Hook
Ninety addresses. Each holding over 10,000 BTC. A six-month high, according to Santiment’s latest on-chain data release. The immediate reaction among crypto Twitter was predictable: whales are accumulating, the bull run is imminent. But as a smart contract architect who has spent years dissecting on-chain data pipelines, I’ve learned that the most dangerous input to a trading decision is a metric that feels too clean. Address counts are not entities. Clusters are not individuals. And a 90-wallet ATH is a data point that demands rigorous deconstruction before it can be used as a signal.
Context
Santiment is a well-regarded on-chain analytics platform. Their methodology involves entity clustering—linking multiple addresses that are likely controlled by the same entity using heuristics like common input ownership, transaction patterns, and exchange deposit behavior. When they report “addresses holding at least 10,000 BTC,” they likely mean clustered entities, not raw addresses. However, the headline uses “Addresses,” which introduces ambiguity. The distinction matters: one entity (e.g., a custodial exchange like Coinbase) may control hundreds of thousands of BTC across dozens of wallets, but Santiment’s cluster would count it as one entity. Conversely, a single wealthy individual might spread their holdings across multiple addresses, each under 10,000 BTC, and be invisible to this metric.
This is not a critique of Santiment—their clustering is among the best in the industry. But any on-chain signal derived from address thresholds must be contextualized by the underlying aggregation method. The raw data from Bitcoin’s UTXO set shows 90 addresses with a balance > 10,000 BTC. That is a factual statement. The inference that these represent 90 distinct “whales” acting independently is a probabilistic assertion, not a fact.
Core
Let’s parse the three key data points from the report:
1. Elite whale addresses (≥10,000 BTC) increased to 90, a six-month high. The previous low was likely around 84-85 addresses during the mid-2023 bear market. A net increase of 5-6 addresses over six months is not a parabolic surge. It’s a gradual shift. To put it in perspective, the total supply of Bitcoin is ~19.5 million BTC. Ninety addresses holding ≥10,000 BTC each represent at least 900,000 BTC, or roughly 4.6% of the circulating supply. That’s significant concentration, but the number of addresses is small. A single ETF custodian (e.g., Coinbase for BlackRock’s IBIT) could account for multiple 10,000+ BTC addresses if they segregate client funds by wallet. The increase could be an artifact of ETF custodians splitting larger wallets into smaller operational buckets, not new whales accumulating.
2. Mid-level addresses (10–10,000 BTC) added approximately $1.5 billion worth of BTC over the past two weeks. This is a more interesting signal. Mid-level addresses are less likely to be ETF custodians (who typically hold larger aggregated balances) and more likely to represent institutional investors, high-net-worth individuals, or trading desks. The $1.5 billion inflow over two weeks is substantial—roughly 35,000 BTC at current prices. This cohort is often called the “smart money” because they are less reactive than retail and more strategic. However, the $1.5 billion figure is sourced from Santiment’s “Total Supply on Exchanges” metric, which tracks the aggregate balance of addresses identified as belonging to exchanges. A decrease in exchange balances suggests accumulation, but the direction of movement matters: if mid-level addresses are moving BTC from exchanges to self-custody, that’s a bullish structural signal. If they are simply consolidating UTXOs, the signal is noise.
3. Small addresses (<10 BTC) continue to decline. The headline narrative is that retail is selling or moving to exchanges. But the alternative interpretation is that retail is consolidating small UTXOs into larger ones to reduce future transaction fees, especially with Bitcoin’s mempool congestion. The average number of UTXOs per address has been decreasing for years as users batch transactions. The “small address” decline could be a technical optimization, not a bearish sentiment indicator.
From a market structure perspective, the combination of elite whale addresses increasing and mid-level addresses accumulating suggests a bifurcation: the top tier is becoming more concentrated, while the middle tier is actively adding exposure. This is consistent with a market where institutional capital is flowing in through ETFs and custody solutions, while semi-professional investors are buying dips. The retail decline is either a redistribution or a reflection of changing behavior (e.g., using exchanges as custodians via spot ETFs rather than holding directly).
Contrarian
The primary blind spot in Santiment’s data is the entity identification heuristic. Most on-chain analytics platforms rely on the “common input” heuristic: if two addresses are inputs to the same transaction, they are likely controlled by the same entity. This works well for simple cases but fails for sophisticated actors who use coinjoin, privacy wallets, or cross-chain swaps. A whale using a Bitcoin mixer could appear as multiple independent addresses, skewing the count downward. Conversely, an ETF custodian that operates a single large cluster (e.g., 50,000 BTC across 10 addresses) may be counted as one entity, but the headline “addresses” count would show 10 addresses—inflating the whale count artificially.
Another blind spot: the ETF effect. The spot Bitcoin ETFs that launched in January 2024 now hold over 900,000 BTC. The custodians for these ETFs (Coinbase, Gemini, etc.) maintain segregated wallets for each fund. For example, BlackRock’s IBIT holds over 300,000 BTC spread across multiple addresses, some of which likely exceed 10,000 BTC. When a new ETF custodian adds a wallet with 10,000+ BTC, it registers as a new “whale address” in Santiment’s data, even though it’s not a new whale—it’s an administrative split. The increase from 84 to 90 elite addresses could easily be explained by ETF custodians creating new wallets for operational reasons, not by new billionaires entering the market.
Unintended consequences of misinterpreting this data: if traders buy BTC based on the “whale accumulation” narrative, they may be trading against the actual flow. The real sign of new money entering Bitcoin is the premium on Coinbase versus Binance, or the net inflow to spot ETFs, not the number of high-balance addresses. The address count metric is a lagging indicator—it only updates after on-chain activity, which itself is a delayed reflection of off-chain purchases. By the time Santiment reports the increase, the buying may have already been completed.
Takeaway
The 90-wallet ATH is a data point, not a thesis. The more reliable signal is the $1.5 billion accumulation by mid-level addresses, which suggests that a class of informed investors is building positions. But the elite whale count is a distraction—a metric that sounds impressive but lacks the resolution to distinguish between genuine accumulation and institutional infrastructure. The real question for the next six months: will the ETF custodians continue to consolidate their wallets, or will they fragment them further? The answer will determine whether the “90” becomes a new plateau or a temporary spike. And if the market wakes up to the fact that half of those addresses are corporate vaults, not individual whales, the narrative breaks faster than a UTXO split.
