I posted a note on the Yahoo Groups Board showing the standings of my MFI stock contest. So far, just the top 25 over $50 M has been a strong performer. Toss on top of that the fact that my actual portfolio is only about 3% above the benchmark, while my tracking portfolios are probably more like 6-8% higher.
It makes one think. Have I been unlucky or do I bring a bias to stock picking? Do we all bring a bias?
I am going to think about this question. I have to say if I were starting over, I'd be strongly tempted to just buy the top 25 on a single day and then make no changes for a year. It'd be duller, but for an extra 5% I could live with dull.
I took a look at my portfolio to see if I could transition over the next year to that approach. It would not be easy unless I wanted to take a substantial tax hit (and I don't). The best approach would be for me to simply hold my current portfolio for a year (meaning some stocks would be more than a year) and then sell everything and "start over" by picking the top 25 over 50m.
I will think about this shift over the next couple of months as my next batch of sales will be in September. Then I wonder if there is a best time of the year to do this? There is always "noise" on the list as stocks fall off when earnings are updated, it seems you want a point in time when noise is minimal. But again, I am probably over-thinking.
What do you think? Post a comment!
MEOH did announce their earnings late last night (Methanex quarterly earnings fall). They didn't seem very good to me. But they did say a lot of positive things about the state of the industry.
Subscribe to:
Post Comments (Atom)
Posts
12 comments:
I think we all bring biases to everything including stock picking the challenge would be to find out what they are. By buying all your stocks at one time you of course risk buying at a market top. Maybe a better approach to eliminate bias and avoid tops would be to choose your stocks randomly, say by a throw of the dice.
You can buy 24 and 50 sided dice!! I think buying all 25 at one time would make you restless all year. I'm making purchases 4 times a year and that's hard.
As Dennis stated, I am not sure about the "buying everything at once" approach. I want to avoid throwing all the chips on the table at a market high. Would it not be better to average in monthly over a 6-10 month period?
The benefits of buying in all at once would be taking luck out of it. Whether it's our bias messing things up or just our lucky/unluckiness I'm not sure we can mathematically expect to mimic the top 25 buying 5 stocks at a time, there's too much variation. If you buy 5 every 2 months you could pick 3-5 bad picks every time and the 25 gets shuffled every couple months so its not like your just going to end up with all 25 after 5 pickings. Even if the 25 stayed static you could buy 5 at their high and sell at their low and someone else could buy the same 5 a year later at their low and have a great year the next year. I bought fcx at $65 but it was on the top 25 when it was down to $48. It's since moved to $95 which is a double if you bought at the bottom but only 50% gain for me. Ideally to match the formula you could get better results buying 25 every two months. In the real world though commission costs would eat in to the returns unless you had a large enough amount to invest.
As you know, I think it's a good idea to keep bias out of the picture, and I think that buying over the course of the year is the MFI equivalent of dollar cost averaging.
Of course, I still say it will take several years to prove whether any of this is remotely correct vs. dumb luck. :)
-A
ps. did you notice that I was in 9th place right now with my random portfolio? Maybe that will also add some support to your decision...
You are basically proving out the superiority of the simple method of MFI. "Just do it!" It doesn't really matter much the "when". I liken this MFI system to the "passive investing" approach - - - buying a few all inclusive US and International stock indices and hold. You'll beat over 85% of money managers long-term. Same principle applies here, it seems.
I entered the contest by randomly picking 15 of the top 25 over 50 from scraps of paper. I am #3 out of 55 so far. I know, I know, we still have a long way to go. I do think the random approach will win out in the end. Just look at your results and more important, just look at MFI random results for 17 years.
random.org is calling you...random.org...random.org...
If you want to eliminate biases, I don't see much difference between buying all 25 at once and randomly buying one stock every two weeks. As long as it's random within the top 25, you should be doing the same thing. Although, if you are prone to tinkering, buying once a year will keep you more disciplined.
I don't know about that....what if I had put 100% of funds into 25-30 MFI stocks a week ago? Of course time will tell but I'd much rather average in and either a) catch higher highs or b) buy in at lower prices over a protracted retracement.
It would be interesting to do some back testing of random "invest 100% at once" vs an averaging in approach to starting up with MFI.
I've done some research about Dollar Cost Averaging and have found that buying stocks in lump sums is better 67% of the time compared to buying a little over time. This makes sense if you look at the entire history of the dow, it "tends" to go up. In theory, the sooner you buy, the better.
This is for a long term approach. I'm not sure if if this would work if you sell all 25 stocks after one year and buy 25 more and repeat the process.
My 2 cents.
-Travis
I appreciate all the commennts. Mathematically there is no difference (over the long run) between buying 25 all at once or phasing it in and buying 25 over a year and being totally random. I just worry that if I went with option two, I could not be totally random and that I would refuse to buy several stocks. I agree with Nick that I could get restless, that is a fault of mine. But I might be willing to take dull for more $ every year.
Of course if you follow the book then Joel recommends averaging in over a 9-10 month period...then selling each stock after being held a year.
That does not sound at all like an all in at once approach!!
Post a Comment