NEW YORK (Real Money) -- We all want to think that we can spot bottoms. We all want to think that bottoms aren't lockstep affairs but, rather, movements that take time to build. You get some confidence here, some short-covering there, some positive news from the left and some upgrades from the right.
But none of that happened yesterday. Nope, it was all one gigantic two-stepped algorithm, and we should just suspend any judgment that there was more to it than that.
The first step? At 12:58 p.m. EDT the NYSE Arca Biotech Index, which had been in slow-motion-crash mode, made a sudden pivot and rallied convincingly. Don't bother looking, as nothing happened at that moment -- nothing at all. I checked the headlines of every single major biotech stock before I wrote this. There wasn't a single story of any import.
Then, at almost exactly 1:02 p.m., the Nasdaq Composite and the S&P 500 both bottomed simultaneously, seemingly within a few seconds of each other. No doubt that's a lifetime in the algo high-speed world, but I don't think our eyes could discern much difference.
After that, it was off to the races and the market never looked back.
Now, as I scanned the headlines for anything that could have caused that pivot, I couldn't find anything of true import.
However, if you examine the trading in the iShares Barclays 20+ Year Treasury Bond (TLT), you'll see it peaked at 1:01 p.m., approximately one minute before the S&P and the Nasdaq bottomed.
So, the sequence: At 12:58 p.m. the biotech index stops going down. At 1:01 p.m., bonds peak for the day. At 1:02 p.m., the S&P and the Nasdaq pivot and roar.
There can be no denying that sequence. It's right there in the numbers.
After that, there's one thing that sticks out among a sea of nothingness: a Twitter (TWTR) post at 1:10 p.m. from Daniel Graf, a man acknowledged as one of the geniuses behind Google's (GOOG) Maps. It read simply, "Followed Maps to find that the flock was just around the corner - excited to take wing with the @twitter product team." Or, in non-Twitter English, a real hot-shot developer had just left the so-called hottest company in the space to become the vice president of the consumer product group at Twitter.
At that point Twitter shares accelerated their gains more quickly than did all of the other social, mobile and cloud-computing stocks -- which had already been rallying. Twitter never looked back, traveling from $41.14 at 1:11 a.m. straight up to $45.52 at the close -- the most vicious of all the ralliers from the ne'er-do-well crowd of both tech and biotech.
I only point this one out because Twitter had been the weakest in the cohort. It was certainly the most "shortable" name -- because, until Monday, we had thought that billions of dollars in insider selling would be coming in Twitter. We learned that holders responsible for the vast plurality of stock would not be selling any time soon, and when that happened, it took away the potential supply that would be needed to handle the massive short position that had been developing.
Do not minimize the significance of that announcement. Forty-five million shares, or 40% of the current float of Twitter, are sold short. The non-sellers, co-founders Evan Williams and Jack Dorsey, control 9.4% and 4%, while Dick Costolo owns 1.4%. Benchmark, the key venture capitalist, has a 5.4% stake. Together that's 20.2% of the locked-up stock. You also have to believe that those four shareholders are powerful enough to intimidate others not to sell. That means the short position is way too big for the float on any good news, like an acceleration in sales or earnings -- which are distinct possibilities.
Given the vociferous way this stock traded after the Graf announcement, lots of people were led to believe that we had just experienced a Twitter-led rally -- that the market had a new leader that had previously been the biggest decliner of this ugly era, a stock that had decline 30% for the year. People thought that this rally signaled, "Enough is enough." But I think we saw one of those "enough is enough" rallies just last week, so I am not going to be so sanguine.
Nonetheless, this move at least felt as if it had a bit of news associated with it. It also coincided with the possibility that interest rates on the 10-year Treasury may have troughed at about 2.5%.
In any case, let's go back to that bottom in the stock market. This much is clear from the sequence: Two minutes before 1 p.m., forced selling ended, regardless of the price level.
What happened at that time? That was really the last safe moment you had to beat the margin clerks who would otherwise margin you out or sell your stuff for you at any price they wanted, so they'd be able to ensure that you wouldn't stiff them.
Now, it isn't clear how the bond peak fits in. It's possible the shorts and high-frequency traders all know the sellers' game plans. That's easy enough to discern. You probably wouldn't get as strong a pivot at that hour off the margin action without the bonds.
Maybe the whole thing is coincidence -- although you would have to be a complete moron to believe that.
So if we put it all together, we can conclude several things. First, the damaged flailing sellers were being margined out first in their biotechs, and then in their techs. Second, the buyers knew this, and they also were attuned to the peak in bond prices.
Third, money flowed out of stocks and into bonds until two minutes after the bottom in the biotechs. Fourth, that whole process reversed at the bottom in the Nasdaq and the S&P. Fifth, everything accelerated to the upside when the Twitter news broke.
Or in other words, the whole thing was dreadfully artificial and entirely done by a combination of "break the sellers" margin call and short-selling and HTF pressure, no doubt encouraged by the rally in the bond proxy TLT. Once the sellers were broken, the process was free to reverse.
Now we have no idea if the sellers are done. They tend to come in when the market lifts, as soon as they think there's enough cushion. Again, that's typical of the way margin clerks work. They want to try letting the bids build before they slice through them. But once the process starts, it doesn't stop until 1 p.m.
My conclusion: If you think for even one second that this selloff has been motivated by earnings shortfalls, think again. That's not just because all the major stocks that have reported, save JPMorgan Chase (JPM), have announced much-better-than-expected earnings.
It's all artifice.
That's why it is so darned hard to make a judgment as to whether the selling, at last, is complete.
At the time of publication, Action Alerts PLUS, which Cramer co-manages as a charitable trust, was long GOOG and JPM.
Editor's Note: This article was originally published at 8:02 a.m. EDT on Real Money on April 16.