# Google's AI search is starving Reddit of visitors — short the traffic loser, not the whole market _AI-generated trading idea · BEARISH · GOOGL, RDDT_ > Canonical page: https://commonquant.ai/research/for-you/google-s-ai-search-is-starving-reddit-of-visitors-short-the--9ce22602-7d65-5abc-a4c5-74e59123e368 Unlike the broad tech selloff, Reddit's fall has a specific structural cause: Google's AI search answers users' questions on the results page, so people click through to Reddit less. This is not a dip that reverses when markets calm down — every improvement in AI search makes the pressure worse. While other tech names may bounce when panic fades, Reddit faces a shrinking core business, so weakness is more likely to persist. That makes it a cleaner short than the crowded, panic-driven trades elsewhere in tech. ## Idea Unlike the broad tech selloff, Reddit's fall has a specific structural cause: Google's AI search answers users' questions on the results page, so people click through to Reddit less. This is not a dip that reverses when markets calm down — every improvement in AI search makes the pressure worse. While other tech names may bounce when panic fades, Reddit faces a shrinking core business, so weakness is more likely to persist. That makes it a cleaner short than the crowded, panic-driven trades elsewhere in tech. ## Advanced Analysis ### Verdict: the Google-AI short on Reddit has a story problem — the numbers aren't shrinking yet The verdict is wait: the durable thesis — Google's AI answers keeping users on the results page and eroding Reddit's click-through traffic — is plausible and structurally motivated, but over a multi-quarter horizon the reported evidence does not yet confirm the premise. The Reuters report of October 28 ("Reddit plunges as Google AI search squeezes traffic") and provisional insider filings for the September 30, 2026 period showing roughly $32.6M of net open-market selling at Reddit are the strongest points for the bearish view. The strongest point against is Reddit's own latest quarter (ended June 30, 2026): revenue of $804.9M, up 21.3% sequentially, with net income up 24% to $252.8M and a 31.4% net margin — a shrinking business is not what those numbers show. The strategy's own completed backtest is lukewarm: over 24 months it lost 2.3% across 48 trades at a 50% win rate with a 10.7% maximum drawdown, and the 5-year window could not run due to an unresolved daily-data gap, with no robust parameter setup established. On top of that, the one unmet entry condition — trend strength above the 20 threshold (currently 15.7) — means the setup is still waiting, and an RSI of 7.8 flags a heavily oversold stock where a bounce is a live risk to any fresh short. What would flip the verdict: Reddit's Q3 2026 print showing clear revenue deceleration together with continued free-cash-flow erosion (free cash flow already fell 16.2% quarter over quarter to $261M), alongside confirmed net insider selling in the final ownership filings. #### Conviction score breakdown Composite score computed by the server from the applicable evidence-tier dimensions. | Measure | Value | | --- | ---: | | Thesis support | 55/100 | | Trade readiness | 35/100 | | Risk quality | 50/100 | | Backtest evidence | 35/100 | | Fundamentals trend | 30/100 | | Score | 41/100 | | Composite Score | 41/100 | | Evidence Tier | backtested | #### Decision scenarios Bull, base, and bear cases synthesized from the cited evidence tier. Likelihoods are rounded evidence-weighted judgments, not statistically calibrated forecasts. | Measure | Value | | --- | ---: | | Evidence Tier | backtested | ### Trade now: the short setup is one reading away from being live — but the evidence for it is thin Wait for a named condition. The strategy behind this idea is a daily short signal on Reddit (RDDT). As of the Oct 31 observation (market data computed 12:29 UTC), four of its five entry rules are already met: the price of $91.73 sits below both the 20-day EMA ($141.52) and the 50-day EMA ($150.76), the 20-day EMA sits below the 50-day, and RSI at 7.8 is far below 50 — a reading that oversold points to deeply washed-out momentum. The one missing condition