Transcript of THE $163B ALGO SELLOFF: The Secret Market Level That Triggers the Machines!
Wall Street Truthbombs
0:00Bank of America's own research desk put0:02a number on how much stock the machines0:04would dump if this market cracks, $1630:08billion0:10in one week. By the end of this video,0:12you're going to understand where that0:13number actually came from, why the0:15version going around the internet is 30:17weeks out of date, and the one piece of0:20it that nobody's actually printing,0:22which is the exact level where the0:24selling really starts.0:27My friends, I was looking out the window0:29around 4:00 the other day and the block0:31below was completely jammed. Four0:34delivery trucks double parked without a0:37within about 60 ft of each other.0:40Different companies, different drivers,0:42none of them talking to each other, and0:44they all arrived at the same moment on0:46the same block because they are all0:49routed by software that reads the same0:51traffic data and reaches the same0:54conclusion. Tell me that hasn't happened0:57to you. Nobody coordinated that. Now,1:00there was no conspiracy there. There1:02were just four independent actors1:04following nearly identical rules, and1:08identical rules produce, well,1:10synchronized behavior, good and bad.1:13Hold that picture in your mind. It is1:15the whole episode. Okay.1:18So, let us start with what is actually1:20true. That's a good place to start,1:22right? Because there is real news here1:25and then there is a number that, well,1:27got welded onto it. The real news is1:30that Bank of America stock fell about1:325.14%1:34on Monday on almost 61 million shares.1:38It closed at 59 bucks and 47 cents.1:41That's a big move for a bank that size,1:43and it wasn't about algorithms at all.1:46It was about Brian Moynihan opening his1:49mouth at the Barclays Financial Services1:51conference. He guided investment banking1:54fees to between 1.6 and 1.8 billion1:58dollars for this quarter. A year ago,2:00that line did about 2 billion. Run that2:04arithmetic and that is down somewhere2:06between 10% and 20% year-over-year. I2:08did the math so you don't have to. These2:10were his own words and this is where I'm2:12going to quote him. What we're seeing is2:15the market generally in investment2:17banking is down 10% or so. Trading he2:20called roughly flat and the whole sector2:22went with him. Goldman down almost 4%,2:26City, Wells, and JP Morgan all down2:28between 1% and 2%. The bank index was2:32off 2.7%2:34and that's the news. Now, here is a2:36number that got bolted onto what2:39Moynihan actually said. I'm going to get2:42into it. Before we do, please click2:44like. Don't forget to subscribe. It's2:45important to know about how this stuff2:47works. This is exactly how you do it.2:49Okay.2:51Bank of America publishes a weekly2:53research note called the Systematic2:55Flows Monitor. It's written by their2:57derivatives and quant team um and uh3:01not by the strategists whose names keep3:04getting attached to it on the internet.3:07It's a completely different thing. I was3:08going to name the names, but you don't3:10need to know the names. Different note,3:12different desk. What it does is model3:15the mechanical funds, okay? Commodity3:17trading advisors that follow trends,3:20volatility control funds that must sell3:23when volatility rises. Risk parity.3:25These aren't people making judgment3:27calls. They're rule books and because3:30the rule books are public and similar,3:33you can estimate what they're going to3:34do. The $163 billion figure is very3:38real. It's from the August 28th edition3:42and it's one week conditional estimate,3:44meaning if the market falls over the3:46following week, this is roughly what the3:49machines would sell based on those rules3:52that I kind of pointed out to you just3:55before. Now, here is the part that3:57should actually bother you a little bit.3:59That is not the current number. The4:01September 11th edition, the most recent4:04one, says down market minus $1574:08billion. But, in a flat market, the4:10machines would buy 20 billion. And in up4:14market, they would buy 15 billion. So,4:17the line going around that is that the4:20algorithms have only $9 left to buy.4:24That's the August 28th number. The4:26current figure is 15 to 20. The4:29imbalance got better, not worse. And4:33now, look at the series because this is4:35a weekly publication, and the weekly4:38numbers move around a lot. 77 billion in4:41early May, about 100 in late May, 114 in4:46late August, and 163.4:49Then, 162. Now, 167.4:53163 is the highest print in the series.4:56Okay, I know this is complicated. Keep4:58stick with me, though. Which is exactly5:00why that is the one that went viral.5:03Yes. And here's the part that should5:06really settle you down. The same warning5:08went out on May 9th with the S&P at5:117730.5:13Today, it's trading roughly around, I5:15don't know, 76 or and change. 