Analyst Forecast Accuracy: How Often Wall Street Gets It Right (2015-2025)

Lambda Finance compiled analyst forecast accuracy data across four dimensions—earnings estimates, price targets, S&P 500 year-end forecasts, and economic projections—using FactSet Earnings Insight reports, Federal Reserve research papers, Houlihan Lokey analyst estimate studies, and S&P Dow Jones Indices data. This dataset covers the period 2015 through 2025, aggregated in March 2026. Analyst forecast accuracy varies substantially by metric type: 77% of S&P 500 companies beat quarterly EPS estimates on average, but only 30% of 12-month price targets are reached at expiration, and Wall Street consensus S&P 500 year-end targets have missed by an average of 12.4 percentage points over the past decade. The tables below segment these accuracy rates by forecast type, sector, time horizon, and analyst tier.

1. Analyst Forecast Accuracy: Quarterly Earnings Estimates

The most widely tracked measure of analyst forecast accuracy is the quarterly EPS beat rate—the percentage of S&P 500 companies that report actual earnings per share above the consensus analyst estimate. The table below tracks this rate over the past six years using FactSet data.

Period	EPS Beat Rate	Revenue Beat Rate	Avg EPS Surprise	Avg Revenue Surprise
2019 Average	74%	57%	+4.8%	+0.9%
2020 Average	76%	68%	+17.1%	+2.5%
2021 Average	84%	71%	+15.8%	+3.4%
2022 Average	73%	68%	+3.6%	+1.7%
2023 Average	75%	62%	+5.9%	+1.4%
2024 Average	77%	64%	+6.8%	+1.3%
H1 2025 Average	80%	65%	+7.2%	+1.5%
5-Year Average	78%	65%	+6.4%	+1.7%
10-Year Average	75%	63%	+5.5%	+1.2%

The data reveals a structural pattern in analyst forecast accuracy: companies beat consensus EPS estimates approximately 75–78% of the time, with an average positive surprise of 5–7%. This persistently high beat rate is not evidence of analyst skill—it reflects a systematic downward bias. Analysts routinely lower estimates during the quarter (a process called “estimate walk-down”), making their final numbers easier to beat. FactSet data shows that consensus EPS estimates fall by an average of 2.9% during any given quarter over the 10-year average, and by 1.6% over the five-year average.

2. Analyst Forecast Accuracy by Sector

Analyst forecast accuracy varies dramatically across sectors. The table below shows the average absolute error between analyst earnings estimates and actual results for S&P 500 and Russell 3000 companies, segmented by GICS sector.

Sector	Avg EPS Error (1-Year)	Avg Revenue Error (1-Year)	EPS Beat Rate	Predictability Tier
Communication Services	4.2%	1.1%	80%	HIGH
Financials	5.1%	1.4%	79%	HIGH
Consumer Staples	5.8%	1.2%	78%	HIGH
Information Technology	7.3%	1.8%	81%	MEDIUM
Industrials	8.4%	2.1%	76%	MEDIUM
Consumer Discretionary	9.7%	2.4%	75%	MEDIUM
Materials	12.6%	3.1%	72%	LOW
Real Estate	11.8%	2.8%	70%	LOW
Healthcare	18.4%	3.1%	74%	LOW
Energy	42.7%	5.8%	68%	LOW

Communication Services and Financials produced the highest analyst forecast accuracy, with average EPS errors of 4.2% and 5.1% respectively. Energy was the least predictable sector at 42.7% average error—more than 10 times the error rate of Communication Services. This is driven by commodity price volatility making earnings inherently difficult to forecast. Revenue estimates were consistently more accurate than earnings estimates across all sectors, with the gap widest in Energy (5.8% revenue error vs 42.7% EPS error).

3. Analyst Forecast Accuracy by Time Horizon

Analyst forecast accuracy degrades significantly as the forecast window extends. The table below compares weighted average errors for earnings and revenue estimates across 1-year, 2-year, 3-year, and 5-year horizons for S&P 500 companies.

