Business Predictability | Robinhodl21

Have you ever wondered why a company beats earnings estimates yet its stock barely moves—or even drops? It might be because the market already expected a beat. Companies that consistently outperform forecasts tend to attract higher expectations over time, so another “+20 % surprise” may no longer surprise anyone. In other cases, investors may weigh sales growth more heavily than earnings, especially in growth sectors where top-line momentum matters more than margin control.

This indicator was built to explore exactly that dynamic. It helps you quantify how predictable a business truly is, how consistently it beats (or misses) expectations, and how well management seems to understand and guide its own performance. It’s not a timing tool, but a quality lens for long-term stock pickers who want to identify stable, well-managed companies with disciplined forecasting and execution.

What the indicator is
its is designed to quantify how often and how well a company beats-or-misses expectations (earnings and sales) over multiple years, then map that into a “predictability” and “quantile” score that you can use to compare across stocks. Its core logic combines deviation from estimates, rolling history, and statistical ranking to highlight companies where the management and the business appear to be aligned, stable and reliable.

Key features:
(• Choice of financial data frequency: you can select FQ (quarterly) or FY (annual) mode so the indicator adapts to your preferred horizon.
(• Deviation calculation: earnings surprise and/or sales surprise can be combined via a weighted setup so you pick which metric drives the score.
(• History buffer: you choose how many “commit points” (i.e., past surprises) to include in the statistics and quantile calculations.
(• Quantile ranking: the tool computes how the company’s recent deviation stacks up versus its own history; in FY-mode we still use quarterly density for statistical robustness.
(• Predictability & volatility metrics: beyond the quantile, you get a predictability score (low recent deviation + low volatility) and a simple “moat” / management-quality overlay via the SLOAN ratio.
(• Status and CI table: the indicator comes with a visualization panel summarizing mean surprise, standard deviation, sample length, and your computed quantile and predictability grades.
(• Future box: optional forward-map showing the next earnings date, estimated deltas and flagged surprises.

What it is not
It is not a timing indicator (i.e., it won’t tell you when to buy or sell precisely). It does not predict short-term price movements. Instead, it is tuned for fundamental stock picking: look for companies that repeatedly deliver surprise results, for which you believe management and business model give an advantage. Use it to add an extra dimension of “earnings surprise stability & management forecasting quality” to your dashboard.

My usage case
I developed this indicator as part of a broader portfolio strategy: I screen for companies that are both highly predictable (i.e., rarely miss) and have the capacity to beat earnings by a meaningful margin, because I believe this reflects strong business execution and good internal alignment. Over time I plan to expand the dashboard with more indicators geared toward company quality and moat (quantitative metrics built from financial statement data). This version is still work in progress (there may be bugs), so consider the output as one more input—do not rely on it exclusively.

Important caveats
The code is relatively computation-intensive, especially with large lookback windows and quarterly frequency. On my Mac Pro it runs smoothly—but depending on your device and market data feed you may experience slower performance. Also: synchronising earnings release timing and sales release timing across companies is tricky—sometimes data lags or is updated later, so there may be discrepancies. Because of this the indicator’s output should be treated as a guide rather than a guarantee.

Empirical background
The academic literature supports the idea that consistent surprises and management execution can matter—but the relationship is complex. For example, research on post-earnings-announcement drift (PEAD) shows that markets often under-react to surprise earnings and that returns continue to drift in the direction of surprise for weeks or months.  At the same time, studies such as Skinner & Sloan (2000) show that when you control for growth expectations the relation of surprise to future returns becomes weaker.  In other words: just beating earnings by 20 % repeatedly does not guarantee outsized share-price gains, because market expectations adjust, estimates bake in the beat and other factors (discount rates, fundamentals) still dominate.

Final word
Use it as part of your fundamental stock-analysis toolkit to gauge how well a company consistently delivers relative to expectations, how volatile those surprises are, and whether you think management has a competitive edge in forecasting or executing. As mentioned, this is a work in progress and should not be your only tool—but used wisely, it can add a meaningful extra dimension to your decision-making. I’ll continue to improve it and add new quality-and-moat oriented indicators in future releases.