The Jordan Premium
Why the market cannot price a drug-discovery engine, and what Arrowhead's is really worth. A guide to valuing the machine, not just the medicines it has made so far.
Robert Toczycki, JD, MBA
bioboyscout.com
bioboyscout@gmail.com
847.227.7909
X: @BioBoyScout
I have written about Arrowhead’s engine before, in The Execution Engine, in Zero, in the platform pieces, and each time I tried to value it carefully and kept the numbers conservative. I now think I undervalued it every single time. Not because the analysis was wrong, but because the engine keeps proving itself faster, and in more places in the body, than any careful model would dare to assume in advance. Every quarter Arrowhead adds another target, another tissue, another program that made the jump from animals to humans, and every quarter it reminds me that the thing worth valuing was never the drugs on the board. It is the machine that keeps putting them there. This paper is my attempt to fix that mistake, to value the engine itself instead of its output, and to explain why the market has had the same blind spot I did.
This paper makes one argument, then tests it. The argument is simple: the market knows how to model a list of drugs, but it has no explicit way to model the machine that produces them. Any company whose real asset is a proven, repeatable discovery engine will therefore tend to be undervalued by output-based models, in a way that is not a mistake anyone makes on purpose but a blind spot built into the math. Arrowhead is the clearest living example. The first half of what follows builds the idea on neutral ground, using nothing about Arrowhead at all. The second half holds Arrowhead up against it.
1. A list of drugs is not an engine
Start with a distinction most investors skip past. A pipeline is a list of drugs. You can name each one, count them, guess each one’s odds and market size, and add it all up. Analysts are good at this, and it is exactly what a sum-of-the-parts model does, SOTP for short: put a value on every drug on the board, total it, and call that the company.
An engine is a different animal. It is not a drug. It is the thing that makes drugs. Its real output is a stream of future programs, most of which do not exist yet and have no name today. What makes an engine valuable is not any one medicine it has produced but its proven ability to keep producing them, in new tissues, against new targets, year after year. The value is in the repeatability, not in any single result.
The market has good tools for the first thing and much weaker ones for the second. Markets price intangible things all the time, through multiples, sentiment, and the premium paid for a good story. What they cannot do is model an engine explicitly. When a company’s main asset is the engine rather than the list, the standard model has nowhere to put it, and whatever credit the engine receives arrives indirectly, as a multiple nudged by mood rather than a number anyone has built. The result is not that the engine is valued at zero. The result is that it is valued by accident.
2. Why the math cannot see the engine
The reason is mechanical, not a knock on any analyst, and it is worth walking through slowly because it is the heart of the paper.
A sum-of-the-parts model values things it can name, with odds and markets it can defend. That is its strength. It is also its cage. An engine’s value lives in drugs that have not been invented yet. They have no name, no trial, no disease, no odds you could write in a footnote and defend to a skeptic. A careful analyst therefore cannot put a line in the model for a drug that does not exist, and the value of the ability to create that drug quietly rounds to zero. Not because anyone decided it was worthless. Because there is no row to write it on.
This is the key idea. The engine is not underpriced because the Street is lazy. It is underpriced because the ruler cannot measure it. A model that paid in advance for drugs it could not yet name would be a fantasy, not a model. The very discipline that makes SOTP trustworthy is the same discipline that blinds it to the engine.
The chain of reasoning is worth keeping in this order, because each link is much harder to argue with than a blanket claim about market blindness. An SOTP model cannot explicitly value assets that do not yet have names. Public markets lean heavily on that method for companies like this one, so they systematically underprice the engine rather than ignore it outright. A strategic buyer, whose whole purpose is to own the future output, can explicitly value the engine’s expected productivity. That is the argument. Companies whose worth is mostly the visible list of drugs can at least be modeled explicitly. Whether the market gets those models right is a separate question, and often it does not. Companies whose worth is mostly the engine get priced on a placeholder plus whatever sentiment supplies, and the more of the value that sits in the machine instead of the list, the wider the gap between the price and the truth.
3. The Jordan Premium
There is a version of this you already understand, from a completely different world, and it makes the whole thing click.
