SystemTrader's Account Talk

Status
Not open for further replies.
Sort by newest Sort by date
imported post

The buy-and-hold crowd (especially those in the financial services industry) often discourage market timing by claiming that you may miss the best performing days, thereby drastically reducing your return. They often show figures and charts like this.

The average S&P 500 return from January 1984 through December 1998 was 17.80%. Here's what it (annual return) would look like if you missed the best days during that period:

If you missed the BEST:

[font=arial,sans-serif]0 days (this is buy-and-hold) [/font]
[font=arial,sans-serif]17.80%[/font]

[font=arial,sans-serif]10 days[/font]
[font=arial,sans-serif]14.24%[/font]

[font=arial,sans-serif]20 days[/font]
[font=arial,sans-serif]11.99%[/font]

[font=arial,sans-serif]30 days[/font]
[font=arial,sans-serif]10.01%[/font]

[font=arial,sans-serif]40 days[/font]
[font=arial,sans-serif]8.23%[/font]



The chart and textright above it is from an email from D.R. Barton, Jr at Traders U. Barton goes on to demonstrate what buy-and-holders don't tell you...(Note: allof the notes in italics are fromBarton; the restare from me.)


And now, here's what happened if you missed the worst days for the same period (the numbers buy-and-holders never let us see):

[font=arial,sans-serif]If you missed the WORST:[/font]

[font=arial,sans-serif]0 days (this is buy-and-hold)[/font]
[font=arial,sans-serif]17.80%[/font]

[font=arial,sans-serif]10 days[/font]
[font=arial,sans-serif]24.17%[/font]

[font=arial,sans-serif]20 days[/font]
[font=arial,sans-serif]27.04%[/font]

[font=arial,sans-serif]30 days[/font]
[font=arial,sans-serif]29.45%[/font]

[font=arial,sans-serif]40 days[/font]
[font=arial,sans-serif]31.66%[/font]


Cue the Suspense Music: Here's the Shocker

What happens if we combine the data? What would happen if you sat out the best AND worst days? Here are the results:


[font=arial,sans-serif]If you missed the BEST and WORST:[/font]

[font=arial,sans-serif]0 days (this is buy-and-hold)[/font]
[font=arial,sans-serif]17.80%[/font]

[font=arial,sans-serif]10 days[/font]
[font=arial,sans-serif]20.31%[/font]

[font=arial,sans-serif]20 days[/font]
[font=arial,sans-serif]20.68%[/font]

[font=arial,sans-serif]30 days[/font]
[font=arial,sans-serif]20.80%[/font]

[font=arial,sans-serif]40 days[/font]
[font=arial,sans-serif]20.87%[/font]


This gives us a rather startling conclusion: If you miss an equal amount of the best and worst days, you outperform buy and hold regardless of how many days you miss!

Some food for thought...Of course, no one can predict the best or worst days in advance. But highly volatile days in the markets are often clustered closely together (think recently to early Nov 04, early Jan 05 and mid-Apr 05).

ST
 
imported post

ST,

Therefore, if you missed the 40 best days out of the estimated 3,750 trading days during the period 1988-1994 (easy) and avoided the 40 worst days (very hard), you could have averaged 3.07% a year over buy-and-hold. Correct?

Did DR Barton mentionwhat return you'd receive formissing all of the best days and hitting all of the worst days? I've always assumed that the 8.23% quoted for missing the 40 best days didn't depend on what you did on the other days, i.e. once you missed the 40 best days the maximum you could earn was 8.23%.

Regardless, you convinced me to stay buy-and-hold. I'm not that smart,lucky, or good looking. :D
 
imported post

To me, the point of it is that the "best days" argument doesn't hold any water. No one, not even the world's worst timer, will only miss the best 40 days and be fully invested the rest of the time. So this hypothetical return that subtracts the 40 best days is just plain silly.

