Asymmetric Beta Dashboard
503 securities
A Gyre Research study of beta across benchmarks and time frames
This study set out to test two hypotheses about asymmetric beta in S&P 500 stocks: that a stock’s asymmetry lasts long enough to exploit in a long/short equity portfolio, and that short-term asymmetry reverts in a way that can be traded. We tested both on S&P 500 history back to 1992, survivorship-free, with every test fixed before it ran, and published the results in full. Both hypotheses were rejected.
What we continue to publish, and why. The study’s daily calculation stays available as a research tool. For any S&P 500 stock you hold, see its beta measured against the benchmark that matches your mandate — SPY, the equal-weight S&P 500, the cap-weighted S&P 500 rebuilt from each day’s members, its own sector, or its own size tier — and over the time frame you care about, from 10 days to 2 years. Up- and down-beta are shown alongside, with their 95% intervals.
Read the study: what we asked, what we did and what we found The tests
Daily digest
Interested in receiving daily results, sign up here!
Custom & data feed
For additional universes, or receiving daily results in machine-readable format? Contact us at team@gyreresearch.com
Latest in Markets and Macro: No released issue covers Asymmetric Beta yet. Read Markets and Macro
Beta against the benchmark and time frame you choose
HIGH Highest beta (top 10)
| Symbol | Sector | Beta | Up Beta | Down Beta |
|---|
LOW Lowest beta (top 10)
| Symbol | Sector | Beta | Up Beta | Down Beta |
|---|
Overall beta uses every day of the time frame; up beta and down beta use the days the benchmark rose and the days it fell, and asymmetry is up beta minus down beta, a description of the time frame measured (the study tests how much of it lasts). ± is the half-width of the 95% confidence interval: up beta 1.242 ±0.221 means the estimate is 1.242 and the interval runs 1.021 to 1.463. A wide interval is a beta estimated from few observations — the number is still shown, because withholding it tells the reader less than showing it with its precision. Tail beta is beta measured only on days the benchmark fell by more than one standard deviation. It is published where at least 20 such days fall inside the window; in the newest data of SPX, the S&P 500 index as measured by SPY that is the 2y and 1y lookbacks. Other windows show “-” because the sample does not support an estimate rather than because the security has no tail behaviour.
| Symbol | Sector | Industry | Cap Tier | Window | Benchmark | Overall Beta | Up Beta | Down Beta | Asymmetry | Spread % | Tail Beta | Tail Obs | R² | Obs |
|---|
Backtest: six claims about asymmetric beta, tested daily
Beside the study, this backtest tests six claims about asymmetric beta, every day.
Every month since the first month with two years of history we took the companies that were in the S&P 500 on that day (including those later removed, acquired or bankrupt), computed their up-beta, down-beta and asymmetry exactly as this page does (252 trading days, against the index as it stood then), and then looked at what happened next. The questions, the measures and the pass rule were fixed and published before the first run (method), and nothing was tuned. It is re-run every day.
How this relates to the study. A Supported verdict means an effect at least 2 standard errors from zero, not one large enough to trade. In the study, only about 6% of a stock’s own asymmetry carries into the next year, and both hypotheses it tested were rejected (the study, its tests).
How to read a verdict. Supported: the effect has the sign the claim states, is at least 2 standard errors from zero over the whole history (Newey-West t-statistic ≥ 2, which allows for overlapping periods), and has the same sign in 2021 onward, a period held back from the design. Not supported: weaker than that. Contradicted: at least 2 standard errors in the opposite direction.
| Claim tested | What was measured | Whole history | t | 2021 onward (held out) | Verdict |
|---|
Growth of 1: highest vs lowest asymmetry quintile, monthly
Equal-weight portfolios re-formed each month end, entered at the next day's close. The shaded band is where 90% of 1,000 portfolios of the same size, picked at random from the same members, ended up: a line inside the band is what luck alone produces.
Growth of 1: highest vs lowest down-beta quintile, monthly
The same construction for down-beta (claim B4, a downside-beta premium), with the same random band.
Does asymmetry persist? Rank correlation with the next year
Each point: across the members at a month end, the rank correlation between their asymmetry then and their asymmetry measured on the following 252 trading days. 0 means today's ranking says nothing about next year's; 1 means it is identical.
By period, costs, data coverage and the live record
| Quintile spread | Gross, per year | Net at 10 bp | Net at 25 bp | One-way turnover per month | Break-even cost | Top quintile beat this share of random portfolios |
|---|
| Year end | Index members | With price history | Share of index weight |
|---|
Limits. Each company carries one sector for its whole history. The last-day move of a company that is bought out or goes bankrupt is often missing from price data; such a holding earns its return to its last price and then 0%. Membership changes are seen at the index snapshot dates (quarter-end before 2005, month-end after). Coverage before 2000 is below 100% (table above). Trading costs are fixed estimates. Past behaviour is not a promise, and nothing here is investment advice.
Technology leads the highest-beta names; energy names dominate the lowest, as market-wide averages hold steady on November 11, 2026.
On November 11, 2026, this page covers 499 S&P 500 stocks against the S&P 500 index as measured by SPY over a one-year lookback, with prices through November 10. The mean beta is unchanged from the previous day at 0.693, the mean down beta edged up from 0.698 to 0.699, and the mean up beta rose from 0.636 to 0.649, while the median beta slipped from 0.542 to 0.539. The highest-beta list is dominated by technology, with Bloom Energy (BE) at the top, and the lowest-beta list is dominated by energy. The page's tests support the persistence of asymmetry, down beta, and tail beta into the next year, but not the claims that high asymmetry or high down beta earn more the next month.
These figures describe beta over the past year's lookback, with prices through November 10, and the page's own study found that asymmetry does not predict returns or persist strongly enough to trade, so none of this is a signal.
This analysis was created with the assistance of large language model technology. It is written from this page's numbers for the day and describes the results; it is not a forecast or investment advice.
Past daily commentary