trading GLD with 1/4 positions (an exchange traded fund that holds gold) I was not ready for the “gap down” and “gap up” behaviour that happens. Gold trades world wide 24 hours a day. GLD has to “catch up” at the market open. It took me a while to adjust to this, and during that adjustment I wasn’t up to my eyeballs in a full position that gapped down at the open and took me out with a tight stop loss then moved up.. That use of an 1/8 or 1/4 to learn a new ticker is a good rule.

Portfolio Construction

At this point you know you need ten to twenty positions, and you know you want to make money. You’ve figured out your position size and decided what to do to scale in or trade in to positions. But what positions?

I run two styles at once. Some of my postions are investments. Others are rabid day trades. Most folks will pick one style and stick with it. A generally useful tool is something I call “racing stocks” (see the tab up top). This lets you graphically compare what stock is moving up fastest. Realize that the time period you pick for a “race” will to some extent determine the winner. A stock that has run up dramatically for a whole year is likely getting “tired” while the laggard in the same field may finally be getting a lift. The tired stock will show as the winner in the 1 year chart, but the new darling will show as the winner on the 10 day chart. Often the 6 month chart is best for markets that have just changed a major direction (recent top or bottom – picking the new winners early) while the 1 year is better for markets in a well established trend (markets tend to run up for 8 years, then down for 2 – strangely in rough sync with the sunspot cycle. Go figure.)

So what to put in your portfolio? Until you get good at picking and trading, start with broad market indexes and sector funds. What you put in also depends a great deal on who you are. A young kid investing for 50 years will want more risky growth stocks. An old guy planning to retire in 5 years will want more stable income funds and bonds (but bonds drop in value when the Federal Reserve starts raising rates after a market collapse, so right now is not a good time to be in bonds. They will be dropping in value “soon”). A woman may find that retail stores, especially apparel, or Mall REITS hold her interest (and frankly, most women know more about what’s “hot” in retail and fashion more than most guys ever will have the slightest clue about. I ask my wife and daughter for advice before I buy any retail stock, and I do it for a reason. It works.) Holding interest matters. It’s hard to trade what you don’t know, or want to know, anything about…

So first, know yourself. Then pick 10 “buckets” you want to be in. I’ll present a sample based roughly on what I’m doing as a risk taking older person. You will need to decide for yourself what is right for your age, risk profile, and needs. The first 10 buckets tend to be persistent (I’m tempted to say permanent). I add buckets 11+ as sectors come into or go out of “style” (i.e. start winning a race or showing up as a “bottom fishing” candidate with a new bottom ready to turn up.)

1) Broad Market Indexes

I race the SPY vs just about everything else. Right now the QQQQ Nasdaq 100 is winning. So I’ll trade into and out of tickers that track the SPY and QQQQ. Both can also be used as longer term investments. Sometimes I use IWM that tracks the Russell 2000 U.S. stocks Presently I have no index tickers for U.S. Stocks, but I probably ought to have QQQQ as a Tech component. The Tech sector is “hot” and I’m ignoring it.

2) Bonds & bond like high dividend stocks

See the Bonds Race under the racing stocks tab. Lately I’ve been using TBT as a “short sell bonds” ticker (a bet that bonds will drop in value). When it’s time to be in bonds, I typically use TIP – Treasury Inflation Protected Securities.

I also have an LNG tanker stock (TGP) with a 10% dividend and an Israeli cell phone company (CEL) paying 12% along with a couple of other fairly stable high dividend stocks that act more like a bond. I’m more of a risk taker, so I count this toward my “bond component”. A typical rule of thumb is to own the same percent of bonds as your age – so a 50 year old would own 1/2 bonds. I want more “juice” than that (and that silly rule ignores the cyclical nature of bond prices with the Fed) so I have some “high yield stocks” in with my “bond component”. Buy everybody needs a high yield component… I also have a couple of financial preferred stocks that act a bit like bonds, including a Bank of America preferred with 10+% yield right now.

3) International Market Indexes & near index funds


Brazil, Australia, India, China. About a 8 : 4 : 2 : 1 ratio for me. Probably ought to be more of a 4 : 2 : 2 : 1

It is worth noting that both EWA and IAF have large dividends, so they also count toward my “high yield” component in my higher risk approach.

