Recently, I’ve been looking at some classic indicators in the crypto market, and honestly, I’m a bit shaken. The tools that were once treated as benchmarks are now almost all failing—and they’re failing in a strangely consistent way.

Let’s start with the most obvious phenomenon. The S2F model once predicted $500k, but now BTC is only a bit above $80k, with an error of more than 3x. The four-year cycle theory has been effective since 2012, yet this year it’s showing absolutely no response. Indicators like Pi Cycle Top, the Rainbow Chart, and the MVRV Z-Score—either they go quiet, or they give signals that are completely opposite. The most painful one is the Fear and Greed Index: it fell to below 10 in April 2025 (even lower than during the FTX collapse), but BTC didn’t rebound at all. The altcoin season has been slow to arrive, and BTC’s market share is still around 57%, completely breaking the kind of capital rotation pattern from the past.

This isn’t a problem with a single indicator—it's that the foundational assumptions behind the entire indicator system have changed.

Institutional capital entering the market has changed everything. After Bitcoin spot ETFs launched, they continued absorbing funds and broke the simple logic driven purely by halving cycles. Institutions tend to buy on dips and hold long-term, directly smoothing out the kinds of crazy volatility that retail investors used to create. When volatility drops from 100% to 50%, indicators that depend on extreme upside moves naturally become ineffective. Pi Cycle Top requires short-term moving averages to deviate significantly from long-term moving averages to produce a crossover signal; with volatility lower, this condition simply can’t be met anymore. The Rainbow Chart is the same—those fixed-width color bands no longer get touched.

There’s also a deeper change. Bitcoin is shifting from a “digital commodity” to a “macroeconomic financial asset.” In the past, indicators mainly looked at on-chain data and halving cycles, but now the price drivers have long shifted toward off-chain variables such as Federal Reserve policy, global liquidity, and geopolitical factors. MVRV Z-Score used to be considered overheated when it exceeded 7; now even at the historical high in October 2025, it’s only 2.69. Why? Because institutional buyers have purchased at high prices and held for the long term, systematically lifting the cost basis to near market value and compressing the range of volatility. High-frequency operations by short-term traders also keep “refreshing” the cost of active supply to current price levels, further shrinking the gap between market cap and cost basis.

The NVT ratio (the crypto version of the earnings multiple) is now sending signals that are completely the opposite. In April, before prices surged, NVT showed overvaluation; when prices reached 120k in October, it said undervaluation. The core problem is that on-chain transaction volume can no longer represent real economic activity. Trends like Layer 2 transactions, settlement within exchanges, and ETF custody are all eroding the representativeness of on-chain data.

The supply side also isn’t that important anymore. After the 2024 halving, Bitcoin’s annualized inflation rate fell from 1.7% to 0.85%, and the proportion of new supply each year relative to the total is less than 1%. Compared with a market value in the trillions of dollars, the real impact of this supply reduction is negligible. The S2F model is off precisely because it only looks at supply and completely ignores demand, leading to predictions that are wildly unrealistic.

The altcoin season index has been staying below 30 for a long time, and BTC Dominance has reached a peak of only 64%—far from the critical thresholds of past capital rotation. But this isn’t because funds are flowing into altcoins; it’s because institutional risk appetite for BTC is higher than for altcoins themselves. The incremental capital attracted by ETFs flows directly into BTC, and structurally, it doesn’t “rotate.” In addition, the siphoning effect of AI and precious metals also reduces the incremental inflow into the crypto market.

Retail sentiment is no longer a decisive force behind prices. The logic of the Fear and Greed Index is contrarian trading: buy when extremely fearful, sell when extremely greedy. But now, when retail investors are fearful, institutions may be buying the dip; when retail investors are greedy, institutions may be hedging with derivatives. The transmission mechanism between sentiment and price has been disrupted by institutional capital.

In the end, these indicators failing isn’t surprising. Most classic indicators are essentially curve fits based on 3–4 halving-cycle trends, with extremely small sample sizes. When the market environment undergoes a qualitative change—when the participant structure shifts from retail-led to institutional-led, and when the source of funds shifts from pure crypto believers to traditional financial institutions—tools trained on past conditions naturally stop working.

The signal being communicated is actually very simple: rather than obsessing over finding the next “universal indicator,” it’s more important to understand the assumptions and applicability boundaries of each indicator. Over-reliance on any single tool can lead to misjudgments. The underlying rules of the market are being rewritten, and staying cognitively flexible may be more practical than searching for a perfect predictive tool.