🔁 RDDT $4.5M LEAP Roll — Our OI Work Shows a Week-Long Move from Sept-2028 to Dec-2028 $250 Calls

📅 2026-08-12 | 🤝 Floor Blocks, Not a Sweep

🔄 Updated 2026-08-13 pre-market — the next-day OPRA open interest confirmed the six-session roll, on the exact branch we published. December-2028 open interest climbed 924 → 1,403 (+479), clearing the ≈1,000+ we predicted, while September-2028 fell 770 → 666 (−104), landing on the ≈650-or-lower we predicted. The inverse relationship held for a seventh session. One label correction: the September leg's provisional BTO ⏳ is retired — its open interest fell, so that side is the closing half of the roll, not an open. See the ✅ RESOLVED box.

---

🎯 The Quick Take

Two floor blocks printed in Reddit on 2026-08-12: 518 December-2028 $250 calls at $45.05 (≈$2.33M) and 518 September-2028 $250 calls at $41.40 (≈$2.14M), both tagged BUY, with spot at $154.58. That's a real number — ≈$4.5M combined — but it's not the story. Our own six-session open-interest history shows the December-2028 $250 line has been climbing by roughly this size every single day for a week while the September-2028 $250 line falls by almost exactly the same amount. That's a calendar roll in progress, not two fresh bullish bets, and it buys the position exactly one extra earnings report — Q3 2028 — that falls between the two expirations.

---

🏢 Company Overview

Reddit, Inc. runs the community-based discussion platform organized around user-created subreddits. It went public on the NYSE on March 21, 2024 at $34.00, is headquartered in San Francisco, and is led by co-founder/CEO Steve Huffman with CFO Andrew Vollero (company profile). Reddit sits in the Communication Services sector, Internet Content & Information industry (StockAnalysis overview).

The business has two revenue lines of very different sizes: $762M of advertising revenue in Q2 2026, up 64% year over year, and $43M of "other" revenue (which contains AI data licensing), up 24% (Q2 2026 results). That gap matters for reading this trade — data licensing is still roughly 5% of revenue and growing slower than advertising. It's a valuation story right now, not yet a revenue story.

Reddit trades a Class A share on the NYSE and is a Russell 1000 member. Market cap is ≈$29.73B, enterprise value ≈$26.96B (≈$2.8B of net cash), with 192.40M Class A shares outstanding and 207.0M diluted shares across all classes (StockAnalysis market cap page, Q2 2026 release). Public float is 137.95M shares, and short interest sits at 8.83% of shares outstanding (StockAnalysis statistics).

---

💰 The Trade, In Plain English

At 13:33:22 ET, two floor blocks printed within the same second — both December-2028 and September-2028 $250 calls, both tagged BUY on the capture, both at a strike that sits ≈62% above today's $154.58 spot. Neither option has a shred of intrinsic value: 100% of what was paid is time value. These are LEAP (long-term) options expiring more than two years out.

🤝 Both legs printed as floor blocks — a negotiated, off-book transaction between a broker and a known counterparty, not an order that hit the lit screen. There was no aggressor lifting an offer here; someone arranged this trade in advance. That means the BUY tag on the capture is reported, not tape-proven — floor trades take no liquidity, so there's no NBBO aggressor signature to check it against.

Ticker: RDDT on AInvest

Both legs printed at a size below their prior open interest, which means neither is provable as a fresh opening trade from today's tape alone. That's exactly why the open-interest history below matters more than the BUY labels.

---

✅ RESOLVED — The Roll Continued, Exactly on the Published Branch

Updated 2026-08-13 pre-market. Resolving OPRA snapshot timestamped August 13 (reflects the August 12 close, after this print); baseline is the August 12 snapshot (reflects the August 11 close, before this print).

The pattern read was the right one, and it is the reason this article was written. Three branches were published: roll continues, both lines tick up together, or the roll completes. December rose and September fell — branch one, for the seventh consecutive session. The seven-session arc is now December 69 → 268 → 478 → 524 → 724 → 924 → 1,403 against September 1,389 → 1,466 → 1,266 → 1,025 → 847 → 770 → 666.