When a team drafts a generational talent, it is not paying for next season’s points. Next season’s points you can estimate, and a decent role player who scored the same total would cost a fraction as much. What the team is paying for is the near-certainty of greatness it cannot spell out yet, year after year, in forms nobody can predict. The extra money, the amount over and above the projected stat line, is the price of proven repeatability. It is rational precisely because the greatness has already been shown, not just hoped for.
Call that gap the Jordan Premium: the difference between what an engine is worth and what its current, visible output is worth. It is the value of the machine on top of the value of what the machine has made so far. For a generational athlete, the premium is huge and the market pays it without blinking, because sports learned long ago to price proven repeatability. The stock market, handed a company whose real asset is a repeatable discovery engine, has not learned the same lesson. It pays for the box score and hands you the player for free.
One rule keeps this honest, and the rest of the paper leans on it. The premium is only earned when the repeatability is proven, not just promised. Most prospects scouts call can’t-miss do miss. The reason a true generational talent earns the premium is a real track record, not the hype in the scouting report. For a company, it is the same. The premium is earned by evidence, and by nothing else. A story about a platform is not an engine. The next section is how you tell them apart.
4. How to spot a real engine
Most companies that call themselves platforms are, when you look closely, one good drug wearing a costume. The jump from animal studies to human results is hit-or-miss, every new program is a fresh gamble, and the platform talk is marketing draped over what is really a short list. The real job, then, is telling a genuine engine from a good story, and it can be made concrete. A real engine leaves four marks you can actually check.
It translates. Does what works in the lab reliably work in people? A real engine turns targets into human-stage drugs at a rate well above the industry average, and it does it again and again, not once. This is the hardest mark to fake, because it shows up in the clinic, where stories cannot follow.
It generalizes. Does the same core trick work in more than one place? One drug in one tissue is an asset. The same technology working in tissue after tissue is proof that the underlying ability is general, not a fluke. Each new tissue, moreover, is not just one more market. It is a multiplier, because every tissue holds dozens of new targets to aim at.
It is efficient. Does it make more per dollar than its rivals? A better engine turns the same research budget into more drugs than the competition. That is the number-based fingerprint of a superior machine, and you can check it against public spending.
It keeps a beat. Does the output arrive on a schedule, or in random bursts? An engine produces. A story goes quiet between press releases. A steady, rising count of new drugs entering trials, year after year, is the mark of a process rather than a lucky streak.
A company that scores high on all four has earned the Jordan Premium, and the next question is whether the market has paid it yet. A company that scores high on one or two has a good drug and a good story, which is common and much cheaper. These four marks have nothing to do with any particular company yet. They are just the measuring stick. Now it is fair to pick a company up and hold it against the stick.
5. How rarely anything passes this test
The four marks are common one at a time and rare together, and that gap is the entire reason a premium exists. If proven repeatability were ordinary, it would be priced the way ordinary things are priced. Plenty of companies translate well inside a single modality. Plenty run cheaply, usually because they are small rather than because they are productive. Some keep a decent cadence for a while. What almost none of them do is generalize, meaning take the same core capability into domain after domain, which is the mark that separates a machine from one good drug with a story attached.
Set the bar at all four marks together, held over years, and the historical list gets short fast. Genentech’s recombinant protein method generalized from insulin to growth hormone to a clot-dissolving drug, and for years the market bought the products rather than the method. Regeneron’s antibody-generation platform produced a cholesterol drug, then an immunology drug that grew into one of the largest in the industry, then an oncology franchise, and a string of others behind them. Through much of that stretch the stock was discussed primarily as a bet on the company’s eye franchise, even while the antibody engine was quietly producing assets behind it. Alnylam’s liver-targeting approach generalized across a series of approved medicines, and it too was talked about as a rare-disease company well after it had become a delivery platform. In each case the engine is obvious in hindsight and was priced late.
Moderna is the case worth studying most closely, because there the frontier fell all at once. The generality of the platform was the entire thesis, and the market paid something for that promise. A pandemic then supplied the demonstration, and the repricing arrived in months rather than years. The sequel is just as instructive. When the demonstrated use case faded, much of the premium went with it. A Jordan Premium can be awarded and then withdrawn, which is a useful corrective for anyone inclined to read this paper as a one-way argument.