4) Cash – U.S. Dollar and Foreign

Typically FXF, FXY or FXE for my foreign cash bets. Swiss Frank, Japanese Yen, or Euros. Sometimes the resource currencies of FXA and FXC – Australia and Canada. Right now I have a low cash level, and I’m looking at what to ease out of to raise cash back toward my goal level. Then again, the US Dollar has been dropping like a rock against other currencies so having a low $ bias is a reasonable position.

5) REITs – Real Estate Investment Trusts

Another traditionally high yield component. I’ve been picking up some “Mall REITs” at a low price. I’ll be adding some office and industrial REITS over time too. I briefly held some medical REITS but since we don’t know what Mr. Obama wants to do to the medical field, I’ve stepped out of them. I can get in AFTER The Ministry of Stupidity Speaks and tells me if they will survive or not… I also own a Timber REIT – PCL Plum Creek Lumber as both a high yield future (as lumber recovers in price), inflation hedge, and Real Estate play.

6) Gold, Precious Metals, and related Miners

GLD is a basket of gold. Sometimes I hold a chunk. Right now I’ve got GDX, the Gold Minders basket instead. As the economy recovers, platinum runs up and down. Right now it’s down. I’ve started buying platinum miners (SWC and related) I ought to have some SLV silver or silver miners like PAAS but you can only own so much… About 10% in the group stabilizes portfolio swings against sudden panics. When folks panic, they run to gold. It goes up while the other part of the portfolio drops. When the panic ends, the gold drops while the rest of the portfolio goes up. I’ve also got some FCX – Freeport McMorRan Copper and GOLD.

7) Oils

For some reason I’m fascinated with oils. Probably the tendency to move a lot (both up and down – called “high beta”) along with the fact that oil drives the costs of other industries. So oil goes up and cruise ships go down. Oil goes down, cruise ships go up. SOME oil stabilizes the rest of the portfolio. I also have a significant part of my oil holdings in “Oil Trusts” that pay a big dividend so I count them as part of my “high yield” component. PHG, PWE, LINE all have fat dividends. PBR is finding more oil deeper than anyone else on the planet.

8) Financials

BRKA is my major position here. BRKB is 1/30th of a BRKA share. I had a big gain in GS Goldman Sachs and some others and have stepped aside for a bit to evaluate them again (and moved a bit of money to China FXI)

9) Speculative and Toys

A couple of examples: SYNM and RTK make synthetic oil. I’m interested in the field. THPW has a Thorium based nuclear fuel cycle – it’s a penny stock that may do something some day, maybe. So I own about $200 of it. Yeah, a toy.

10) Agriculture and Ag commodities

MOO, JJG One is a basket of agriculture input providers – seeds and fertilizers. The other is a basket of grain futures contracts (JJG). Looking to add some farms, but no decision yet. CZZ counts here, as a sugar maker, and counts in my Foreign bucket as a Brazilian company, and kind of counts as an “oil” due to making fuel (ethanol from sugarcane). Sometimes It’s a judgement call, sometimes I just count it in all three…

11) Retail

JCP JC Penny Co., WMT WalMart, TIF Tiffanys, etc. See the retail races tab. I’ve also started adding some restaurant stocks that could be thought of as a whole new sector or, since it’s just a small part right now, put under “retail” of a sort. Technically not correct, but “good enough” for now. BNHNA Benihana and YUM Yum Brands (KFC, A&W Root Beer, Pizza Hut, …) Yeah, bracketing both high end and low and with a “China kicker” for YUM since they are opening KFCs like crazy in China (and visit a KFC in any asian dominated community and it is PACKED. Those China restaurants will do very very well…) Retail is a new sector for me (I’m fashion blind in the extreme) and it’s an early economic recovery play so still a bit volatile. But you make the most percentage gains “off the bottom” as long as you have enough Maalox …

