How do you guys choose from the list of 25? The book says to pick the top 5-7 stocks from the list every two months or so. What does it mean by 'top'? The web-site doesn't put them in order of their rankings. Anyway, there would be no point generating a list of 25 or more if you're always going to pick the top nine at most.Or is the list larger because you should pick ones you don't own yet? How different are the lists generally two months down the line? And a year?Another reason I can see why the list is larger is that it doesn't seem to be all that rare for a stock to have become delisted.I'd probably be tempted to start with the ones with the highest earnings yield. And then I think I'd avoid the ones with stupidly high PE. Greenblatt already advises to exclude the ones with stupidly low PE. Next I exclude the ones unavailable from my broker. The last point is a bit arbitrary, as they would add it upon request. I have to do this on a regular basis already with quite a few hidden gems.So my list of 5 becomes: DLX MTEX PWEI PNCL VPHMNotice that I excluded a few for other reasons. PLAY I already own, SCC is being acquired close to its current price and FDG has been on my watch-list for a while already but I keep seeing problems with it.I'd be interested what people think of this way of selecting and I'd be interrested to hear how other people do it.Mark
You might want to look at Vaalco Energy (EGY) as well which is on sale today, due to analyst downgrade... It's still one of the few in my 13 that are in the plus column, despite the 12% punishment so far today.... I have DLX - which had gotten some good press on the MF site as a potential turn-around candidate. I also liked the dividend. But last week they announced they threw millions away on an IT project they are abandoning - so I ma not overly impressed with the turn-around leadership potential of the current managament.Anyway, the reason that stocks are on this list in the first place is usually due to some problem - so, as others have said, we can either be experts in the field and make an educated guess which companies are unfairly punished or poised for turn-around, or, as in my uneducated case, we can play the odds that a couple of them will come back big enough to make up for all the bona fide losers.Maggy - Long EGY, DLX, ALDA, AMS, CALL, KSW, KSWS, MTEX,PTSC, PTEN, TGIS, FORD, JAKK
Anyway, the reason that stocks are on this list in the first place is usually due to some problemYes, I noticed this. Many of the companies listed make me first want to run away from them. But on closer look they also seem to provide a lot of room for gains when the coming year they turn for the good.I have the impression that on a global scale this method milks the concept that you have a maximum downside of 100%, whereas the upside can be many times that. If among the 20 stocks you'd pick there are two 10-baggers, ten of them become go bankrupt and eight go down 20% you're still looking at an overal gain of 25%. That would be an extreme case, but you get the gist.Mark
If we are to follow the MF system, we have to pick NEW stocks every few months, so that after a year we own about 30 stocks.I've been generating different lists to begin with by beginning with a different market cap. I haven't been as organized about this yet, but I've been thinking about making each quarter of the year a different cap size, so that my January buys and sells will be the smallest cap size (say), and the April buys and sells mid-caps. That way, I could generate the same MF list based on the exact same cap.... every Jan, look at the list for cap size 53, for example.This is my first year using MF, and so my portfolio is about half and half MF picks and prior picks.... it's messy and confusing at the moment!As to which stocks I choose from any list, I also run several companies through other tests I like (mostly from here at Fool), and other web sites' evaluation routines. If any stock passes several tests, I go with that one.Greenblatt says that if you are in for the long haul, it shouldn't matter which stock you pick from the list, that is, picking 5-7 stocks randomly should work just as well over time. Still, I like to understand the business I'm buying, and my little research routines help reduce anxiety.Hope this helps.Impish
This is my first year using MF, and so my portfolio is about half and half MF picks and prior picks.... it's messy and confusing at the moment!As to which stocks I choose from any list, I also run several companies through other tests I like (mostly from here at Fool), and other web sites' evaluation routines. If any stock passes several tests, I go with that one.Based on what you're saying in the second paragraph, I would advise you (not that you asked for advice) to accept the messy confusion in the first paragraph. You seem to feel compelled (like me) to do your own research. So I'd recommend keeping a portion of your portfolio set aside for pure MFI picks and another set aside for stocks where you do your own due diligence, following some other investment strategy. Otherwise, it sounds to me that you're at risk of mixing apples and oranges a bit too much. The problem with multiple screening a stock is that you run the very high risk of imposing incompatible investment strategies onto your list of candidates and therefore ending up with a company that is in some very weird situation which will lead to its eventual demise. So far, the only "second test" that's been very seriously looked at here has been using Piotroski to try and identify financially strong micro caps from the MFI screen. This approach makes sense for a number of reasons--both schemes are well-backtested mechanical approaches to value investing, and Piotroski would seem to offer the possibility of reducing one's exposure to companies that would go bankrupt after you bought them. (One might be better off just eliminating companies that Piotoroski would short rather than only buying those that he would buy. At any rate, nobody should use this approach without actually reading Piotroski's research paper and spending the time to really understand how the two approaches might interact.)However, I see this as a fairly risky endeavor in that the resulting sample size of companies is so small as to be potentially meaningless next to the numbers of companies that Piotroski used in his research. It is probably far wiser to simply stick to a random process of picking companies from the MFI screen. If one does choose to use something like Piotroski's F_Score to narrow down the field, it would seem prudent (necessary) to limit such purchases to a portion of your port set aside for gambling.So how to pick?Greenblatt makes it pretty easy by alphabetizing his list. You could definitely just pick the top 5 stocks from the list 5 times per year. Perhaps you'd want to increase the min cap each quarter by 500m or so. Or, you could pick 1-5 then 6-10 and so on from the top 25, 5 times per year, skipping stocks that were duplicates from previous purchases. I seriously doubt the usefulness of discriminating between companies on the list based on EY or ROC, or any other metric. We pretty much pounded into the ground the idea of, for example, limiting purchases to those companies that meet classic value criteria such as steadily increasing EPS--what we see is that those companies that make this list with such qualities tend to be in even worse shape than those with less compelling books. (Very small companies, particularly, with strong historical trends yet discounted share prices very frequently have some fundamental flaw, e.g., they just lost their only customer, etc.)Stick to randomly choosing from the list, if you ask me. In fact, over time you'd be just fine picking them all on one day of the year--the only reason to spread it out is to guard against starting the process at a market peak (i.e., it would have been a mistake, most likely, to start a full MFI portfolio three months ago).TDT