Now the warning that belongs on that list, because leaving it off would be the same error this paper keeps trying to avoid. Every company named above is an engine that worked, which is precisely why anyone remembers it. For each of them there were companies with equally confident platform stories whose generalization never arrived, and they are missing from the list because they failed and were forgotten. The base rate of platform claims in biotech is enormous. The base rate of platform claims that turn out to be engines is small. That asymmetry is not a reason to discard the framework. It is the reason the framework needs a test at all, and it is why the four marks insist on demonstrated behavior rather than stated ambition.
The honest scale, then, is a few per generation rather than a few per year. Note also where Arrowhead sits relative to that list, which is not on it. The companies above had already demonstrated substantial generalization by the time their platform value became impossible for the market to ignore. Arrowhead is a candidate standing at the moment before its defining test, which is a less comfortable position than membership, and also the only position in which the premium has not already been priced.
6. Arrowhead against the stick
Run the four marks in order.
It translates. Arrowhead’s own line is that just about everything it has taken into the clinic has made the jump from animal models to humans. Treat that as a claim to be checked, not swallowed, and it is still a strong one, because a false claim like that gets exposed fast, failed translation shows up as failed trials for everyone to see. A translation rate near the top of the industry, held across many programs, is the first and hardest mark, and Arrowhead appears to carry it.
It generalizes. Here the record is plain. The same delivery technology has reached the liver, then the lung, then muscle, then fat, and the brain program is in the clinic now, with its first data expected shortly. That is not one trick used once. It is one trick that keeps working in a new place each time, and each new place widens the field of targets by a large multiple. Generalization is the mark that most cleanly separates a real engine from a single lucky drug.
Arrowhead has taken this one into the clinic across more tissues than any other RNAi company. Competitors are working on the same frontier, and Alnylam in particular has clinical programs aimed at the central nervous system and adipose tissue, so this is a lead rather than a monopoly. On that record, Arrowhead has a strong claim to being the leading extrahepatic siRNA developer, which is the more useful way to state it: every approved siRNA drug to date targets the liver, and Arrowhead has gone furthest in leaving it.
It is efficient. The numbers were the subject of an earlier paper and they are blunt. Arrowhead has advanced more clinical programs per research dollar than either Alnylam or Ionis, its two closest peers, spending roughly $607 million on research and development in fiscal 2025 against their $1 billion and $916 million.
That comparison is rough, and it is worth saying why rather than leaving a skeptic to say it. Both peers carry heavier late-stage and commercial spending, and a single Phase 3 program can consume more cash than several discovery programs combined, so part of the gap reflects where each company sits in its life cycle rather than how good its engine is. Fiscal periods do not line up perfectly either. The gap is wide enough that the direction survives those adjustments. That is the claim being made here: not that the ratio is precise, but that more output per dollar is the number-based signature of a better machine, measurable rather than asserted.
It keeps a beat. The count tells the story. The clinical portfolio has climbed steadily toward a guided twenty-three drugs, owned and partnered, entering trials on a schedule rather than in fits and starts. There is a concrete reason the beat holds, and it is worth knowing, because it is the least glamorous item in this paper and one of the most important. Arrowhead built its own manufacturing plant. RNAi timelines across the industry have often been gated by waiting for batch slots at contract manufacturers, which means a rival’s program can sit idle for reasons that have nothing to do with its science. Owning the plant takes that queue out of the schedule. That is production, not luck. Arrowhead scores high on all four marks at once, which, measured against the record in the previous section, is the unusual case rather than the normal one.
7. The tell that the engine is still accelerating
Here is the part that should stop an investor cold, because it turns the engine from a story about the past into a story about the future.
Arrowhead is not done adding tissues. Behind the brain, two more are already in the works at the preclinical stage: the eye and the heart. The eye program aims at glaucoma, delivered by a small injection into the eye itself. The heart program, aimed at the heart muscle cell, the cardiomyocyte, is the one to sit up for, and it was not even on the company’s public pipeline slides until recently.
