Sesgo de Recencia
EN: Recency Bias PT: Viés de Recência
La tendencia a dar peso desproporcionado a eventos recientes al predecir futuro. Traders sobreestiman la continuidad de tendencias actuales — bulls at market peaks, bears at market bottoms. Todo el mundo es "growth investor" en 2021 y "value investor" en 2022.
Qué es el Sesgo de Recencia
El Sesgo de Recencia (Recency Bias, en portugués Viés de Recência) es la tendencia psicológica a dar peso excesivo a eventos recientes al predecir el futuro, mientras underweighting historical patterns o long-term trends. Simply: traders extrapolate recent performance indefinitely. Bull market continues? Expect more bulls. Bear market continues? Expect more bears. Tech outperforms? Assume forever. Emerging markets lag? Abandon. Real estate rising? Buy more. Each extrapolation ignores historical base rates. El bias opera en múltiples timeframes: Intraday: stock down 3% today? Trader expects more downside, sells. Often bottom of daily move. Weekly: rally 10% this week? Trader expects continuation, buys. Often peaks within days. Monthly: sector outperforms 20% this month? Trader chases, misses reversal. Annual: 2021 tech surge → FOMO buyers enter late 2021 → lose 50% in 2022. Academic research: Kahneman y Tversky (1974 y related work) documented recency bias via availability heuristic — easily recalled events (recent) feel more probable. Benartzi y Thaler (1995) showed recency bias affects long-term investors, reducing returns significantly. Famous examples: Japanese investors 1989: extrapolated Nikkei rise indefinitely. Peak at 39K, subsequent 30-year bear. US investors 1999: dotcom peak, extrapolated tech forever. Nasdaq crashed 78%. Real estate 2006: "Prices always rise." 2008 housing crash. Crypto 2021: "Bitcoin to $100K." Subsequent -77% to $15K. Pattern universal: extrapolation of recent trends leads to peak-buying or capitulation-selling.
Neuroscience y Evolution
El recency bias has deep evolutionary roots. Early humans benefited from pattern recognition — if predators present recently in area, expect more. If food abundant recently, expect continuity. This worked in stable environments. Financial markets are emphatically NOT stable environments — they're mean-reverting over time with periodic structural shifts. Same cognitive mechanism that kept ancestors alive costs modern investors. Neurologically: amygdala y hippocampus interact during memory encoding. Emotional recent events encode strongly. Non-emotional historical data fades. Brain's "current probability" estimates heavily weight recent emotional experiences. Availability heuristic: Kahneman's concept that easily recalled events seem more probable. Financial crises recent: remembered vividly. Long bull markets: fade into "normal." Effect: investors underestimate frequency y severity of crises después de long bull runs. Conversely overestimate recovery difficulty during crises (forgetting how many recoveries historically happened). Modern media amplification: 24/7 financial news emphasizes recent events. CNBC, Bloomberg TV constant emphasis on "today's action," "this week's movers," "most active." Historical context rarely featured. Social media especially recency-biased — algorithms promote current engagement. Creates investor conditioning for recency-biased decisions. Trading psychology studies: retail investor surveys show consistent pattern. After 6-month bull market, >70% expect continuation. After 6-month bear, >60% expect continuation. Reality: mean reversion is the historical norm. Extended trends are exceptions. Hedge fund positioning data: Commitment of Traders reports reveal professional funds often positioned opposite to retail recency bias. Pros fade recent trends; retail chases them. Asymmetric outcomes over long periods.