4 months5:19later and about 1 and 1/2% lower. No5:22cascade, no crash, also no melt up,5:26either. These warnings haven't been5:28wrong, exactly. They've been non-events.5:30So, now let me give you the thing that5:32nobody actually printed. And this is why5:35you watch me at Wall Street Truth Bombs.5:38If you work long hours, like you know I5:39do, you know how important it is to get5:41a good night's sleep. Working on Wall5:44Street, you guys know that my days start5:46really, really early and they end late.5:49My brain doesn't always get the message5:51that it's time to shut down. That's why5:54the SomniPods 3 by FitNexa have made5:57such a difference for me. I've been5:59using them at night and they've6:00dramatically improved my ability to get6:03long, quality sleep. And believe me, my6:05friends, I really need it. They're6:07specifically designed for sleeping.6:09They're ultra-thin, you can see that6:11right here, and they're incredibly6:13lightweight and comfortable even when6:15I'm sleeping on my side or even when I6:18fall asleep right here at the desk. And6:20with hybrid active noise cancellation up6:23to 24 decibels, they help block out the6:25little noises that can wake you up6:27throughout the night. You know, in the6:29city, we have lots of those. I can also6:32choose from built-in sleep sounds like6:34white noise, rain, ocean waves, and6:36other relaxing soundscapes without6:38needing another subscription. And I use6:41every one of them. I like to switch6:43between them. But, one of my favorite6:45features is the sleep tracking.6:47SomniPods can monitor things like when6:49you fall asleep, when you wake up, your6:51sleep stages, movement, and even your6:54sleeping position. Their AI sleep coach6:56then uses that information to provide6:58personal recommendations designed to7:01help improve your sleep routine. And7:04when you have to get up early like I do,7:06there's even a private alarm that plays7:08directly through the earbuds so you can7:11wake up without waking up your partner.7:14Or in my case, my partner and my dog7:17Eloise. For me, better sleep means being7:19sharper and more focused when the7:21markets open the next morning. [music]7:23And that is critical. And when you work7:26the hours that I do, that matters.7:28Really, it does. So, if you're looking7:30to improve your nighttime routine, check7:32out the SomniPods 3 by FitNexa. You're7:35going to thank me.7:38It was buried in that September 11th7:40note is the trigger band. The model says7:44the selling starts to fire up when the7:46S&P futures gets down into the range of7:497349 to 7558.7:52With the index around 7620, that's7:55roughly a 0.8% decline to reach the top7:59of the band and about 3 and 1/2% to8:02reach the bottom of it. Now, read what8:05that actually means. This isn't a trap.8:07A trap is something that's hidden. This8:10is a stop loss ladder with the rungs8:12printed right on it. The machines have8:15published where they start selling. That8:17is less frightening than the headline8:19and far more useful because now you're8:22not afraid of a ghost. You're actually8:24watching a level. And let me explain why8:27these funds are forced to sell because8:30this is the piece that makes the whole8:32thing click.8:34Take a volatility control fund. It's8:36mandate says hold fixed level of risk.8:40Not a fixed amount of stock, a fixed8:42amount of risk. So when the market gets8:45choppier, the same pile of stock now8:47carries more risk than the mandate8:50allows and the fund must now shrink the8:53pile. It's not an option It's not an8:55opinion. Nobody calls a meeting. The8:57spreadsheet says sell and somebody9:00sells. A trend follower works the same9:02way from the other direction. It's long9:05because the trend is up. Break the trend9:07line and the rule flips. Again, no9:10judgement, just a threshold. Now, put9:13those two together with my four delivery9:15trucks that I talked about earlier. None9:17of these funds talk to each other,9:19either. They just read the same price9:22and run nearly the same rule. So they9:24arrive at the exit at the same moment.9:27That's the real