Reaching heart muscle with a gene-silencing drug is close to a holy grail. Heart muscle has been one of the great locked rooms in medicine, enormous numbers of patients, and until now almost no way to switch off a disease-causing gene inside those cells. Cardiovascular disease is the largest cause of death in the world. A delivery engine that reaches the heart muscle would not open one more drug. It would open a whole new front in the biggest disease category there is, at the genetic root. Two words of restraint belong on that sentence: this program is preclinical, years from a human test, and it may not work. What it demonstrates today is not a heart drug. It is the engine picking its next locked room.
Count the tissues: liver, lung, muscle, fat, brain, eye, heart. Seven, and the last two were added while nobody was pricing the first five. That is not a company with a platform. That is a machine that reaches a new part of the body on a schedule.
Here is the real point, the one that matters most and gets priced least. The named tissues are not the prize. The prize is the increasingly credible expectation that the engine keeps reaching new ones. A machine that has gone from the liver to the lung to muscle to fat to the brain in a decade, and is now engineering toward the eye and the heart, gives no good reason to assume seven is the limit. The reasonable expectation for an engine with that record is that it reaches an eighth tissue, and a ninth, in places nobody has named yet. Those unnamed tissues, the ones not on any slide, not in any model, not in anyone’s price target, are where the largest and least-priced value of all is hiding. You cannot write them into a spreadsheet. That is precisely why they are free.
8. What the market pays for all of this today
The framework predicts that conventional asset-by-asset valuation will pay almost nothing explicitly for a discovery engine, however good, until evidence forces that engine into the model. Arrowhead gives you the cleanest proof of that prediction you could ask for, and the number is worth saying out loud.
In the sum-of-the-parts breakdown JPMorgan published with its $95 target, the line covering the brain platform, the central nervous system programs and the delivery technology behind them, was carried at $1 a share. Not a billion. One dollar. JPMorgan has since raised its target to $100, on a quarter driven by commercial and cardiometabolic progress rather than any human brain data, so there is no reason to think the extra value reflects a new credit for the platform. Most of what any sane person would call the engine, the proven, generalizing, efficient, steady machine described above, sits in that model at a placeholder next to zero.
The preclinical tissues, the eye and the heart, sit inside the same platform line, which is to say they are carried at very close to nothing at all.
The stock does not trade at exactly the sum of those parts, and sentiment supplies some credit the model omits. The point is narrower and sturdier: the one part of the company that most deserves an explicit valuation is the one part no analyst has a row for.
The $1 is not an insult to the engine. It is the math doing exactly what the math does. An SOTP model cannot explicitly value a discovery engine, so it parks it at a placeholder, and given the method, the placeholder is honest. Arrowhead is not the exception that breaks the framework. It is the framework’s clearest proof.
9. What the premium is worth
The fair question is how big the premium should be, and the honest answer is a range with its assumptions visible, not a single number pretending to a precision it cannot have.
The premium is the gap between the engine-inclusive value of Arrowhead and its SOTP value. Its lowest layer is visible, and its highest layer is not. The lowest layer is revealed strategic value: what buyers have already shown they will pay for the engine’s output. Novartis put up $200 million up front and committed to as much as $2 billion in milestones for one program and a few targets built on this platform, before a shred of human brain data existed. In just the most recent quarter, Madrigal paid $25 million up front and committed to up to $975 million for a single clinical-stage liver program. Read those numbers precisely. The upfronts are cash that changed hands. The milestone figures are contingent, payable only if the programs succeed, so they are not money in the bank, but they do establish that sophisticated counterparties were willing to contract around substantial future economics for slices of this engine. That is observable evidence in signed contracts, not hypothetical value invented by a model.
The ceiling is the value of all the drugs and tissues the engine has not reached yet, including the ones with no name. It cannot be itemized, which is the exact reason the market leaves it out, and it is also potentially the biggest part of the premium. It is the value of the machine’s future, the same thing a team pays for when it drafts a generational talent instead of a year of projected points. No honest analyst can type it into a model. That same analyst can nonetheless see that its absence from the model is the mispricing, not a verdict on its worth.
