Semiconductor Puts vs. Dark-Pool Bids: Diverging Signals in AMD and MU Demand a Resolution

1The Contradiction, Stated Precisely

On 07-31, our desk recorded a $246M dark-pool block in AMD equity - 506,770 shares at $485.39 - printing within the same dossier window as $2.4M in AMD 490-strike puts, $7.4M in MU puts across the 830 and 850 strikes, and $1.2M in TSM 390-strike puts. Every options position carries a 1-day expiry (07-31) and was filled at 100% of the ask. The equity block and the put flow are directionally inconsistent on their face: one large actor absorbed a quarter-billion dollars of AMD equity while simultaneously, in the same session, other institutional actors paid full ask for near-term downside across the semiconductor complex.

Two coherent structural interpretations exist. First: the dark-pool block is a hedge of an existing short position or a portfolio completion trade, not a net-new directional long - making the put flow the cleaner signal of near-term institutional sentiment. Second: the dark-pool block is a net-new strategic long initiated by an actor with a multi-week or multi-month time horizon, making the 1-day puts orthogonal noise from a separate counterparty. The dossier does not contain sufficient information to rule out either interpretation with certainty. This note documents what each interpretation requires to be true and identifies the observable that resolves the ambiguity.

2Anatomy of the Options Flow

The options flow on record totals $11.0M in premium across three names - AMD ($2.4M), MU ($7.4M), TSM ($1.2M) - all at 1-day expiry, all filled at 100% of the ask. Paying the full ask on 1-day expiry contracts is a structural signal: the buyer is not price-sensitive and is prioritizing immediacy over execution cost. This is characteristic of either directional conviction on an imminent catalyst or a portfolio manager closing a hedge gap before a known event.

The MU flow is the heaviest in absolute premium: $3.1M at the 850 strike, $2.4M at the 850 strike (a second tranche), and $1.9M at the 830 strike - $7.4M total. MU's current composite score is 6.0/10, with a defensive score of 3.0/10 and a value score of 3.3/10. The defensive score is the lowest of the three names in the dossier. A composite defensive score of 3.0 indicates the name offers limited cushion against drawdown as scored by our methodology, which is consistent with, though not sufficient to explain, the scale of put protection purchased.

AMD's put flow ($2.4M, 490 strike) sits at a strike level above the current last price of $457.06. AMD's last was $457.06 at the time of this writing; the 490 strike was therefore already in-the-money relative to the block fill price of $485.39 but out-of-the-money relative to the spot price. TSM's 390 put was 6.1% out of the money relative to the $415.50 last. The cross-name breadth of the put flow - spanning AMD, MU, and TSM - is more consistent with a macro sector hedge than a single-name event, though the dossier does not record whether these flows share a common counterparty.

Strike Geometry and Price Displacement

AMD's dark-pool block filled at $485.39. AMD's last recorded price is $457.06 - a displacement of $28.33 per share, or approximately 5.85% below the fill price. On 506,770 shares, this represents an unrealized mark-to-market loss of approximately $14.35M relative to the block price as of the last recorded price. This displacement is material and is consistent with either: (a) the block having been executed earlier in the session when AMD traded higher, or (b) the block representing a loss-tolerant strategic accumulation. The dossier does not record the intraday timestamp of the block relative to the options flow, which limits precision here.

The AMD 490 put, struck at $490, is currently in-the-money relative to both the block fill ($485.39) and the last price ($457.06). A 490 put with AMD at $457.06 carries $32.94 of intrinsic value. On a 1-day expiry, the time value component is near zero, meaning the $2.4M in premium paid was purchasing primarily intrinsic value at the time of fill - unless AMD was trading above $490 when the puts were purchased, which cannot be confirmed from the dossier alone.

3The Hedge-of-Long Interpretation

The structurally simpler explanation for the simultaneous presence of a $246M equity block and $2.4M in AMD puts is that the dark-pool buyer is not a new entrant but an existing long covering tail risk. Under this interpretation, the block represents a portfolio rebalancing or index-related absorption - common in dark pools - and the put buyer is the same actor or a correlated actor managing the downside of that position.

This interpretation requires the following to be true: the dark-pool buyer had a pre-existing long position in AMD that was being extended or rolled, not initiated from zero; the put premium of $2.4M represents a cost-of-carry hedge on a position worth multiples of that figure; and the 1-day expiry was chosen because the specific risk event (a catalyst, earnings-adjacent window, or macro print) was dated to 07-31. AMD's composite score of 8.8/10 - with momentum at 9.0, tactical at 9.0, and defensive at 10.0 - is consistent with a name that institutional holders would want to maintain exposure to while managing near-term event risk, not exit entirely.

The defensive score of 10.0 for AMD is the highest in the dossier and is notable: it suggests, by our scoring methodology, that AMD offers strong cushion characteristics relative to the names scored. A holder of a large AMD long with a 10.0 defensive score has structural reasons to maintain the position and tactical reasons to hedge the near-term tail - which is precisely what the 490 puts, paid at full ask, accomplish.

