A single institutional investor just committed $13.58 million to Palantir call options expiring in late 2027. At almost the same moment, Michael Burry has been publicly arguing the stock could shed 75% of its value.
Both positions cannot be right. Understanding why each exists is more useful than picking a side.
Palantir Technologies (PLTR) has become one of the most polarising names in U.S. equity markets this year, trading near $166 after recovering roughly 51% from its August low, yet still sitting about 20% below its all-time high of $207.52. The gap between the company’s fundamental performance and its valuation multiples has opened a genuine analytical fault line, not simply retail chatter. Institutional players on both sides are putting real capital behind their convictions.
This piece maps the mechanics of the bullish options bet, the fundamentals driving it, the specific valuation logic behind Burry’s warning, and the sector context that makes both views defensible. After reading, you will have a structured way to work out where you sit on this debate, grounded in the actual numbers rather than the loudest voices.
What the $13.58 million options trade is actually saying
The trade broke into two legs, and the specifics matter more than the headline figure.
The primary leg was 3,150 contracts of 17 December 2027 $190 calls, executed across six tranches at exactly $40.60 per contract, for total premium of roughly $12.76 million. Prior open interest on that strike was just 1,527 contracts, and volume subsequently pushed above 4,100. The break-even sits at $230.60, meaning Palantir must climb about 39% from current levels before this position returns a cent.
The secondary leg was smaller but far more aggressive: 300 contracts of 21 January 2027 $250 calls, at premium of roughly $819,000, with a break-even of $277.30.
That second break-even is the tell. At $277.30, this buyer needs Palantir to trade above the highest published Wall Street target on record. This is not positioning for a moderate recovery. It is a bet on a full rerating beyond any analyst’s stated view, and you should register that conviction level before reading the trade as a simple price signal.
For context, on 4 September 2026 total options premium on Palantir reached approximately $95.3 million, split between $45.98 million in calls and $49.28 million in puts. The market is heavily two-sided.
| Leg | Strike / Expiry | Premium Paid | Break-Even | Approx. Probability of Profit |
|---|---|---|---|---|
| Primary | $190 / 17 Dec 2027 | $12.76M | $230.60 | ~1 in 3 |
| Secondary | $250 / 21 Jan 2027 | $819,000 | $277.30 | ~1 in 6 |
Reading the probability, not just the premium
Long-dated calls carry a quiet cost: theta, the erosion of an option’s value as time passes. If Palantir fails to reach the break-even by expiry, the entire premium is lost. There is no partial credit.
The primary leg’s roughly one-in-three chance of expiring profitably means a two-in-three chance the premium disappears entirely. The secondary leg’s one-in-six odds are steeper still.
The execution style reinforces the read. Working the order across multiple sweeps, splits and blocks is how institutions minimise market impact, not how retail traders behave. For comparison, a Microsoft December 2027 calls structure involving 7,870 contracts carried roughly $10 million in net premium, which places Palantir’s trade squarely within normal institutional scale rather than as an outlier.
Institutional options positioning at comparable scale appeared in Microsoft during August 2026, where a $18.8 million net premium commitment across 9,215 contracts revealed deliberate payoff shaping through a three-leg structure rather than a simple directional bet, confirming that multi-leg sweep execution is a recurring pattern among large actors managing risk profiles rather than placing binary wagers.
When big ASX news breaks, our subscribers know first
Palantir’s fundamentals are not the dispute
Here is what bulls and bears actually agree on: the business is performing exceptionally.
In Q2 2026, Palantir generated $1.935 billion in total revenue, up 93% year-over-year. The growth engine was the U.S. commercial segment, which expanded 149% to $764 million. U.S. government revenue rose 90% to $809 million, driven by what CEO Alex Karp frames as demand for “AI sovereignty,” the idea that governments increasingly want mission-critical AI capability they control directly.
The margin profile is what separates Palantir from most software peers.
- GAAP operating margin: 47%
- Adjusted operating margin: 62%
- GAAP net income: $1.062 billion, a 55% net margin
- Net dollar retention: 157%
- Rule of 40 score: 155%
The Rule of 40 is a software benchmark: add a company’s revenue growth rate to its profit margin, and a score above 40 signals a healthy balance of growth and profitability.