Let's assume your market timing is random and you end up missing abouthalf the best days and half the worst days. According to this study (and several others I've seen), it won't hurt you and will probably help you. This is surely not enough to settle the timing vs. buy-and-debate, but it does debunk the "best days" argument in my book.

ST



rokid wrote:
ST,

Therefore, if you missed the 40 best days out of the estimated 3,750 trading days during the period 1988-1994 (easy) and avoided the 40 worst days (very hard), you could have averaged 3.07% a year over buy-and-hold. Correct?

Did DR Barton mentionwhat return you'd receive formissing all of the best days and hitting all of the worst days? I've always assumed that the 8.23% quoted for missing the 40 best days didn't depend on what you did on the other days, i.e. once you missed the 40 best days the maximum you could earn was 8.23%.

Regardless, you convinced me to stay buy-and-hold. I'm not that smart,lucky, or good looking. :D
Click to expand...
 
imported post

SystemTrader wrote:
To me, the point of it is that the "best days" argument doesn't hold any water. No one, not even the world's worst timer, will only miss the best 40 days and be fully invested the rest of the time. So this hypothetical return that subtracts the 40 best days is just plain silly.
Click to expand...
ST,

I've seen this argument made. However, I haven't been able to locate it in any of my references. Which study/article/book is DR Barton referencing? I'd like to review it.;)

 
imported post

rokid wrote:
ST,

I've seen this argument made. However, I haven't been able to locate it in any of my references. Which study/article/book is DR Barton referencing? I'd like to review it.;)

Click to expand...



Rokid,

I'm not sure where this idea originated, but it seems to propagated mostly by mutual fund companies and brokerage houses. That's not surprising, since buy-and-hold investing keeps their administrative fees down. I'm not sure if any serious analysts or academic finance types have used this argument.

Below are a couple of links. The first is an article by Franklin Templeton that uses this argument (near end of the article). It lists the Standard & Poor's (S&P) Corporation as the source of its study. The next is an article from Motley Fool that approaches this from a "value investing" angle, which is interesting since it'ssort of a third party perspective: neither market timing nor buy & hold.

Regards,

ST

http://www.franklintempleton.com/retail/jsp_cm/education/fund_basic/types/us_stock_funds.jsp

http://www.fool.com/news/commentary/2005/commentary05030404.htm
 
imported post

Update: Earlier, I said Iwould pull out of the F Fund for one day whenever the Non-Farm Payroll numbers are released. This wasn't intended to be a permanent strategy, justsomething to try during the next fewNFP releases, oruntilthe strategyquits working.(Since April of this year, the F Fund has taken a big hit during each NFP release.)

The NFP numbers will be released tomorrow morning. However, due tothe Philadelphia Fed's comments yesterday, which hinted at a possible change in interest rate policy, I've decided to stick with the F Fund. The Philly Fed based their comments on the Katrina tragedy and how it may impact the economy. In addition to this, the markets may not react to news and economic reports as they "normally" would during a time like this.

ST



SystemTrader wrote:
A heads-up: I'm going to add a little tweakwhenever I'm in theF Fund, as I am right now.

I've noticed the F Fund almost always gets slammed on days when theNon-Farm Payroll (NFP) numbers are announced. NFP numbersalways seem to be better than expected and this drives up bond yields temporarily.

So, whenever I'm in the F Fundand the monthly NFP dayis approaching, I'll allocate 100% to the G Fund for that one day, then switch back to the F. If I'd done this for the whole year, my return would probably be 1-2% higher. The modest gains I've made in the F Fund have pretty much been erased by NFP days so far.

Note: I'm not going to comment on the accuracy of theemployment numbers. There's been some discussion elsewhere on this board about the "hedonic" adjustments. Or you can read one person's view here:


http://www.[[financialsense.com/fsu/editorials/kirby/2005/0304.html
Click to expand...
 
imported post

New allocation: 50%C Fund & 50% I Fund

IFT date:22Sep 05

Moving from 100%G Fund to 50%C Fund and 50% I fund as of COBThursday,22Sep 05.