The label correction, stated plainly. Both of today's legs carried a BUY tag off a negotiated floor block, and the classifier defaulted both to BTO. September open interest fell, which rules out an open on that side. The honest description is a roll: the December leg opened, the September leg is being retired. This does not change the article's thesis — it was always framed as a roll — it corrects the per-leg order-type label to match.

Magnitude caveat. September retired only 104 contracts against a 518-lot print (20%), so most of that leg's volume transferred between holders. December captured 479 of 518 (92%). The migration is real but it moves in smaller net steps than the print sizes suggest.

Control check. The move is confined to the $250 line: September-2028 $200C 28 → 29 (+1), $300C 11 → 10 (−1). No chain-wide event.

---

✅ Label note (2026-08-13): the roll thesis below is confirmed by next-day open interest (December +479, September −104). One correction: the September-2028 leg's provisional BTO label is retired — its open interest fell, so that leg is the closing half of the roll, not an open. See the ✅ RESOLVED box above.

🤓 What This Actually Means — Plain English

What's actually happening here is a calendar roll, not two new bullish bets. Think of it like this: someone already owned September-2028 $250 calls. Instead of just holding them or letting them run, they're gradually selling down that position while buying up the equivalent December-2028 $250 calls — same strike, same underlying view, just a later expiration date. Over six sessions, roughly 850 contracts have made that migration.

Why does the extra three months matter? Because of earnings math. Counting forward from the next scheduled Reddit report, roughly 8 quarterly earnings releases land before the September-2028 expiration, and roughly 9 land before December-2028. The difference — Q3 2028, expected around late October 2028 — sits in the gap between the two expiration dates. Someone paid a bit more (≈$3.65 a contract, the difference between the December and September premiums) to make sure their position survives long enough to see one more earnings print before it expires. That's a precise, mechanical explanation for why anyone rolls a LEAP forward: they want more chances to be proven right.

What "100% time value" means for a retail trader: neither of these options is worth anything if exercised today — RDDT would need to more than double just to reach the strike, let alone turn a profit. Every dollar paid here is pure bet-on-time-and-volatility, which is why these options decay fastest of all if the thesis doesn't play out, and why they're only sensible as a small, defined-risk slice of a portfolio, never a core position.

The real hurdle is the breakeven, not the strike. The September-2028 leg breaks even at $291.40 (+88.5% from today's spot); the December-2028 leg breaks even at $295.05 (+90.9%). That's not "RDDT needs to go up" — that's "RDDT needs to nearly double from here, and stay there, over roughly two years." For context, $250 times the 207.0M diluted share count works out to ≈$51.8B of equity value (this is the researcher's own arithmetic, not a consensus estimate) — only about 19% above Reddit's all-time-high market cap of $43.56B set on December 31, 2025. On today's 9.71× EV/Sales multiple applied to a much larger 2028 revenue base, that's demanding but not fantastical — it doesn't require the stock to re-rate to a richer multiple than it already touched once, just to compound revenue into a bigger number at roughly today's multiple.

---

📈 Chart Check-Up

Reddit's one-year chart tells a clear story: a run to an all-time high near $282.95, then a grinding decline of roughly a third from that peak. The sourced figures put the drawdown at −30.10% over 52 weeks and −31.76% versus the 2025 year-end market cap — use those numbers, not a chart-window-dependent −35% figure, though the chart visually shows a decline of roughly a third.

Gamma positioning at a spot of $154.66 shows resistance at $160 (Moderate strength) and support at $150 (Moderate strength) — both roughly 3% away from spot in either direction. These are dealer-hedging-implied levels, not hard floors or ceilings; they shift as the day's flow changes, but $150–$160 is where the options market currently expects the most friction.

The options market is pricing:

• ±4.37% by August 14 ($147.88 – $161.40)
• ±8.50% by August 21 ($141.50 – $167.78)
• ±17.53% by September 18 ($127.53 – $181.75)

None of those windows get remotely close to the $250 strike on today's LEAP trade — which is exactly the point. The near-term implied move tells you what the market expects over days and weeks; the LEAP roll is a two-year structural bet that doesn't need to resolve inside any of these windows.

