This Week in Brief

The last 7 days produced 149 total trigger events across 6 active anomaly signals between 2026-04-16 07:19:47 and 2026-04-23 05:00:00.

This week's cleanest read came from the social-data side of the package. aave was the clearest cross-signal name, leading social_dominance_spike and also topping the non-BTC/ETH side of project_in_trends, this is likely due to the recent ~$292 million KelpDAO exploit; Aave became heavily involved as a major victim of the fallout because rsETH was an accepted collateral asset on its V3 and V4 markets (Ethereum and Arbitrum).

ethereum was also visible in multiple sections: eth_whale_dump remained active through most of the window, while project_in_trends and social_price_correlation also signaled on ETH later in the week.

price_network_activity_divergence led by signal count again, but the useful part was narrow. The most frequent recurring signals concentrated in pufeth, kelp-dao-restaked-eth, and maker, while the rest of the tail looked mostly like noise.

Trigger Count Table

Social Dominance Spike Spotlight

This signal favors repeat activity, peak strength, and market readability. I excluded stablecoins from the main basket.

jasmy, just, and origintrail also showed up, though with lighter repeat counts.

Projects in Trends Beyond BTC and ETH

Raw project_in_trends counts were led by ethereum and bitcoin, but the more useful non-BTC/ETH data this week was centered on aave, with xrp ranked second. tether also appeared repeatedly, though it’s likely due to the KelpDAO incident.

This section was much less diverse than last week, telling us there has been less broad attention towards altcoins.

All Top Projects in Trends & Sentiment Analysis

Weighted sentiment note: the table below is computed from ClickHouse metrics using the published weighted sentiment formula, sentiment_balance_v2 × unique_social_volume_1d, then normalized as a 60-day rolling z-score.

The main takeaway from this cross-check is that project_in_trends did not line up with a broad positive-sentiment breakout this week. Outside of ETH's short-lived positive burst on 2026-04-22, most of the top trend names were appearing against neutral-to-negative weighted sentiment conditions, especially aave, bitcoin, and tether.

Price-Network Activity Divergence Top 3

These are the top price_network_activity_divergence rows ranked by weekly peak value. The signal stayed fragmented overall, but the first three names were clear outliers before the tail dropped into lower-value one-offs.

This is more of a narrow ETH-restaking and DeFi read than a broad market signal this week. After these three rows, the rest of the table quickly faded into isolated low-value observations.

What Else Happened

eth_whale_dump was one of the most persistent signals. This anomaly is ETH-only by definition, so the useful read here is the length and intensity of the sequence rather than any cross-asset comparison. It triggered 31 times on ethereum, with an average value of 3,143,985.74 and a weekly peak of 7,639,002.00 at 2026-04-20 10:26:35. The signal stayed active from 2026-04-16 07:19:47 through 2026-04-22 18:23:35, which makes it a clear week-long run within this ETH-only signal. We are working on an eth_whale_buy signal as well to help paint a fuller picture in the future.

large_usdc_usdt_mint appeared 4 times and peaked at 1,526,622,101.01 on 2026-04-21 00:00:00. That is large enough to note, but not extreme enough to say it is surprising.

social_price_correlation returned, but only on ethereum, with 3 triggers between 2026-04-20 10:00:00 and 2026-04-20 12:00:00. The values stayed narrowly negative at -0.21, so this was a real but very concentrated signal. Social price correlation is an anomaly that measures whether social sentiment and price are behaving in an unusually aligned or unusually decoupled way. This time it was more decoupled than usual.

Where Signals Overlapped

Nothing very notable or valuable. Overlaps were due to the KelpDAO news event in the case of AAVE and natural popularity and an eth-specific anomaly (eth_whale_dump) in the case of ETH.

Worth Watching

• aave: strongest overlap between the social and trend sections.
• ethereum: worth checking because project_in_trends and social_price_correlation both printed on ETH while eth_whale_dump also stayed active across the week.
• nexo: the clearest repeat social-attention name after aave.
• pufeth and kelp-dao-restaked-eth: the top price_network_activity_divergence rows hint at a narrower ETH-restaking pocket worth checking separately.

About the Anomalies We Track

Covered anomalies and assets per anomaly type

1. Network Activity & Price Divergence Anomaly - detects price surges that are not accompanied by a corresponding rise in network activity (real growth). Triggered only if the price growth exceeds 3% on the same day, ensuring the alert is relevant for significant bullish moves.
2. Project in Trends Anomaly - detects when a specific crypto token gains significant traction in online discussions, placing it among the top 10 most-mentioned words in crypto-related conversations.
3. Large USDC-USDT Mint Anomaly - detects daily combined $1 billion or larger minting events of USDC and USDT stablecoins on the Ethereum blockchain.
4. Social Price Correlation Anomaly - detects whether social sentiment and price are behaving in an unusually aligned or unusually decoupled way. The current implementation runs on 1h X.com social data for bitcoin, ethereum, solana, dogecoin, xrp, trx, and cardano.
5. Social Dominance Spike Anomaly - detects abnormal spikes in social dominance metrics. It alerts when values exceed a statistical significance thresholds of 3 standard deviations over 30 days.
6. Social-Dev Score Anomaly - detects unusual activity compared to recent 30-day history of social media metrics and development activity, the score itself is weighted at 60/40 for social and dev activity.
7. Hyperliquid Average Funding Rates Anomaly - detects when rates exceed predefined high/low thresholds of +40% and -20%, measured hourly.
8. ETH Whale Dump Anomaly - detects potential large-scale ETH dumps by monitoring whale wallet behaviors, threshold of 500 ETH and uses a 30-day holding period as a key parameter to distinguish between normal trading activity and potential dump scenarios.

Disclaimer: The opinions expressed in the post are for general informational purposes only and are not intended to provide specific advice or recommendations for any individual or on any specific security or investment product.