is trend strength: ADX at 15.7 needs to be above 20. ADX measures whether a downtrend has real force behind it; until it crosses 20, the setup is a falling knife without confirmed trend, not the strategy's signal. Separate the two questions. The business thesis — that Google's AI answers reduce click-through to Reddit, making RDDT's weakness structural rather than panic-driven — is the idea's argument, and some numbers complicate it: RDDT's latest quarter (ended June 30, 2026) actually showed revenue up 21.3% quarter over quarter to $804.9M, net income up 24% to $252.8M, and a 31.4% net margin. A shrinking core business is not what those prints show yet. Meanwhile, the strategy's own completed backtest is unimpressive: over 24 months it lost 2.3% across 48 trades with a 50% win rate and a 10.7% max drawdown; the trailing 12-month window made 1.6% with a 54.5% win rate. The 5-year backtest could not run at all because of an unresolved market-data gap, and no robust parameter setup was established. This is a weak edge, honestly stated. If the ADX condition triggers (a sustained daily reading above 20 while the other rules hold), the trade's risk controls are fixed: a hard 12% stop, a 20% take-profit (roughly 1.7:1 reward to risk), a 60-bar time stop, and exits on a close above the 50-day EMA ($150.76) or the first resistance level ($156.39). The downside that matters for the thesis itself: any sign that Google's AI search is sending traffic back to Reddit — or another RDDT quarter of 20%-plus growth — invalidates the structural story well before price stops hit. \#\#\# Technical details \- Symbol: RDDT, daily timeframe. Current price $91.73 (last close, market state as of 2026-10-31 12:29 UTC); 2.3% above its range low and 64.6% below its range high. \- Entry rule 1 — price below 50-day EMA: $91.73 versus $150.76, met (price is $59.03 below). \- Entry rule 2 — price below 20-day EMA: $91.73 versus $141.52, met ($49.79 below). \- Entry rule 3 — 20-day EMA below 50-day EMA: $141.52 versus $150.76, met. \- Entry rule 4 — ADX (14) above 20: currently 15.7, not met (needs a rise of 4.3 points). \- Entry rule 5 — RSI (14) below 50: currently 7.8, met (extremely oversold; a bounce from here is a real risk to a fresh short). \- Exits: signal exit on a close above the 50-day EMA ($150.76, currently far away, roughly 64% above price); stop below the first support level ($139.55 per the supplied level set); take-profit near the first resistance level ($156.39); 12% hard stop; 20% hard take-profit; 60-day time stop. Position sizing: fixed 2% risk, max 25% of capital per position. \- Backtest evidence: 24-month window, 48 trades, -2.3% return, 50% win rate, 10.7% max drawdown; 12-month window, 22 trades, +1.6%, 54.5% win rate, 2.3% drawdown. Exit fills were approximated on daily bars, so treat win rate and drawdown as coarse. Note also: the idea tags GOOGL alongside RDDT, but the strategy only trades RDDT; GOOGL sits at $279.84, 30.5% below its range high, with its own ADX at 25.9. #### GOOGL price and trigger map Uses the idea timeframe and keeps price levels on the price axis. | Measure | Value | | --- | ---: | | Ticker | GOOGL | | Timeframe | 1d | #### RDDT price and trigger map Uses the idea timeframe and keeps price levels on the price axis. | Measure | Value | | --- | ---: | | Ticker | RDDT | | Timeframe | 1d | ### The short case: traffic squeeze signals, insider selling, and a cash-flow inflection The idea's core claim — that Google's AI search answers keep users on the results page and erode Reddit's click-through traffic — is supported by the one news item in the evidence set, by insider behavior, and by a softening in Reddit's cash generation, though the fundamental tables do not yet show a shrinking business. Note the backtest evidence is weaker than the narrative: over 24 months the strategy produced a -2.3% return with a 10.7% maximum drawdown. \#\#\# A sourced demand shock headline Reuters reported on 2026-10-28 that "Reddit plunges as Google AI search squeezes traffic," which is the event the thesis is built on. The causal chain is direct: Reddit monetizes user attention through advertising, so if AI answers reduce