---

🎪 Catalysts

Keep two clocks separate: the events below happen over the next six months, while the options expire in September and December 2028 — roughly two years out. No single catalyst below resolves the LEAP position by itself; it's the cumulative effect of 8-9 earnings prints between now and expiration that the roll is really betting on.

Already happened

Reddit's Q2 2026 report on July 30, 2026 beat on every headline number — revenue of $804.91M (+61.1% YoY) against a $731M consensus, EPS of $1.25 against $0.95 consensus, adjusted EBITDA of $343M at a 42.6% margin, and free cash flow of $261M, up 135% (MarketBeat earnings, Q2 2026 release). Q3 guidance of $860–870M came in above the $829.7M consensus. And the stock still fell ≈22% that day — its worst single-day drop on record (MarketBeat).

What broke it wasn't the numbers — it was the users underneath them. U.S. daily active users fell ≈300,000 sequentially, and U.S. logged-in users were roughly flat at ≈23M (+1% YoY), even as U.S. revenue per logged-in user jumped from $17.85 to $27.62 (earnings call transcript, MarketBeat analysis). CEO Steve Huffman told analysts that "visibility into referral traffic remains low" and that AI Overviews from Google "has yet to make a similar level of positive impact" as traditional search links (Q2 2026 call transcript). No new AI licensing deal was announced on the call, which the contemporaneous news roundup flagged as a driver of the selloff alongside the user metric (MarketBeat news roundup).

This looks like a multiple event, not a fundamentals event — revenue accelerated all the way down. The de-rating actually started before the print: BofA cut its target from $205 to $175 on April 2, 2026, and Citizens JMP cut from $250 to $240 on May 1, 2026 (price-target history), while revenue was still growing 69% in Q1. July 30 delivered the capitulation leg on top of an already-compressing multiple.

Be precise on data licensing here: it's currently a valuation story, not a revenue story — that $43M "other" line growing 24% sits well behind advertising's $762M growing 64%. The pricing lever for the whole AI-licensing category is Reddit's active federal lawsuit against an AI search company, filed in October 2025, alleging unlawful scraping of Reddit's data (case background); the case remains active as of this writing, and Reddit reportedly advanced the suit on the same day as the earnings drop (news roundup).

Post-earnings, analyst reaction split sharply: Wells Fargo cut its target to $142 (below today's price) while Truist raised to $275 and B. Riley raised to $270 — into a 22% crash (price-target history). Consensus across 34 analysts averages $216.35 with a range of $120–$300 (StockAnalysis forecast).

Upcoming

Q3 2026 earnings are expected ≈October 29, 2026 — this date is estimated, not company-confirmed (MarketBeat earnings). The number that will decide the tape is the U.S. DAU trajectory, testing whether the flat-logged-in-user problem persists or stabilizes.

Any new or renewed AI data-licensing agreement is unscheduled — management describes the marketplace as "expanding" across training, post-training, grounding and search indexing but gave no timeline on the July 30 call (Q2 2026 call transcript). Its absence cost Reddit roughly 22% in a single day once already; its arrival would be the single highest-beta discrete catalyst for both the near-term stock and the long-dated licensing thesis this LEAP roll is ultimately expressing a view on.

Q4 2026 / FY2026 results are estimated for ≈February 2027 by reporting cadence — also not company-confirmed.

Between now and the December-2028 expiration, roughly 9 quarterly reports will print, with the extra one relative to the September-2028 expiration being Q3 2028, expected ≈late October 2028 — the specific quarter this roll appears to be buying.

---

🎲 Four Ways to Read This

🎲 The YOLO trader

You're looking at a floor block that's already ≈62% out of the money with a breakeven near +90%. If you want exposure to this exact structural bet — "Reddit's AI-licensing optionality plus advertising diversification eventually justifies a near-$52B valuation" — buying a handful of contracts at the same $250 strike costs real money for a long shot. Size it as a lottery ticket, not a position: cap it at what you'd be fine losing entirely, because 100% time value on a two-year LEAP can bleed steadily even if you're eventually right about direction, and being early is functionally the same as being wrong if the contract expires first.