The Rule of 40 methodology defines a healthy SaaS company as one whose combined revenue growth rate and profit margin equals or exceeds 40, a threshold that makes Palantir’s 155% score genuinely anomalous rather than incrementally impressive.
Palantir’s Rule of 40 score of 155% is nearly four times the benchmark threshold, and sits far above the median for S&P 500 software names, which clusters well below 40.
That figure is not a normal software metric. It is genuinely anomalous, and it means the bullish case starts from an unusually strong operational base. That is precisely why the valuation debate is so consequential rather than obvious.
Management’s confidence shows in the pattern of upward guidance revisions through 2026. Initial full-year guidance called for 61% growth ($7.182-$7.198 billion). After Q1, that lifted to roughly 71% ($7.65-$7.66 billion). After Q2, it rose again to 82% growth, targeting $8.15-$8.16 billion.
Three consecutive raises tell you execution has consistently outrun management’s own forecasts. If you were looking for cracks in the operating story, they are not here. Which means the entire debate collapses into one question: valuation.
Why Palantir’s multiples are genuinely hard to defend, and what Burry is actually arguing
To understand the bear case, you first need to understand multiple compression as a mechanism.
A valuation multiple is what the market pays for each dollar of a company’s revenue or earnings. Multiple compression happens when that price falls not because the business deteriorates, but because the market lowers the multiple it is willing to apply. The company can keep growing while the stock falls, purely because expectations for future growth get re-rated downward.
Multiple compression mechanics operated in exactly this pattern following Palantir’s Q1 2026 earnings: the stock fell 7% and erased roughly $26 billion in market capitalisation in a single session despite 85% revenue growth, a direct demonstration that exceptional fundamentals do not insulate a high-multiple stock from re-rating.
Palantir’s multiples leave enormous room for exactly that. The stock trades at roughly 450x trailing earnings, somewhere in the 117-230x range on forward sales and earnings, and around 295x on next-year estimates.
Set those against the baselines. The S&P 500 averages around 28x earnings. The “Magnificent Seven” mega-cap tech names average roughly 35x. On enterprise value to revenue, Palantir trades at 7.6x implied 2029 revenue versus 5.4x for Microsoft.
This is the coherent core of Michael Burry’s thesis. His argument is not that Palantir is a bad business. It is that a market cap above $400 billion prices in so much future growth that any normalisation of AI software multiples could drag the share price toward sub-$42, below a $100 billion valuation.
Burry has publicly argued Palantir faces roughly 75% downside, with its market capitalisation potentially falling below $100 billion, implying a share price near $42, driven entirely by multiple contraction rather than any business failure.
The reason this cannot be waved away is that the same compression has already happened once this decade.
| Metric | Value |
|---|---|
| Palantir trailing P/E | ~450x |
| Palantir forward multiple | ~117-230x |
| S&P 500 average P/E | ~28x |
| Magnificent Seven average P/E | ~35x |
| Post-compression SaaS median (EV/Revenue) | 5-7x |
After the 2021 tech peak, public SaaS median EV/Revenue multiples fell from 18-21x to 5-7x. The Bessemer Cloud Index compressed from 18.4x to 6.3x, and broader software P/E ratios collapsed from 51x to 27x.
That data tells you Palantir’s valuation risk is not theoretical. It has already played out across the software sector once. Adding weight to the bearish read, hedge fund managers Israel Englander and David Shaw have reduced their Palantir holdings, and the most cautious analysts sit well below the market: Morgan Stanley at Underweight with a $60 target, RBC Capital Markets at Underperform with $50, and Brian White with a $20 Sell rating.
The open question is whether Palantir’s AI sovereignty moat is strong enough to exempt it from that same gravitational pull.
Ecosystem switching costs are the mechanism that would structurally exempt Palantir from sector-wide multiple compression; the diagnostic question is whether its embedding in defence and intelligence workflows creates re-platforming costs comparable to those Nvidia has built through its developer ecosystem, or whether open-source AI models and cheaper alternatives erode that lock-in faster than the growth trajectory implies.