12) Cyclical Recovery stocks

RCL, CCL Cruise ships, for example. To some extent the Mall Reits fall in the bucket too as would some of my other stocks. Retail perhaps. So you pick a bucket to count something in and don’t worry about it too much… I probably ought to add a “transports” sector, since I own some SEA (basket of shipping) but I count it under Cyclical Recovery for now since it will move with the global business recovery. I’m looking at industrial / farm machinery (DE Deere, JOYG Joy Global, BUCY Bucyrus, Cummins engines, etc.) but haven’t made any decisions yet. I’ve also put my DIS Disney and DWA Dreamworks Animation into this bucket, even though technically they are “media”. As the recovery develops I’ll need to stop using this bucket as an “Else” clause and clean up my sorting a bit more (i.e. decide if I really need a “media” sector or if I need to stop calling this a recovery bucket and start calling it Misto or Random Picks or…) Similarly, I could put all my oils, miners, grains and ag, etc. into a bucket named Commodities. There is some discretion here. Just keep the buckets different enough so that they all don’t move too much together…

So that’s what you need to do

Pick your sectors and buckets. Sort your proposed “buys” into those buckets. Try to keep the money more or less balanced between the buckets (or at least know why you have stepped out of one and doubled up another – for example my Brazil holdings have roughly doubled and I’ve not sold yet…)

I’ll add or remove categories as one sector or another starts to win a race. I’ll “shop” a new sector for a while, sometimes buying a “tiny” in my toys bucket to see if I really want to go there. Then, when the day comes, I have to decide: Drop an old friend sector to add the new one, or take on Yet Another Sector?

That’s how you design and manage a portfolio. You could add target percentages if you like (start with an even split between them – then adjust). You can be convinced that BRIC is going to win (Brazil Russia India China) and put BRIC funds in your Foreign sector bucket AND BRIC individual stocks in your sector buckets (that will increase your exposure to The Minstry of Stupidity in those foreign countries, but does maintain sector and company diversification)

So pick your list of buckets. Set your targets. Then go hunting in the Racing Stocks tab for interesting stocks and sectors (and countries!) Fill it out with 10 to 20 positions. That’s it. Now each day or each week examine your holdings. Race them against the SPY benchmark and those foreign country and some sector funds. If one of your buckets is not winning the race any more, start edging into a new sector (as you learn it’s quirks…)

Remember to allow for a risk premium when looking at a race; (FEED is winning a pig farmers race, but it’s a speculative stock in China while SFD Smithfield is a well established giant global player. One has great risk with the POTENTIAL for great reward, the other lets you sleep will with some reliable performance.) Sometimes the loser of the race will have the greatest percentage gains as it has a “dead cat bounce” off the bottom as a speculative “day trade”. So I usually go with the race winner, but only after assessing the risk and deciding if it’s PURPOSE is to be a fast trade (“dead cat bounce”?), income earner, speculative Vegas money, or steady earner steady growth investment.

Remember to allow for dividends. PBR trounces PGH in a race, but PGH pays me a fat dividend (trusts must pay out the bulk of their gains) and pays that dividend every month. So I hold one for capital growth with a bit of risk, the other as my” income with a bit of growth” bond like component. I buy PGH for my “Bonds” bucket. PBR for my “oils” bucket.

In Conclusion

Please let me know if there is part of this that is not clear of if there is a part where I need to add something. I’ve been doing this a long time and there are parts that I know so well I don’t even think about being aware of them. If they are missing, please let me know and I’ll add a chunk.

Click for Disclaimers, Disclosures, and Where To Get Charts

About E.M.Smith

A technical managerial sort interested in things from Stonehenge to computer science. My present "hot buttons' are the mythology of Climate Change and ancient metrology; but things change...
This entry was posted in Economics - Trading - and Money and tagged , . Bookmark the permalink.

1 Response to Portfolio Design – the basics

  1. BobN says:

    This was very good information for people starting to trade or for people that may not have had success.
    I do short side trades quite extensively and have made as much on it as I have the upside. Curious as to your views on doing this.
    I also trade options as a good hedge on my stocks, with Covered calls and collars whenever the stock hits an all time high. For option trading in general I find selling options to work better for me than buying options. Vertical spreads cuts the risks and margin requirements. I actually make more on the option trading than the stock.
    I would be most interested in any comments on options.

Comments are closed.