mechanism and it's not9:29sinister, it's just crowded. That's a9:32Wall Street term and I owe you three9:34honest things before I go any further.9:37first, the most quoted flow strategist9:40on Wall Street, his name is Scott Rubner9:42at Citadel Securities.9:44He says the opposite of the headline.9:47His words, quote, "Positioning is not9:49stretched." He does not add the market9:52no longer has the same reservoir of9:54unused systematic buying capacity it had9:57after the July reset, and he says he9:59would rather reduce exposure and buy10:01cheap protection than chase this market10:04higher. So, he's cautious, but he's not10:08saying the machines are maxed out, and10:10he's the one who would know. Second,10:13Bank of America is an outlier on the10:15size of this. Goldman's comparable10:17estimate earlier this year was over a10:19hundred billion of selling but over a10:22month. Bank of America is over a week.10:25That's roughly six times more aggressive10:27per unit of time. Don't let anyone tell10:29you the banks agree on this. They don't.10:32Third, this is the one that will annoy10:35you guys, trust me. The buy up The10:37buyback blackout story is weaker than it10:39sounds. The mechanic is real. Companies10:42go quiet before earnings and stop buying10:44their own stock. That demand has been10:47estimated at roughly five billion10:49dollars a day when the window is open.10:51That is Rubner's number at Citadel10:53Securities, and he says the blackout10:56started to accelerate around September10:5812th. So, that bid has been thinning out11:01recently. But, State Street actually ran11:03the regressions on whether the blackout11:05window hurt markets. Raw buyback dollars11:08correlate about 77% or so11:12with the index, which sounds pretty11:14damning, but when you normalize buybacks11:16as a share of market cap, which is the11:18only honest way to do it, the11:20correlation actually collapses to about11:235%, and their tests across and excuse11:26me, and their tests across before,11:29during, and after earnings windows came11:31back with no statistical significance at11:33all. In other words,11:35throw them out, okay? Tear them up and11:37throw them in the garbage. Uh there11:39that's their phrase for it. Uh well,11:41they called it noise. I just told you11:42what to do with it, right? Because yeah,11:44I sat through that PhD level class11:46myself. I also want to correct something11:49that is everywhere. You're going to hear11:51that corporate buybacks are the largest11:52marginal buyer of American stocks. Uh11:55what Goldman actually wrote is that with11:57roughly 1.4 trillion of buybacks and12:00about 700 billion of new share issuance,12:03excuse me, new share issuance, corporate12:05America should remain a uh significant12:07net source of demand. A significant12:10source, not the largest. And the same12:12firm expects households to lead equity12:14demand this year. So, where does all12:17this leave you? All this noise I just12:19sent your way. It leaves you with a12:21market about 2 and 8/10% off its record12:26high with volatility in the mid-teens.12:29There is no sell-off to be trapped in.12:31The trap has not sprung because nothing12:33has stepped onto the trap. What would12:36step on it is a catalyst big enough to12:39push the index through that 7558 that I12:42said earlier. And there was one two days12:45ago at 2:00 when the Federal Reserve12:47announced a decision that futures had12:49been pricing at better than 2012:52than 90% odds of a rate hike. That's the12:55real setup, not algorithms lurking in12:57the dark. A known level, a known date,12:59and a crowd that has already decided13:02what it thinks is coming. The dangerous13:04trade this week is not the one where the13:06Fed surprises you. It's the one where13:08the Fed does exactly what everyone13:10expects and the market sells off anyway13:11because when you find out the machines13:14were never reacting to the Fed at all.13:16That's what you got to pay attention to.13:18So, your truth bomb for today is this.13:20The $163 billion number is not a13:23prediction. It's not even a the one. It13:26is a stop-loss ladder with the rungs13:28printed on it at 734913:31to 7558.13:33And the only thing that makes a13:34published level dangerous is everybody13:37knowing that it's there.13:39That was complicated. I hope you picked13:41up at least a little of it. Join me13:42every day for Wall Street Truth Bombs13:44where I drop them right here before the13:45market figures it out.
2,242 words · 334 lines