click-throughs from Google search — historically a meaningful discovery channel — ad impressions and pricing come under pressure before revenue optics do. The price reaction cited by Reuters is consistent with the market beginning to price that channel erosion. \#\#\# Insiders are net sellers, and heavily so at Reddit The ownership filings for the period ending 2026-09-30 (provisional coverage) show net open-market insider selling at both tickers, but the magnitude differs sharply in context: Reddit insiders sold a net $32.6M against 103 reporting holders, while Alphabet's net selling of $3.1M is spread across 1,086 holders and a vastly larger market value. Interpretation, not a sourced fact: proportionally heavier insider selling at the smaller name is consistent with the idea's bearish tilt and signals reduced confidence at the margin from those closest to the company. \#\#\# Reddit's cash generation cooled… #### RDDT Free cash flow Free cash flow trend from CommonQuant fundamentals/XBRL data; +226.3% from first to latest point. | Measure | Value | | --- | ---: | | 2022-12-31 | $-100254000 | | 2023-03-31 | $3719000 | | 2023-06-30 | $-54883000 | | 2023-09-30 | $-11639000 | | 2023-12-31 | $-84838000 | | 2023-12-31 | $-22035000 | | 2024-03-31 | $29213000 | | 2024-06-30 | $27183000 | | 2024-09-30 | $70269000 | | 2024-12-31 | $215820000 | | 2024-12-31 | $89155000 | | 2025-03-31 | $126599000 | | Latest Value | $126599000 | | Change Pct | $226.2782532367786 | | Ticker | RDDT | | Timeframe | reported periods | #### RDDT Operating margin Operating margin trend from CommonQuant fundamentals/XBRL data; +103.9% from first to latest point. | Measure | Value | | --- | ---: | | 2022-12-31 | \-0.25822970117038974% | | 2023-03-31 | \-0.43118358373030413% | | 2023-06-30 | \-0.2895083346536926% | | 2023-09-30 | \-0.09438190334830465% | | 2023-12-31 | \-0.17432331420881583% | | 2023-12-31 | \0.012072072072072072% | | 2024-03-31 | \-2.4252869778525947% | | 2024-06-30 | \-0.1103867929896438% | | 2024-09-30 | \0.019675557124853953% | | 2024-12-31 | \-0.43113816667371685% | | 2024-12-31 | \0.12361733616704892% | | 2025-03-31 | \0.00994747184353185% | | Latest Value | \0.00994747184353185% | | Change Pct | \103.8521795898947% | | Ticker | RDDT | | Timeframe | reported periods | #### GOOGL sector percentile check Ranks GOOGL against 239 companies in its sector using CommonQuant fundamentals. | Measure | Value | | --- | ---: | | Free cash flow | \98.32635983263597th percentile | | Operating margin | \96.5909090909091th percentile | | Rnd Intensity | 80th percentile | | Revenue growth (YoY) | \76.75276752767527th percentile | | Ticker | GOOGL | | Sector | Communication Services | | Peer Count | 239 | ### Scores - **Conviction score breakdown:** 41 - **Thesis support:** 55 - **Trade readiness:** 35 - **Risk quality:** 50 - **Backtest evidence:** 35 - **Fundamentals trend:** 30 ### Watch items - **RDDT — ADX (14), daily** - **RDDT — Q3 2026 revenue growth** - **RDDT — Close vs 50-day EMA** - **RDDT — RSI (14)** - **GOOGL — AI search product changes affecting outbound links** ## Key details - Symbols: GOOGL, RDDT - Timeframes: 1d - Tags: \#canonical-demand, \#cluster-version:1, \#direction:bearish, \#entity-kind:instrument, \#entity:GOOGL, \#entity:RDDT, \#horizon:unspecified, \#intent:research, \#symbol:GOOGL, \#symbol:RDDT ## Community - Upvotes: 0 - Views: 0 - Copies: 0 - Cosigns: 0 ## News sources - [Reddit plunges as Google AI search squeezes traffic - Reuters](https://www.bloomberg.com/news/videos/2026-10-28/fed-leaves-rates-unchanged-cook-and-hammack-dissent-video) — Reuters ## Related - [GOOGL trade ideas](https://commonquant.ai/markets/googl) - [RDDT trade ideas](https://commonquant.ai/markets/rddt) - [Latest market news](https://commonquant.ai/news) ## About CommonQuant Research Ideas and Advanced Analysis are generated with fresh, LLM-selected news headlines and grounded in SEC XBRL fundamentals and peer percentiles. Strategies built from these ideas are automatically backtested (over a timeframe-dependent historical window with walk-forward tuning) and stress-tested before they can go live, then monitored against the news hourly while they run.