Between that lowest layer and that ceiling the premium separates further, and naming them is more useful than pretending the whole can be reduced to a single figure. The bottom layer is revealed strategic value: the cash paid and the milestone economics sophisticated counterparties have already committed to for slices of the engine. Above it sits demonstrated platform value, implied by the fact that several independent partners have each chosen to build on the same architecture rather than on anyone else’s. Above that is unmodeled option value (targets in tissues already proven), the future programs inside tissues the engine has already reached, which are real, foreseeable, and still nameless. Higher again is frontier option value (tissues not yet proven), covering the eye, the heart, and whatever comes after them. At the top sits strategic control value, which is what an acquirer gains by owning the machine outright rather than renting pieces of it.
Each layer is progressively harder to defend with a number, even as the potential value it captures expands, which is exactly why conventional valuation stops at the first one and sentiment does uneven work on the rest. The point of this paper is not to squash the layers into a single figure, which would be false precision painted over an honest range. The point is that the premium is real, that its lowest layer is already visible in signed deals, and that its largest layers sit outside what the standard method measures. That is why it has never been priced in full. That is why it is still there for the taking.
One caution belongs on all of this, and it is the objection a valuation specialist would raise first. Platform confidence does not only live in a line labeled platform. It can also sit quietly inside the rest of a model, in a probability of success set a little higher because the technology has worked before, in a terminal assumption made a little friendlier, in a multiple that drifts up because the company is respected. To the extent that credit is already there, some of what this paper calls the premium is being counted a second time. The honest equation is not the sum of the parts plus the whole premium. It is the sum of the parts plus whatever engine value is not already embedded elsewhere.
Two things suggest that overlap is small rather than large. The first is the $1 itself. A model that had quietly folded substantial engine value into its other assumptions would have little reason to carry the explicit platform line at a placeholder, and the placeholder is what the published breakdown shows. The second is that the credit which does arrive through sentiment and multiples is unstable by nature, since it moves with mood rather than with evidence, which is a different thing from value a model has actually assigned. Neither observation proves the overlap is zero. Both suggest that the great majority of what is described here is genuinely incremental, and the premium in this paper should be read as the incremental portion rather than the whole.
10. The one buyer who can see it
There is one kind of buyer for whom the premium is not invisible at all, and this is where the argument completes itself. A strategic acquirer does not value a company with a public-market spreadsheet and a $1 placeholder. Its deal team values exactly the thing the spreadsheet excludes, the future programs, the tissues still to come, the repeatability itself, because the future output is precisely what an acquirer is buying. The blind spot this paper describes is a public-market artifact, not a universal one. The one participant structurally equipped to price the engine is the one that intends to own it.
Look again at the partner deals through that lens and they change meaning. Novartis contracting around as much as $2 billion for one program and a few targets, Madrigal around up to $975 million for a single liver asset, those are not just evidence that the engine has value. They are the premium being paid, in fragments, by the actors structurally equipped to price it. Each license is a strategic buyer pricing a slice of the machine that the public market has no explicit way to price. The fragments are already trading. Only the whole is not.
Follow that one step further and you reach the scenario where the premium gets priced most completely: a competitive buyout. In a contested acquisition, rival bidders do not converge on the sum of the visible parts, which any of them could assemble more cheaply elsewhere. They bid up the one thing that cannot be assembled elsewhere, the proven machine, which means a bidding war for Arrowhead would be, almost by definition, an auction for the Jordan Premium itself. Earlier papers of mine walk through who those bidders would be and why each may be cornered into competing. This paper deliberately does not re-argue or price that scenario, for the same reason the Keystone kept the control premium outside its numbers: the case here does not need a deal to work. The engine is underpriced today, on the evidence already in hand. A buyout is simply the venue where the premium would be priced most completely, and in public. The market prices the box score. An acquirer prices the player.
11. What would make me wrong
A paper that only accumulates reasons to agree with itself is advocacy, not analysis. Here are the four ways this thesis breaks, stated as plainly as I can manage, because a reader deserves to know where the load-bearing walls are.
The translation record could break. Everything above rests on a track record, and a track record is a statement about the past. The first mark, that the engine translates, is exactly the kind of claim that a single high-profile failure in a new tissue would damage badly, and the newer the tissue, the thinner the evidence behind it. Liver delivery is proven many times over. Brain delivery has not yet been proven once in a human. A miss in September would not merely disappoint on one drug. It would put the generalization claim itself back in question, which is the mark the whole premium leans on.