4The Directional Bear Interpretation

The alternative interpretation treats the dark-pool block and the put flow as originating from separate, non-coordinated actors - the block from a strategic or passive buyer (index rebalance, ETF creation, pension allocation), and the put flow from a directionally bearish institutional actor making a 1-day catalyst bet. Under this reading, the put flow is the signal and the equity block is noise relative to the short-term thesis.

This interpretation is strengthened by the cross-name coherence of the put flow. AMD, MU, and TSM puts all printed on the same date with the same structure (1-day, ask 100%). A single-actor sector-level bear would produce exactly this pattern. MU's weak defensive (3.0) and value (3.3) scores make it the most structurally vulnerable of the three names to a sentiment-driven sell-off, which may explain why it attracted the largest absolute put premium ($7.4M vs. $2.4M for AMD and $1.2M for TSM).

This interpretation requires the following to be true: the dark-pool block was not initiated by the same actor as the put buyer; the put buyer had visibility into or conviction about a 07-31 catalyst; and the $11.0M in total put premium was deployed with the expectation that all three names would close below their respective strikes by end of session on 07-31. If AMD closes above $490, MU above $850, and TSM above $390, this interpretation is falsified by the data - the puts expire worthless and the bear read was incorrect.

Falsification Criteria

The directional bear interpretation is falsified if: AMD closes at or above $490 on 07-31 (AMD 490P expires worthless); MU closes at or above $830 on 07-31 (both MU strikes expire worthless); TSM closes at or above $390 on 07-31 (TSM 390P expires worthless). Full falsification across all three names would be strong evidence that the put flow was pure hedging rather than directional expression.

The hedge-of-long interpretation is weakened - though not falsified - if: AMD continues to trade below the $485.39 block fill over the subsequent five sessions with no additional equity accumulation observed; the dark-pool block does not recur in follow-on sessions; and MU or TSM composite scores deteriorate in subsequent readings, suggesting the sector-level weakness the put buyers may have been pricing was real.

5Composite Scores as Structural Context

AMD's composite of 8.8/10 - driven by momentum (9.0), tactical (9.0), and defensive (10.0) - places it as the highest-scoring name in the dossier. The defensive score of 10.0 is the maximum recorded and indicates, by our methodology, strong downside cushion characteristics. This is structurally incongruent with a name that a directional bear would target first, but it is precisely the name a large existing long would want to hedge given the size of the position ($246M) and the event risk on 07-31.

MU's composite of 6.0/10 is the weakest of the three, with defensive at 3.0 and value at 3.3. Momentum (8.0) and tactical (8.0) scores are elevated, suggesting the name has been in a momentum regime, but the absence of value and defensive support makes it vulnerable to a momentum reversal. The $7.4M in MU puts is the largest single-name put allocation in the dossier and is directionally consistent with a bet on momentum exhaustion in a name with no value or defensive floor.

TSM's composite of 8.1/10 includes a quality score of 9.1 - the highest quality score in the dossier - and a defensive score of 8.0. TSM is structurally the most defensible of the three names, yet $1.2M in 390-strike puts was purchased against it. The 390 strike is 6.1% below the $415.50 last, making this the most out-of-the-money put in the set on a percentage basis. This is consistent with a tail-risk hedge on the strongest name in the sector rather than a high-conviction directional bet - which paradoxically supports the macro hedge interpretation over the single-catalyst bear interpretation.

Sources and method

This note was written from evidence our own data desks recorded, not from a general model's recollection. The lines below are what the desks captured for this topic.

• AMD $2.4M · 490 P 07-31 · 1d out · ask 100%
• MU $3.1M · 850 P 07-31 · 1d out · ask 100%
• MU $2.4M · 850 P 07-31 · 1d out · ask 100%
• MU $1.9M · 830 P 07-31 · 1d out · ask 100%
• TSM $1.2M · 390 P 07-31 · 1d out · ask 100%
• AMD $246M block · 506,770 sh @ $485.39

Underlying sources

• Parallax Research Desk - Options Flow Recording, 07-31 (AMD 490P $2.4M, MU 850P $3.1M, MU 850P $2.4M, MU 830P $1.9M, TSM 390P $1.2M)
• Parallax Research Desk - Dark-Pool Block Recording, 07-31 (AMD 506,770 sh @ $485.39, $246M notional)
• Parallax Composite Scoring System - AMD 8.8/10 (momentum 9.0, tactical 9.0, defensive 10.0, quality 7.3)
• Parallax Composite Scoring System - MU 6.0/10 (momentum 8.0, tactical 8.0, defensive 3.0, value 3.3, quality 7.6)
• Parallax Composite Scoring System - TSM 8.1/10 (momentum 8.0, tactical 9.0, defensive 8.0, value 6.2, quality 9.1)
• Live Price Feed - AMD $457.06, MU $956.08, TSM $415.50 (as of note compilation)

Written by Parallax Research's research writer on 2026-09-03 and reviewed against the dossier it was given. Figures are as of that date and are not updated in place. Our scoring model's out-of-sample AUC is published in the whitepaper; nothing here is a forecast.