SPY: Longer-Term Outlook

Time series Time!
As promised, I will finish up the year with a larger look of what 2023 and beyond may look like via time series. As well, I think its time to recap the year as a whole, because its been an interesting one, that’s for sure!
So let’s kick it off:

2022: The year of bear markets, history and disasters.

The chart above shows the Z-Score of SPY over its lifetime since its inception. The reason we do this is because z-score “standardizes” the data. For example, when a stock falls 50 % or 20% in, say, 1990, its not the same as a stock falling 50% or 20% in 2022. The reason being, we need to account for inflation, the stock’s value, etc. Z-Score removes the need to control for inflation by standardizing everything and representing it as a standard value that remains constant over time. This permits us to look at stock movement in terms of standard deviations from its mean.
In the chart above you will see boxes of various colours. They show previous bear markets. We have th tech bubble burst in Red, the housing bubble burst and 2008 financial crisis in purple, 2018 correction and COVID crash in green and finally the 2022 bear market in yellow.
Across all of these bear markets and crashes, SPY fell an average of 1 standard deviation away from its mean. The COVID crash caused the market to move a bit more, roughly just over 1 standard deviation. From the high of the year to the low of this year, 2022, the market made a historic 1.19 SD move! That is the most it has seen yet!

So why does this matter?

This matters because SPY is not following its historic behaviour. Previous bear markets/crises have not resulted in this kind of market move and this should concern us because it makes SPY’s trajectory slightly more difficult to predict.

Okay, but predict it anyway.

Well, the best way to do this is to plot the slop of SPY’s Z-Score. Doing that yields this:

We can see that immediate slope support currently rests around 1.5 SDs. That would correlate to a price of around 307 on SPY.
But we can actually be more precise and ask SAS to give us the equation for the Z-Score slope based on the trading days. If we do that, SAS gives the following equation:

y = 0.0004(Trading Days) – 1.485

Since we finished Friday on trading day 7533 we can solve as such:

Y = 0.0004 x 7533 Trading Days
= 3.0132 – 1.485
= 1.53 SDs

Or a SPY price of 310.62.

SPY closed 382.91, quite aways from its mean reversion slope. But if you believe in the powers of mean reversions and regressions, then the thesis here is that SPY will slowly work its what down until it makes contact with this mean.

Now this is a moving target, as the days pass and SPY moves up and down, so to does the mean linear line. So while right now 310.62 is the mean of this slope, that is not to say that 310.62 will be the mean of this slope in 6 months. As time passes, the mean increases and SPY will likely continue to fall until there is contact, i.e. a reversion to the mean.

What about Drifts from the mean?

However, we can make things more interesting by looking at, what I call, drifts from the mean. Regression and statistics accepts the reality of life. And what I mean by that, is it accounts for the fact that nothing follows math perfectly and we can actually account for drifts along the way from the mean.
We do this by plotting how far SPY normally travels away from its mean at any given time over its trading history. When we do that, we get a drift of 0.51 SDs. That means, despite the fact that currently, SPY’s mean Z-Score is 1.53, if we account for SPY’s max drift, the mean could be as high as 2.04 SDs.

And what price does that correlate with?

That is a price of 360.24 on SPY. And the area I would be immediately targeting in the coming months.

What about plotting the actual price?

So, while I plotted the Z-Score, we can do the same with actual price. And I have done it so that I can draw a comparison. I like doing it with both to validate that all the data is adding up and in agreement. The more agreement = the more persuasive.

When we plot the price out this is what we get:


The mean price of SPY currently is around 303, with a max drift of around 353. This is close enough to the Z-Score that I would deem them to be in agreement.

Timelines?

So when will it happen? That is the million dollar question. And unfortunately, it’s a question that really can’t be answered with the upmost confidence. Timing the market is really difficult, if not impossible, and its something not even the biggest, richest hedge funds or banks have been able to concur. However, according to the time series analysis, the move should be anticipated in the NEAR future.

What about the Bottom?

Mathematically, the bottom currently rests in the 280s.
How we get that is just subtract the drift from the average.
But as you see, SPY right now is slightly over its high drift. So there is no saying that SPY is going to stop right at its low drift and could in fact drop even lower. Its impossible to say for sure.

What you need to pay attention to is fundamentals. Sure, the bear market and SPY’s selling can be explained mathematically as a mean reversion, but you have to understand that there are always catalysts for these moves. Bubbles burst because of catalyst that cause the bubble to come crashing down. The 2008 financial crisis resulted from massive defaults on mortgages. Tech bubble resulted from massive bankruptcies of corporations. 2022 resulted from the abrupt halting of Quantitative Easing followed by rapid and successive rate hikes.

However, bubbles lead to said catalyst. It’s a “what came first, the chicken or the egg” type thing. Catalysts are the direct results of bubbles, bubbles catalysts. Its all just very philosophical and existential really.

What this means is, start re-investing when the fundamentals permit for it and when the market is trading around its mean. That's really all any investor can do.

My position:

Leading into next week, I will be starting to build a short position for a longer term hold. The reason being this analysis, but also, SPY has triggered my short que, which was dropping below the 75 DSL on the daily:


Its even more damming if you look at it on the weekly:


Last time this setup happened was 2008:

I will be scaling into SPXS and not options here, because, again, timing the market is really difficult!
And that’s it! That’s all I got for you!

Next Week:

I will be trading next week into Thursday. I will not be posting a SPY analysis leading into next week though. Why? Because everyone who matters in trading (i.e. hedge funds and firms) are all on Christmas vacay. The market is going to be range bound and do nothing interesting. Thus, swinging is not advisable next week and you can expect choppy markets. But I will be around to answer your questions if you have them!

That said, overall probabilities for Monday are looking fairly bullish, so expect an upward bias to be the predominate zest of next week.

Final Thoughts:

Its been a wild year! I can’t say I have enjoyed it. Its been stressful. But its been a learning curve for sure!
I am very happy to have shared it with you and all of the support, comments and feedback from everyone on TradingView! I will be continuing to post into next year and nothing is changing in that regard, but when I started posting on TradingView (initially as a way of journaling), I never anticipated the slew of interest and participation from this awesome community! And I thank all of you for your support, it truly is appreciated!

I wish everyone the best going into next year and safe and green days/years/months ahead! And I hope to spend 2023 with you all again. Let’s make it a great year!