The engine costs money, and shareholders have paid for it. This is the counterweight I would want raised against me, so I will raise it myself. More tissues mean more programs, more programs mean more spending, and Arrowhead’s share count has roughly doubled over the past decade. The engine described in this paper was not free. It was funded, in significant part, by diluting the very shareholders who own it. An engine that requires continuous financing can produce wonderful science and still deliver ordinary returns, because the value created gets divided across a growing number of shares. Anyone paying a premium for repeatability should be honest that repeatability has a running cost, and that the cost has historically been paid in equity.
The premium may be structurally unpayable. Here is the deepest objection to this paper, and it deserves to be stated at full strength rather than waved at. If the public market genuinely cannot price a discovery engine, as Section 2 argues, then perhaps it never will, in which case the mispricing is not an opportunity but a permanent condition. A gap that never closes is not value waiting to be captured. It is simply how the asset trades. The honest answer is that this objection is largely correct about the public market and is the reason the strategic buyer matters more to this thesis than a framework paper would like to admit. The historical record cuts both ways here. In the cases from Section 5 the gap did eventually close, which is encouraging, and in every one of them it closed late and only when something forced it, which is not. Recognition tends to arrive when something forces it, proof across a doubted frontier, or a buyer willing to pay for what the spreadsheet omits. Absent one of those, an investor can be right about the engine and wait a very long time to be paid for it.
Someone else could solve delivery. The premium assumes scarcity. If a competitor demonstrates its own route into the brain or the heart, the engine stops being the only one of its kind and the pricing power that comes with being alone at the frontier erodes quickly. Nothing in the record guarantees Arrowhead stays ahead. It guarantees only that it has been ahead so far.
None of these dissolve the argument, and I do not think any of them is more likely than not. Taken together, though, they mark the boundary of the claim. The engine is real and it is underpriced. That is not the same as saying the gap closes on a convenient schedule, or that the shareholder captures all of it when it does.
12. When the market will finally see it
A framework is only useful if it tells you when the gap closes. This one does.
The pattern is this: an engine’s premium expands when it proves itself somewhere the market believed it might not reach. Not every engine has a single hardest job, and it would be convenient rather than true to invent a universal law that happens to point at this September. What is true is narrower. Doubt attaches to specific frontiers, and when a frontier falls, the doubt attached to it is released, along with some of the doubt attached to every frontier still ahead.
For Arrowhead, the brain is the frontier that carries the most doubt today, because it is a room the field has historically entered with a needle in the spine. That makes the September readout the highest-information event this engine has faced. It also means the same logic applies again later. If the brain falls, the heart becomes the next frontier the market doubts, and clearing that one would expand the premium a second time, on the same principle rather than on a new one.
This is where this paper shakes hands with the last one. The case that Arrowhead’s brain readout reprices the company, argued in full in The Keystone, is just this general idea in one specific instance. That paper is about the trigger. This one is about the machine the trigger reveals. A good readout does not merely add a brain drug to the model. It forces the market to face the Jordan Premium it has never been able to price. Proving the engine across a frontier the market doubted it could reach is the moment the engine stops being a story a skeptic can wave off. It becomes a fact a model can no longer park at a dollar.
The timing, finally, is close. Arrowhead announced initiation of the Phase 1/2a study of its lead brain program in December 2025, and management has guided to topline data from the healthy-volunteer portion in September. Note what that readout is and is not: management has described it as primarily safety and total tau knockdown in healthy volunteers, not evidence of benefit in patients. The argument here does not hinge on it, the engine is underpriced today on the evidence already in hand, but September is the moment the market is most likely to be forced to look.
Notice there are two doors the recognition can walk through, and September could open either. The first is the data itself, which forces the public market to reprice what it can now no longer dismiss. The second is a bid, because a proven engine invites the one buyer who can price it, and nothing teaches a market the value of something faster than watching someone else pay for it.
13. The bigger lesson
Step back from the ticker, because the durable idea is larger than any one stock.