Where the bulls and bears actually disagree, and what Wall Street’s price targets reveal
Strip away the noise and the entire debate reduces to a single assumption.
It is not about financial performance, which both sides accept is exceptional. It is about whether Palantir’s entrenchment in defence, intelligence and mission-critical workflows constitutes a structural exemption from the multiple compression that has hit generic AI software, or merely delays it. Bulls see a durable moat. Bears see a stock temporarily defying the same gravity that pulled the rest of the sector down.
The analyst price targets map that disagreement almost perfectly.
Professional analyst targets on Palantir range from $20 to $255, one of the widest dispersions on any major U.S. equity, a direct readout of how differently institutions are underwriting the durability of its moat.
The 12-month consensus clusters around $192.19 to $201.23, with roughly 60% of 30 surveyed analysts rating the stock a Buy. But the consensus midpoint hides the real story. On the bull side, Citi Research upgraded to Buy at $235 in January 2026, Rosenblatt Securities raised its target to $225 by May, and William Blair sits at $200. On the bear side sit Morgan Stanley’s $60, RBC’s $50, and Brian White’s $20.
That spread is not noise. It is the disagreement over moat durability rendered in numbers, which tells you there is no consensus reality to anchor to.
The bifurcation is already visible across AI software valuations more broadly. LLM and generative AI assets trade at 12-20x EV/Revenue, while enterprise AI applications have reverted to 3-6x. Which bucket Palantir belongs in is the whole argument. Strategists at Allianz note a forward P/E glide path across AI equities from roughly 35x today toward 23-24x within three years, limiting the room for further multiple expansion.
Priced-for-perfection dynamics appear across the high-multiple software sector, not only at Palantir; Snowflake, UiPath, and Datadog all posted strong earnings in the same period and still saw share price declines, confirming that the post-earnings sell-off pattern is a sector-level phenomenon rather than a company-specific signal.
Three variables will ultimately decide which thesis proves correct:
- The pace at which AI software multiples normalise across the sector.
- The durability of U.S. government AI spending through budget cycles.
- Whether commercial AI alternatives commoditise Palantir’s platform before its customer base becomes structurally irreplaceable.
Both sides acknowledge the risk factors that sit underneath these: customer concentration, the historical lumpiness of government contracts, and the possibility that competing AI tools erode Palantir’s edge faster than its growth trajectory implies. Where you land depends on your own read of AI infrastructure adoption, not on an average that papers over a real structural divide.
Making an informed call in a market where both sides have a credible case
The two theses reduce to one question you have to answer for yourself: do you believe Palantir’s defence and intelligence moat is structural enough to sustain premium multiples through an AI software normalisation cycle, or is that moat insufficient to offset the sector-wide pull toward compression?
The bull case is operationally grounded. Full-year 2026 guidance of $8.15-$8.16 billion at 82% growth is real, and the fundamentals are broadly uncontested. But it is valuation-dependent, and that dependency is the entire bear argument.
The options trade is a useful signal of one institutional actor’s conviction, not directional certainty for you. Remember the maths: even the primary leg’s roughly one-in-three chance of profit means a two-in-three chance the entire premium is lost. That asymmetry cuts both ways.
Time horizon changes everything. The bull thesis is a multi-year compounding story. The bear thesis is a mean-reversion story that could play out over months, and the Allianz glide path from 35x toward 23-24x within three years is the compression clock ticking in the background.
The year-to-date price action is the practical warning. Palantir opened near $180, fell below $110, and recovered to $166. Holding through that kind of drawdown is one thing for an equity position. A leveraged options position with a fixed expiry structurally cannot absorb it, which is the real distinction between the institutional trade and a retail equity stake.
Before taking any position, answer three questions:
- What is your conviction on Palantir’s moat durability over a 3-5 year horizon?
- What drawdown can you tolerate if multiple compression hits before fundamental growth catches up?
- Are you underwriting the equity or the options, and does your time horizon match the instrument’s expiry?
The bull case is operationally grounded but valuation-dependent. The bear case is mechanistically sound but timing-uncertain. Both are legitimate, and the value here is in making that distinction navigable, not in choosing between two confident voices for you.
This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.
Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors. These statements are speculative and subject to change based on market developments and company performance.