Cómo se Manifiesta en Trading
Recency bias manifestations en trading: (1) Trend following late: chasing established trends. Momentum investors enter trends late due to recency bias. Professional momentum strategies use rules-based filters, not emotional extrapolation. (2) Mean reversion underutilization: reversions to historical means are frequent but feel "unlikely" recency-biased. Missing reversal opportunities. (3) Sector rotation mistakes: abandoning underperforming sectors just before rotation into them. Late rotation into outperforming sectors just before rotation out. Retail investors consistently buy high, sell low via recency bias. (4) Volatility mispricing: after low-volatility periods (2017, 2019), traders sell volatility aggressively. Then volatility spikes (Vol-mageddon 2018, COVID 2020). After high volatility, traders overpay for protection. (5) Strategy switching: strategy underperforms 6 months? Abandon for new strategy. Just as old strategy begins performing. Chasing strategies that worked recently. (6) Risk assessment: post-calm period (2017), investors underestimate crash risk. Post-crisis (March 2020), overestimate continued risk. Both wrong. (7) Valuation anchoring to recent prices: stock at $100 two months ago, now $120. Feels "expensive." But if fundamentals deteriorating, could be $80 soon. Anchoring to recent price vs. current fundamentals. (8) Earnings expectations: company beats earnings 3 quarters running? Traders expect continuity. Miss follows, surprised. (9) Economic forecasts: recent GDP growth extrapolated. Recent inflation extrapolated. Recessions surprise repeatedly. Fed reacts late. Same mistake financial media makes. (10) Position sizing errors: after winning streak, size up (confident from recent wins). Next big loss wipes recent gains plus more. Opposite during losing streaks.
Contramedidas y Best Practices
Countering recency bias requires systematic approaches. (1) Historical context practice: before decisions, research long-term base rates. "How often does this happen over 50 years?" Forces long-view thinking. (2) Mean reversion positioning: at sentiment extremes (bull or bear), take contrarian position. Extreme investor sentiment surveys (AAII, II), Put/Call ratios, VIX levels — all provide quantified extremism measures. (3) Fundamental analysis discipline: evaluate companies/markets based on fundamentals, not recent performance. Valuation metrics (P/E, P/B, yield) vs. historical averages. (4) Time-diversified decisions: rebalance periodically, not reactively. Quarterly rebalancing ignores recent noise, respects long-term allocation. (5) Long-term backtests: strategies should be evaluated across multiple market regimes (30+ years if possible). Strategy looking brilliant in 3-year bull market often fails in bear market. (6) Opinion filtering: analyst recommendations heavily recency-biased. Most "buy" ratings near tops, "sell" near bottoms. Ignore herd sentiment; focus on individual analysis. (7) Documented strategy: written trading plan defines decisions. Prevents reactive changes driven by recent events. (8) Warren Buffett rule: "Be fearful when others are greedy" — explicit anti-recency positioning. When market sentiment extremely positive, reduce exposure. When extremely negative, increase. (9) Sentiment indicators: track and respect: VIX (fear index), put/call ratio (options positioning), AAII Sentiment Survey (retail bullishness), BofA Bull & Bear Indicator (positioning). Extreme readings typically reverse. (10) Patience cultivation: some of the best trades require waiting for mean reversion. Buffett holds cash during euphoric periods, deploys at troughs. Template for individual investors. Options-specific: recency bias affects IV pricing. After low-vol periods, IV too low — buy protection cheaply. After crisis spikes, IV too high — sell premium. Systematic contrarian IV positioning often profitable. Critical insight: financial markets are mean-reverting in valuation over long periods. Trends extend further than expected, but eventually revert. Recency bias leads investors to trade opposite of mean reversion — buying tops, selling bottoms. Discipline to trade WITH mean reversion (contrarian to recent trends) is professional money management core skill.
Recency Bias vs. Legitimate Trend Recognition
Process differs despite potentially similar outcomes.
| Aspect | Recency Bias | Legitimate Analysis |
|---|---|---|
| Basis | Emotional extrapolation | Systematic indicators |
| Entry timing | Late, near peaks | Early, confirmed signals |
| Exit planning | None defined | Pre-determined criteria |
| Position sizing | Escalating with wins | Rule-based consistent |
| Reversal awareness | Dismissed | Acknowledged, monitored |
| Historical context | Ignored | Referenced regularly |