Forecast Horizon	Avg EPS Error	Avg Revenue Error	Directional Accuracy	Reliability
Current Quarter (90 days)	3.8%	1.0%	82%	HIGH
1 Year Forward	9.4%	2.8%	68%	MEDIUM
2 Years Forward	16.1%	5.3%	58%	LOW
3 Years Forward	24.7%	8.6%	52%	LOW
5 Years Forward	38.2%	14.9%	47%	LOW

Current-quarter EPS estimates have just a 3.8% average error—reasonable accuracy when management has already provided guidance. But by the 2-year mark, EPS error rises to 16.1%, and at the 5-year horizon, analysts miss by an average of 38.2% with directional accuracy dropping to 47%—effectively a coin flip. Revenue estimates degrade more slowly but follow the same pattern. This data suggests that any analyst forecast beyond 12 months should be treated as a directional signal at best, not a quantitative prediction.

4. Wall Street S&P 500 Year-End Forecast Accuracy

Each December, major Wall Street strategists publish S&P 500 year-end price targets for the following year. The table below tracks the consensus median forecast against the actual year-end value from 2015 through 2025.

Year	Consensus Target	Actual Year-End	Error	Direction Correct?
2015	2,215	2,044	+8.4%	Overshoot
2016	2,100	2,239	-6.2%	Undershoot
2017	2,356	2,674	-11.9%	Undershoot
2018	2,850	2,507	+13.7%	Overshoot
2019	2,900	3,231	-10.2%	Undershoot
2020	3,330	3,756	-11.3%	Undershoot
2021	4,000	4,766	-16.1%	Undershoot
2022	4,825	3,840	+25.7%	Overshoot
2023	4,050	4,770	-15.1%	Undershoot
2024	5,000	5,882	-15.0%	Undershoot
2025	6,600	6,468	+2.0%	Close

Over this 11-year window, the consensus was within 10% of the actual result in only 4 out of 11 years. Strategists undershot the actual result in 7 of 11 years, reflecting a persistent tendency toward conservatism. The largest miss was 2022 (+25.7% overshoot), when consensus expected continued gains but the S&P 500 fell 19.4%. The second-largest misses were 2021 and 2023, both undershoots exceeding 15%. Wall Street’s year-end target is better understood as a sentiment indicator than a forecast.

5. Analyst Price Target Accuracy

Individual stock price targets—the 12-month projections published alongside buy/sell ratings—show a different accuracy pattern than earnings estimates. The table below summarizes hit rates using multiple definitions of “accuracy.”

Accuracy Metric	Hit Rate	Notes
Stock reached price target at any point in 12 months	50–60%	Looser definition; benefits from intraperiod volatility
Stock at or above target at 12-month expiration	~30%	Strict definition; most common in academic studies
Directional accuracy (up/down correct)	50–60%	Comparable to chance for many horizons
Mean absolute error of target vs actual	30–40%	Average magnitude of miss (Taiwan market study: 39.1%)
Bias direction (overestimate vs underestimate)	~65% optimistic	Persistent upward bias; targets tend to exceed actuals
Buy ratings as % of all recommendations	~55%	Structural bias: banking relationships incentivize bullish coverage

Analyst price targets carry a persistent optimistic bias: approximately 65% of 12-month targets overshoot the actual stock price at expiration. Only about 30% of targets are achieved when measured strictly at the 12-month mark. Approximately 55% of all analyst ratings are “Buy” or equivalent—a structural consequence of the banking relationships that fund sell-side research. Academic research has found that sell recommendations carry more informational value than buys: the mean post-event price drift for a sell downgrade is -9.1% over six months, while buy upgrades produce only +2.4% drift.

6. Economic Forecast Accuracy: GDP and Interest Rates

Analyst forecast accuracy for macroeconomic variables—GDP growth, interest rates, and inflation—is typically worse than company-level estimates. The table below compares professional economic forecasters’ accuracy using Federal Reserve, IMF, and Survey of Professional Forecasters data.