As usual, I like your clear analysis and presentation, TDT, but I'm wondering -- if they could all be picked on one day of the year, or if they can be picked on several different occasions, what difference would it make to the system if six are picked, the two top performers are sold off and replaced by three more off the list (with those extra $$$ received), and then that's repeated every few months. I don't mind holding the nonperformers for the whole year, maybe they'll perform at some point, but it seems like this would be a combination of Magic and Rule 1. Plus give me a way to work my portfolio up to 20. Also, I can't see that screening for companies you particularly like would be any different than if you had thrown a dart and those same companies had been hit "randomly." As you've said, any on the list should work, so why not pick some of those we feel more attracted to? (I really like the ones that show up both on the magic list and the MOAT list, that sounds like win-win to me!)Of course, Greenblatt had to make something very simple for his book and for backtesting, but so long as one keeps adding more and more underpriced stocks, then at intervals rolling even the laggards over as he suggests to do yearly, how would the "system" know the difference? Maybe the underlying benefit is provided more from keeping your money in the market and continuing to add underpriced stocks, (seems to be what the Fool experts are saying) and the gift Greenblatt's given us is the list generator. I like the sound of it ... will try to learn something about stop limits so I can "buy insurance" and wait for the good ones to go up a little more, then give it a try!STG
Unless you really know the companies you own I think it is best to keep with the strategy of holding a magic formula stock for a year. That is what has been tried and tested. If you want to see how it works to sell your winners after a few months to reinvest in more magic formula stocks I would recommend a pretend portfolio over doing that with your real money, and keep in mind the transaction costs. We know what the performance of this method is based on holding stocks for a year. We don't know the disparity between the biggest performers and the worst performers in any one portfolio but let me hazard a guess and figure it is large. The high return on average is just that, an average, some do much better, some much worse. Why would you want to sell the best stocks before the method states? You could very well be keeping the worst performers. I also find it easier to sell the outperformers than the underperformers but this is not the correct way to think about your investments imho. Spend more time in thinking about the company's present and future than the past. You don't want to become anchored to your initial price. Many roads to success, just my opinion. I am trying to write my own thoughts down more as I believe that will help my own investments. Whatever you do, track your returns, that is some of the best advice I have had from these boards. I am very interested in how your experience works out. I haven't used this method yet to invest, but I mean to start. (although I happen to have a few stocks that meet the MF criteria) I am afraid of all the transaction costs. :) Even 7 dollar trades add up.
what difference would it make to the system if six are picked, the two top performers are sold off and replaced by three more off the listYou'd lose the tax advantage built into the system, and make it impossible to ever have the really big winners that balance the really big losers. You'd continuously replace your best stocks, culled from a field of 25-30, with unknowns. The odds would be against the two or three replacements including any stocks as successful as the ones you just sold. It's not a backtested approach, so you're a test pilot on this one...but give it a try if you like. Let us know how it works out.J
Thanks for your thoughts, guys, I sure appreciate the caution. I don't see any negative tax consequences since all of this is in my Roth IRA, if I'm missing something there, please let me know. I will look at the trading costs to be sure they don't eat up so much gain as to make it a stupid move. As for replacing the best, I guess I'm comfortable that with a field of 100 to pick from, giving up two to get three give me good odds of choosing winners instead of losers, especially if I can find a site to run the MOAT test on (probably back on hoopmd's post), if Ilanbigfoot shares his most recent screen with me again (I liked your reasoning) and a few others keep talking about which ones they like and why (always interesting). For starters, I'll check to see if ASEI still looks good and is on our magic list. I really like the idea of getting 20 stocks in my magic portfolio faster than I'd be able to do with the IRA contribution each year (let's see, that'd be one stock a year for 14 more years.) Once I have those 20 (I'm trying to put about $4,000 in each stock,didn't Greenblatt suggest about that?) which will still take a few years, then I'll probably be able to see if the trend has been up or down or neutral and can go back to the guaranteed 30%+ method if I'm not doing as well. Since I just pulled these funds (only representing about a third of my total IRA) from a traditional mutual fund IRA (hope I'm in lowest tax bracket I'll ever be in, so did think of the consequences) which was increasing pretty slowly (and added them to this Roth IRA account which needs to do a LOT better so I can retire and spend more time doing research) there won't be much lost over what they would have done left in the old account. So let the fun begin! May the market bump up again on Monday!stg
I have DLX - which had gotten some good press on the MF site as a potential turn-around candidate. I also liked the dividend. But last week they announced they threw millions away on an IT project they are abandoning - so I ma not overly impressed with the turn-around leadership potential of the current managament.There's two ways of looking at this. I see it as a good sign that management admitted their mistake, abandoned the project, cutting their losses financially and eating crow.
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |