A Target-Poor Environment

Screen Shot 2015-08-19 at 12.51.46 PM

I’ve owned stock in Target Corp. (TGT) for quite awhile. ¬† It’s a long-term play and so I only check it a couple of times per week. I noticed they beat earnings expectations today, and out of curiosity I took a look. The stock had opened¬† higher than yesterday’s close (about 4.5%) but then fell like a stone. “What a drag,” I thought, “for all those people who bought it at the open in the hopes of it going somewhere”.

That got me wondering, was this ‘failure to launch’ after an earnings beat a result of the current climate, or does this embarrassing dysfunction happen a lot?

I decided to take a look at the period from July 8th until today (08/19/15). I picked that start date because that was when Alcoa announced their earnings report, which is the unofficial start to the quarterly earnings season. I then also selected those time periods for 2013 and 2014.

I selected the current list of Russell 3000 members as my stock universe, with the historical price >$5 and the median 19-day volume greater than 100,000 shares. I looked at all the stocks whose opening price was 5% or greater than the previous day’s close. And I removed any stock that, in the previous 5 days, had their highest close more than 7% above its lowest close. This keeps out the stocks that are thrashing around because of some previous news. It’s not the same thing as selecting only stocks that bumped up because of earnings news. That’s too much time and effort for me – I don’t get paid for writing! So this will have to work as a stand-in.

Under the theory that you might have bought (or shorted) when you detected a stock opening more than 5% above the previous close, I recorded the gain/loss % over the same day, as well as the gain/loss over five days (counting the initial bump as day 1). I then sorted each set of data by the size of the initial bump, and divided them up into vigintiles.

Screen Shot 2015-08-19 at 12.25.11 PM

Ahhh, remember the heady days of 2013? Back then in the hot July and August sun, if your stock popped up over 8% or so, you were off and running. A positive expectancy for sure, especially if you held for five days instead of just one.

Screen Shot 2015-08-19 at 12.25.23 PM

Summer of 2014 wasn’t so hot however, at least from a market perspective. The upper percentages actually were more likely to fail, but over all just a noisy random assortment of returns.Screen Shot 2015-08-19 at 12.25.31 PM

And then there’s 2015. Lower bumps are actually showing a positive return, this time when the close-to-open bump is in the 5.3% to 6.5% range. Interestingly, the extremes are much wider for 2015. Whereas the other years showed a max/min of maybe +/-3% per vigintile, 2015 has some vigintiles exceeding +4 or -4.

Also interesting to note, across all three years, the lowest vigintile (i.e. right around a 5% bump) has a negative expectancy each time. An opportunity to short? Perhaps. I might want to take a detailed look at the 3-6% range in future.

So what does this all mean? Is it a sign that the ‘end times‘ are upon us? Possibly. But more likely, I think it tells us that a simple ‘earnings beat’ is not enough in itself to predict the short-term behavior or stock. Market conditions will affect whether investors will take profits at the earliest opportunity, or if they’re willing to risk a longer-term reward.

3 thoughts on “A Target-Poor Environment”

  1. So, you grouped the results by ‘initial bump’ (open today – close yesterday). Any interesting pattern observed if you group the results by 1) ‘reaction to initial bump’ (close of announcement day – open of announcement day) or 2) ‘decline/growth of initial bump’ (close of announcement day – close of pre-announcement day)?

    1. Looks that just like when programming, there is this 1 difficult part when posting comments on a website: naming things. “decline/growth of initial bump” is a bad name. But you got the idea.

      (There are only two hard things in Computer Science: cache invalidation and naming things. – Phil Karlton)

    2. Hi Adrian and thanks for your comment. Sorry for the delayed reply…wordpress wasn’t notifying me about comments from people who had posted before (only new commenters). Must fix that!

      I see what you’re asking here. Are there any secondary signals that could create a better chance of success? I will try and take a look at your suggestions, and post if I find anything interesting.

Leave a Reply

Your email address will not be published. Required fields are marked *