Any company whose real worth is a proven discovery engine will be underpriced by output-based valuation, in biotech and well beyond it. The pattern is old enough to have a history, as Section 5 shows, and it has repeated across four decades and several modalities. The very thing that makes SOTP trustworthy, that it prices only what can be named and defended, is the thing that makes the engine invisible. This is not a bug you fix with a better spreadsheet. It is baked into the act of pricing the nameable, and it hands a standing opportunity to anyone willing to tell an engine from a list and put a value on the difference.
The rule that keeps this from being wishful thinking is the stick from Section 4. An engine is not a company that calls itself a platform. It is a company that translates, generalizes, runs efficiently, and keeps a beat, all at once and over years. Hold every hopeful up to that stick, and the premium is earned by evidence instead of handed out on enthusiasm. Arrowhead is today’s clearest case, scoring on all four marks while the market carries its engine at a placeholder and its next two tissues inside it. The rule, though, is the thing to keep. It will outlast this one example.
You do not draft a generational talent for the games you can already put on the schedule. You draft him for the ones you cannot imagine yet. That is what Arrowhead’s engine is, and it is the one thing a spreadsheet built to price the nameable can never fully capture.
Here is the whole paper in a sentence. The market is happy to pay for the drugs Arrowhead has already made, and it will hand you, for free, the machine that keeps making them and the tissues that machine has not reached yet.
The market prices the box score. An acquirer prices the player.
Learn to see the engine, and you learn to see the value the market cannot. That is the Jordan Premium. It is sitting in plain sight, priced at a dollar.
A Note on Supporting Independent Research
If this white paper has been valuable to you, whether it shaped your thinking, validated your conviction, or simply saved you the time of doing this work yourself, a voluntary contribution is genuinely appreciated and directly funds the next paper.
For individual investors and readers
Any amount you feel reflects the value you received is welcome and meaningful. A contribution in the range of what you might pay for a single premium research report is a thoughtful gesture that makes a real difference.
For family offices, investment funds, hedge funds, and research platforms
This paper is the caliber of work that institutional research desks bill significant retainers to produce. If your team referenced it, distributed it internally, or used it to inform a position, a suggested contribution of $1,500 reflects the professional value of the analysis, though any amount is meaningful. Your support makes it possible to continue publishing at this level without a paywall that limits the reach of the ideas. If your organization requires an invoice to process a payment, please reach out directly at bioboyscout@gmail.com and one will be provided promptly.
There is no obligation and no expectation. This is purely a thank you for work that meant something to you.
Zelle: (847) 227-7909
Thank you for reading, and for being part of a community that takes this thesis seriously.
— Robert Toczycki | BioBoyScout
Important Risks, Disclosures, & Disclaimers
The author, Robert Toczycki (aka BioBoyScout), certifies that:
all views expressed in this white paper accurately reflect his personal opinions about the topic discussed;
he was not compensated in any form for producing this white paper; and
he has not received and does not receive compensation from Arrowhead Pharmaceuticals.
This paper is provided for informational and analytical purposes only. It does not constitute investment advice, financial advice, legal advice, or a recommendation to buy, sell, or hold any security, and it is not a recommendation as to any corporate course of action. The author holds a long position in Arrowhead common stock. Past performance is not indicative of future results, and forward-looking analysis is inherently uncertain. The author and BioBoyScout are not registered investment advisors. The author assumes no obligation to update this paper.
About the Author
BioBoyScout is the publishing name for Robert Toczycki, an independent biotech investment research writer based in Chicago. The BioBoyScout series publishes institutional-grade analysis of structural dynamics in RNA-class therapeutics, with particular focus on Arrowhead Pharmaceuticals’ TRiM platform and the broader competitive landscape. Robert is a registered US Patent Attorney with a JD, an Executive MBA completed at the top of his class, and a BS in Mathematics and Computer Science from the University of Illinois at Urbana-Champaign. He has a deep passion for financial analysis, particularly identifying valuation discrepancies and demonstrating them through rigorous, data-driven research and solid analytics.
Comments or questions: bioboyscout@gmail.com.
Copyright © 2026, BioBoyScout. All Rights Reserved.
