Stats › Valuation › Analyst accuracy
Analyst forecast accuracy
Every week someone tells you where the S&P 500 will be in twelve months, or what a company is going to earn next quarter. Those numbers come from sell side analysts, and they get repeated everywhere as if they were measurements rather than estimates. Which raises the obvious question nobody seems to ask out loud: how have the estimates actually turned out?
This page collects what has been measured. The aggregated twelve month price target and its record against what the index really did. The quarterly earnings estimates and the well documented habit of trimming them just before companies report, which is most of the reason you keep hearing that the overwhelming majority of companies beat. The long horizon growth forecasts, where the gap is much wider. And the rating split, where sell ratings have all but stopped existing. I put it together because analyst numbers are among the most quoted and least audited figures in the market. Not because analysts are stupid, they aren't, but because the incentives around the job point in one direction and you can see that pointing in the data. Knowing the size and the direction of the bias is far more useful than arguing about whether there is one.
Five sources at the bottom, each with its date. Two of them disagree with each other, and where they do I've said so rather than quietly picking the number that reads better.
Analysts' aggregated twelve month price target for the S&P 500 stood at 9,060.03 on 31 July 2026 against an index level of 7,437.63, implying 21.8%. Their record on that forecast over 2005 to 2024 is an average overestimate of 5.9%, though they have underestimated in five of the last six years. On earnings the bias is larger and better documented. Over 1984 to 2006 analysts forecast long term earnings growth of 14.7% a year against 9.1% actual, and forecast a decline less than 1% of the time while 31.1% of companies actually delivered one. And estimates are systematically walked down before each quarter, which is why the beat rate is so high.
Members only
The rest of this page is for members
Below this point there are 9 sections, 1 chart and 5 named sources, roughly 2000 words of it. Every figure carries the source it came from and the date the data is from.
You can keep browsing every statistic in the library for free. The intro and the summary are always open.