Forecast Type	Mean Absolute Error	Within ±1 pp of Actual	Overestimate Bias
U.S. GDP Growth (1-year ahead)	1.0 pp	~55%	Slight optimistic
IMF GDP Growth (advanced economies)	1.3 pp	~42%	56% overestimate
IMF GDP Growth (developing economies)	2.1 pp	~30%	55% overestimate
Federal Funds Rate (1-year ahead)	0.8 pp	~60%	Mixed
CPI Inflation (1-year ahead)	1.2 pp	~48%	Underestimate during shocks
Unemployment Rate (1-year ahead)	0.6 pp	~70%	Slight pessimistic

The IMF was within a 0.1 percentage-point margin of error in only 6.1% of its GDP forecasts. Overall, IMF forecasts underestimated GDP growth in 56% of cases and overestimated it in 44% of cases. The St. Louis Fed’s research found that professional forecasters reported 53% confidence in their own accuracy, but were correct only 23% of the time—a significant overconfidence gap. Unemployment rate forecasts showed the highest accuracy, likely because labor market changes tend to be gradual and persistent.

7. The Estimate Walk-Down: Why Companies “Beat” Estimates

The high quarterly beat rate (75–78%) does not indicate analyst forecast accuracy in the traditional sense. The mechanism behind it—known as the “estimate walk-down”—explains why the beat rate is structurally inflated.

Metric	5-Year Average	10-Year Average	Implication
Consensus EPS decline during quarter	-1.6%	-2.9%	Analysts systematically lower estimates before reporting
% of quarters where consensus falls	~82%	~85%	Walk-down occurs in nearly every quarter
Avg positive EPS surprise at reporting	+6.4%	+5.5%	Walk-down creates a gap companies can “beat”
Companies providing guidance	~60%	~55%	Management guidance anchors estimates lower
Guidance-to-actual beat rate	~72%	~70%	Companies set beatable guidance targets

 

 

The walk-down mechanism works as follows: analysts begin the quarter with estimates anchored to management guidance and their own models. As the quarter progresses, analysts lower estimates by an average of 2.9% (10-year average). When the company reports, it “beats” the now-lowered bar. Both sides benefit: companies get positive headlines, and analysts maintain relationships. The Federal Reserve’s 2024 research paper demonstrated that these forecast errors are partially predictable using macroeconomic data—suggesting they are not purely random errors but contain systematic components.

8. Key Takeaways

• Analyst forecast accuracy is highest at short horizons. Current-quarter EPS estimates have a 3.8% average error and 82% directional accuracy. Beyond 2 years, directional accuracy drops to coin-flip levels (52%).
• The 77% EPS beat rate reflects systematic bias, not skill. Analysts lower estimates by an average of 2.9% during each quarter (the “walk-down”), creating a beatable bar. The beat rate is a feature of the system, not evidence of accuracy.
• Sector matters more than methodology. Communication Services and Financials have EPS error rates of 4–5%, while Energy averages 42.7% error—a 10x spread driven by commodity price unpredictability.
• S&P 500 year-end targets missed by 12.4 pp on average. Wall Street consensus undershot the actual result in 7 of 11 years from 2015-2025, with the largest misses occurring at market turning points (2022, 2023).
• Price targets hit their mark only 30% of the time using strict 12-month definitions. Approximately 65% of targets carry an optimistic bias.
• Revenue estimates are 2–3x more accurate than EPS estimates across all sectors and time horizons, because revenue has fewer discretionary accounting inputs.

Methodology

This analysis aggregates data from FactSet Earnings Insight weekly reports (2019–2025), Houlihan Lokey’s “Accuracy of Analyst Estimates” study (July 2024), Federal Reserve FEDS working paper 2024-049, IMF Working Paper WP/21/216, the Federal Reserve Bank of St. Louis professional forecasters analysis (December 2025), and S&P Capital IQ consensus data. EPS and revenue beat rates use S&P 500 company reporting data as compiled by FactSet. Sector accuracy data uses 10-year average absolute weighted errors for S&P 500 companies. S&P 500 year-end consensus targets represent the median of major bank strategist forecasts published in December of the prior year, sourced from Bloomberg and financial media compilations. Price target accuracy data draws on multiple academic studies including ScienceDirect (2024) and peer-reviewed journals. Data compiled March 2026 by Lambda Finance.

Sources