A few notes:

1) I've revised the system moderately, and itnow has a component that will sometimesbuy when the stock market dips. This doesn't happen too frequently, however, and will only happen if certain other conditions are in place. Overall, the system is still mostly trend-following in nature.

2) When the system is in the "buy stock funds" mode, I have a new way of determining which funds to purchase. Unlike the old method, which always held a 100% allocation, this new method always holds 2 of the funds in a 50/50 ratio.

ST
 
imported post

:DThank you System Trader! After a long time Bunkered down in G and F, I stepped out aggressively (40S 30C 30I) and it seems like everybody stepped back. At least there are afew of us left in the stock market, so I can feel a little validated.

I hope that this next hurricane isn't as bad as Katrina. Good luck out there.
 
imported post

Thanks! Several components of my system went from bearish to bullish last week, even before Rita touched down. Some people think we'll have a "relief rally" this week, and the overnight stock index futures are up strongly already. Should be an interesting week...



Citizen wrote:
:DThank you System Trader! After a long time Bunkered down in G and F, I stepped out aggressively (40S 30C 30I) and it seems like everybody stepped back. At least there are afew of us left in the stock market, so I can feel a little validated.

I hope that this next hurricane isn't as bad as Katrina. Good luck out there.
Click to expand...
 
imported post

In my Account folder, I posted a change to 100% F Fund yesterday.I just posted a note in there toignore that transaction, and thought I'd mention it here, too.The system is still in the same holdings until further notice: 50% S and 50% I Fund.

ST
 
imported post

No allocation change yet, but I'm expecting to be out of stocks soon. I'll update my allocations when and if this happens. The system is (barely) hanging on to its buy signal, but this could change very soon--maybe even after today's close.

There are several reasons for this. One is that larger, blue-chip stock indices such as the Dow and S&P 500 have started outperforming small-cap and technology indices like the NASDAQ and Russell 2000. This indicates that traders are looking to take on less risk, and often happens near the end of a rally. Also, as Tom mentioned the other day, commercial futures traders (i.e., the "big boy" institutional type traders) are becoming more bearish. They have to report their positions weekly to the CFTC:

http://www.cftc.gov/dea/futures/deacmesf.htm

We saw a very similar scenario last year around the same time. In addition, the S&P 500 chart is beginning tolookbearish. Notice how the Relative Stength Index (RSI), a measure of momentum, is now decreasing, while prices have inched higher recently without much conviction. This is known by chartists as "divergence." It doesn't always indicate a trend reversal, but it can, especially when other factors are also indicating bearish signs.

I'll have to end with ausual disclaimer: no charting tool or trading system is 100% accurate.I--and the system, if it gives a sell signal--could be wrong. But I have to put myself on the side of what's probable, and I think it could be an end (or at least breather) in the current rally.

z
 
That is my allocation. I do not trust the rally Friday. The S and I fund are at record highs. The C fund almost is and the F fund is doing lousy (as usual). Any thoughts on when to jump back in, what level of drop?
 
Status
Not open for further replies.

Similar threads

TommyIV
  • Article Article
TSP Talk Weekly Wrap Up - Opportunistic Investors
Replies
0
Views
100
TommyIV
TommyIV
tsptalk
  • Article Article
TSP Talk - An inauspicious start to the New Year
Replies
1
Views
168
nasa1974
nasa1974
tsptalk
  • Article Article
TSP Talk - So far so good for the holiday rally
Replies
1
Views
160
WorkFE
WorkFE
tsptalk
  • Article Article
TSP Talk - Stocks pull back and the Fed sounds less dovish
Replies
0
Views
152
tsptalk
tsptalk
TommyIV
  • Article Article
TSP Talk Weekly Wrap Up - Optimism with Caution
Replies
1
Views
360
tsptalk
tsptalk
Back
Top