Down goes the Dow

I was curious about this so I plotted all SP500 data back to 1950, taking each day's index value as indexed to the start of the year. Here is a scatter plot of all data:

ibb.co

Slide32

Image Slide32 hosted in ImgBB

Open link

You can see there's an awful lot of scatter. If we want to see the "typical" behaviour, the following graph takes all that data and shows the average trend (black line in middle, plus the 10th and 90th percentile values that capture most (80%) of the range of the data:

ibb.co

Slide33

Image Slide33 hosted in ImgBB

Open link

We can take a closer look at the mean trend:

ibb.co

Slide34

Image Slide34 hosted in ImgBB

Open link

Not every year has a pattern that mimics the mean trend, but on average the markets are roughly flat from mid-July to end-October, with the biggest (and most of the) gains of the year happening in January to April and then November/December.

If I isolate calendar months we get the following average change:
January
1.31%
February
-0.02%
March
1.36%
April
1.43%
May
0.15%
June
1.42%
July
1.12%
August
-1.53%
September
-1.34%
October
0.93%
November
1.21%
December
2.47%

The best months are usually, in order, December, April, June, March, January, November and July. February, September and August on average show negative monthly change.

Of course there's an awful lot of variability, year to year, so I wouldn't count my late-year chickens before they are hatched.

https://x.com/hussmanjp/status/1692526172834251204?s=61

:-)

https://x.com/elerianm/status/1692267881827033294?s=61

Is it just me or does posting links that are visible on the thread only as: "X.COM" seem lazy and manipulative?

It doesn't read as anything besides a link - no indication of its content or message (unless you click on it).

What is the takeaway here? No comment, no discussion, just some poster thinking his reading list should be everybody's reading list?

And their are many of these each day. I'm not opening them.

Igy’s not too bright.

OK, now filtering either from X or LRC. Here is the content from Hussman:

“Total return of 30-year U.S. Treasury bonds. This is what happens when investors become convinced "there is no alternative" but to chase risk, at any price.
Banks and pensions own this stuff too.

You know I think equities are next.

I know you don't care.”

Thank you. And I do care.

Just a bit more on this...

As of end-July, SP500 was up ~ 20% YTD. If I look at all years since 1950, I can find 13 that make a reasonable comparison, having each been up > 16% by end-July (ranging from 16% to 28%). Of these 13 similar (to July) years, only 1 (7.7%) saw declines to the end of the year. That year saw a big swing, dropping from 28% up to dead flat on the year. If I include that reversion, the average change from end-July to end-December for those 13 similar years was up another 4.9%, which, if meaningful, would put this year up +25% from the 2023 opening bell to market close at the end of December. At any rate, if history is a reliable guide (it's not, but play along), we had, at the end of July, a conditional probability of ~ 93% to see the year-end at more than 20% gain on the year.

I have not looked at those 13 years to see how many of them had an August swoon, so buyer beware... :-)

Nice job, Comfortably retired. Interesting. How long did it take you to do all this?

I spent a couple / few hours on this. Your comments got me curious, and I've got time on my hands. :-)

What year was the negative year, and final yearly return?

I’ll try to remember to look up the year tomorrow. End-July it was up 28%, back to even (lost all mid-year gains) by the end of the year. That was the only year, of the 13 I selected (based on similarity end-July to this year, all up between 16-28%), that didn’t end the year higher than the end-July index value. Pretty big swoon that year though. And therein lies the hidden tail risk… most years, stocks go up. Some years they drop, and can drop a lot…

Of course, the calendar year is an arbitrary timeframe to focus on. Unless you’re looking to beat the DOW in 35 out of 32 years, I suppose… 🤭

There is no period since 1950 like the past fifteen years. We have experienced one of the best possible outcomes. The not so hidden tail risk may be the ringing out of excess.

1987. Up ~ 28% YTD at end-July, back to 0% at the end of December.

If anyone tries to check my math, I've done some data smoothing so one might find minor differences depending on how they decide to calculate things. For example, my baseline value for each year was the average closing value for the first week, rather than the first trading day, which was just a lot easier to work out in Excel without a lot of manual work or tricky formulas.

None of which means anything.

Kinda like your posts.

I suppose that depends on how you mean "like." If we look at 15-year gains to today, the SP500 has increased 295% over 15 years. If I look back through the data back to 1950 making the same calculation for every trading day (back to 1965, which gives me 15 years of prior data), the 15-year change has been more than 295% over half the time (52.7%), which doesn't really square with "no period" very well.

Maybe you mean something different? If so, please elaborate.

And if you meant to cherry-pick from the bottom of the post-housing crisis lows in 2009, or 14.1 years rather than 15 years, the 14.1-year gain to today is ~ 417%. Every day between 9 Sept 1994 and June 2008 had better 14.1-year gains than that.