📈 The swing trader

This particular trade isn't built for you — it's a two-year structural position, not a weeks-to-months swing. But the setup underneath it is useful: gamma resistance sits at $160 and support at $150, both Moderate strength, with the implied move putting a ±8.5% band ($141.50–$167.78) around spot through the August 21 monthly expiration. If you want to trade the next few weeks, that $150–$160 zone is where the options market expects the most chop, and a break of either level on volume is your signal, not the 2028 strike.

💰 The premium collector

There's no credit here to collect — both legs on today's print were debits, and the roll itself (selling September, buying December) nets out close to a wash in premium terms since it's being done gradually rather than as a single defined package. If you're inclined to sell premium against RDDT, the far-OTM $250 strike and its ≈90% breakeven suggest the market is pricing real skepticism into that level — a covered call or cash-secured put strategy at strikes much closer to the current $150–$160 gamma zone would harvest more premium per unit of risk than anything near $250.

🌱 The beginner

Here's the plain-English version: two investors placed bets, months apart in spirit even though today they both traded within one second, that Reddit stock will be worth almost double what it is today — sometime before September or December 2028. They didn't just buy those bets fresh; our data shows someone has been quietly shifting a position from the earlier date to the later date for about a week, buying themselves one more earnings report to be right. That's called a "calendar roll," and it's a sign of patience and conviction, not urgency. Before you copy any options trade like this, understand that a $250 strike on a $154.58 stock means the stock has to rise roughly 62% just to have any real value at expiration — and if it doesn't get there, every dollar paid is gone. Start with strategies much closer to today's price if you're new to options.

---

⚠️ Honest Limits — What the Tape Cannot Prove

The OPRA tape and open-interest history can tell us the size, price, and mechanism of a trade, and they can strongly suggest a pattern across sessions — but they cannot prove intent. Specifically:

• We cannot prove today's prints are "the same" position rolling, only that the open-interest lines have moved in a consistent, inverse pattern for six sessions at a comparable size to today's trade. It is possible — though the pattern argues against it — that today's buyer is a new party unrelated to the prior week's flow.
• Neither leg's open/close status is provable from today's tape alone, since both printed below their respective prior open interest. The next-session OPRA snapshot is the definitive test, not today's BUY tags.
• OPRA cannot identify the broker, the counterparty's identity, the order ID, or whether this trader holds an offsetting position elsewhere (stock, other options, or hedges in a different account).
• Research gaps, disclosed honestly: this session's search budget was exhausted before completion, and several sources (Reuters, Investopedia, Yahoo Finance, SEC EDGAR, Reddit's own investor relations site) blocked or refused the fetch. Share-class voting ratios are unstated because the 10-K/proxy could not be retrieved. FY2027 and FY2028 consensus revenue are paywalled, so the $51.8B-valuation-context math in this article is the researcher's own arithmetic, clearly labeled as such — not a Street consensus figure. The Q3 2026 earnings date (≈October 29) is a third-party estimate, not company-confirmed. No lock-up expiration or index-inclusion event was found in available sources — that is an absence of evidence, not evidence that no such event exists.

---

This is not investment advice. Options trading involves substantial risk, including the potential loss of the entire premium paid, and is not suitable for all investors. Past performance and options flow do not guarantee future results. Do your own research and consider consulting a financial advisor before trading.

---

Last updated: 2026-08-13 (pre-market) — the next-day OPRA open-interest snapshot confirmed the roll on the published branch. Dec-2028 $250C 924 → 1,403 (+479 against 518, cleared the predicted ≈1,000+): OPEN (BTO); Sep-2028 $250C 770 → 666 (−104, landed on the predicted ≈650-or-lower): NOT AN OPEN, the closing half of the roll — the provisional BTO label on that leg is retired. The roll thesis and title are unchanged. The ⏳ callout was replaced with the ✅ RESOLVED box.