Do RSI, MACD and other indicators predict returns?
By Mehmet Saruhan, Founder·Opulence Alpha Research · Published · Data through
Only weakly, and as short-run reversal rather than momentum. Over the next week, US stocks with RSI(14) below 30 beat the median stock 51.6% of the time and those above 70 only 48.5%; a close above the upper Bollinger band 48.6%, a new 10-day high 48.7%. Breakouts were followed by weaker weeks, not stronger ones. MACD and golden crosses showed no reliable edge.
Keywords: does RSI work, RSI oversold backtest, MACD crossover evidence, golden cross does it work, Bollinger bands backtest, technical indicators evidence·JEL classification: G11, G12, G14, C12, C58·Concepts: Multiple testing, Point-in-time data, Backtest overfitting
- Universe
- The same for every study: 1,767 US common stocks in the 11 GICS sectors, 422 of them since delisted; S&P 500 members since 1996 plus large and mid-sized companies outside the index
- Period
- Every Wednesday from 4 Jan 1995 to 19 Aug 2026: 1,634 Wednesdays across 7,959 trading sessions
- Sample
- n = 1,763 stocks of the 1,767-stock universe; 1,889,758 stock-weeks, a median of 1,148 stocks per Wednesday
- Outcome
- Return over the next 1, 5, 21 and 63 trading sessions against the same day's median stock
- Inference
- t on non-overlapping dates; proven only when |t| ≥ 3 and the same sign in at least 60% of years
1Introduction
RSI, stochastics, Bollinger bands, MACD and moving-average crosses appear on almost every trading screen, each with a settled reading: an oversold or overbought extreme marks a turn, and a crossing or a breakout marks the start of a move. Academic tests of simple trading rules, from Brock, Lakonishok and LeBaron onward, found some predictive power in older index data, and Park and Irwin's survey found that evidence weaker in later data and open to data snooping; Jegadeesh documented that a stock's return over one month tends to reverse in the next. This study measures each common reading on one universe, every Wednesday since 1995, against the same day's median stock, and asks which way each leaned and whether the lean held from year to year.
2Data and method
2.1Sample design
Population and frame. The population is US common stocks listed on the NYSE and Nasdaq; funds, ETFs, trusts, preferred shares, warrants and units are excluded. The sampling frame is a fixed universe of 1,767 companies, drawn once when the platform was built and not re-sampled since, in two strata. Stratum 1 is a census of the S&P 500: every company in the index at any time since 1996 whose price history could be recovered, 1,027 companies of which 413 have since delisted; it holds 76% of the index's members in 1996 and at least 96% in every year from 2010. Stratum 2 is 740 large and mid-sized companies outside the index, selected in proportion to the market's sector weights from the stocks that passed a minimum share price of $15 and a minimum average daily trading value of $25 million; 9 of them have since delisted.
Sample. The unit of observation is a stock-week: one stock on one Wednesday. A stock enters a Wednesday's cross-section when it has a valid close that day, a value of the signal and a measured outcome; bars flagged as bad data and returns that cross a change of issuer are left out. This study's sample is n = 1,763 stocks of the 1,767: 1,889,758 stock-weeks on 1,634 Wednesdays from 4 Jan 1995 to 19 Aug 2026, a median of 1,148 stocks per Wednesday (range 820 to 1,371). 331,954 stock-weeks (17.6%) come from the 422 companies that have since delisted. Every stock and every Wednesday carries equal weight.
Representativeness. Table 1 gives the sample by GICS sector beside the S&P Composite 1500: 7.9% of companies would have to change sector for the two to match exactly. By latest market value, 55% of the listed companies with a recorded market value are large (at least $10bn), 38% mid ($2–10bn) and 6% small. Because Stratum 2 was chosen from companies listed at construction, its history carries survivorship bias. Section 4 repeats the lead result among the stocks that were S&P 500 members on each date, the part of the sample largely free of that bias.
Companies in this study's sample, of them those since delisted, the size of the listed companies with a recorded market value, the sample's share of stock-weeks and the sector's share of the S&P Composite 1500.
| Sector | Sample | Delisted | Large | Mid | Small | Stock-weeks | S&P 1500 |
|---|---|---|---|---|---|---|---|
| Information Technology | 282 | 67 | 122 | 76 | 18 | 13.9% | 12.7% |
| Financials | 256 | 60 | 114 | 74 | 8 | 14.6% | 17.2% |
| Industrials | 255 | 53 | 126 | 67 | 11 | 15.9% | 17.5% |
| Health Care | 237 | 47 | 92 | 81 | 16 | 12.4% | 10.9% |
| Consumer Discretionary | 209 | 39 | 69 | 87 | 12 | 12.3% | 12.9% |
| Energy | 105 | 30 | 39 | 31 | 5 | 5.9% | 4.7% |
| Consumer Staples | 101 | 38 | 40 | 17 | 5 | 5.9% | 4.9% |
| Materials | 94 | 34 | 35 | 24 | 2 | 5.6% | 5.1% |
| Real Estate | 84 | 10 | 40 | 32 | 2 | 5.7% | 6.9% |
| Communication Services | 80 | 31 | 28 | 13 | 5 | 3.9% | 3.3% |
| Utilities | 60 | 13 | 35 | 12 | 0 | 4% | 4% |
| All sectors | 1,763 | 422 | 740 | 514 | 84 | 100% | 100% |
Sample: n = 1,763 of the 1,767 companies. Size by latest market value for the 1,338 listed companies with one recorded: large ≥ $10bn, mid $2–10bn, small < $2bn. S&P 1500 shares count the constituents of the S&P 500, MidCap 400 and SmallCap 600 (1,506 companies, lists read 25 Sep 2026). Sectors are each company's current GICS sector.
2.2Signals
| Signal | Definition | Since | Wednesdays |
|---|---|---|---|
| RSI(14) below 30 | The 14-session relative strength index under the usual oversold line | 4 Jan 1995 | 1,616 |
| RSI(14) above 70 | The 14-session relative strength index over the usual overbought line | 4 Jan 1995 | 1,617 |
| Above the upper Bollinger band | The close sits above the upper band, the usual overbought reading | 4 Jan 1995 | 1,630 |
| New 10-day high | The close breaks above the last 10 sessions' high, counted on the day it happens | 4 Jan 1995 | 1,631 |
| MACD crosses up | The MACD line crosses above its signal line, counted on the day it happens | 4 Jan 1995 | 1,633 |
| Golden cross | The 50-day average crosses above the 200-day, counted on the day it happens | 4 Jan 1995 | 1,328 |
| RSI(14) | The 14-session relative strength index, ranked across all stocks each Wednesday | 4 Jan 1995 | 1,634 |
| MACD histogram | The MACD line minus its signal line, in price units, ranked across all stocks each Wednesday | 4 Jan 1995 | 1,634 |
2.3Measurement
Outcome. The return from the close on the Wednesday to the close 1, 5, 21 and 63 trading sessions later, on closes adjusted for splits and dividends, compared with the same day's median stock; half of all stocks beat the median by construction, so chance is 50 in 100 on every date. No delisting return is added.
Each Wednesday, the stocks showing the reading are scored against the same day's median stock. A state counts on every Wednesday it holds; a cross or breakout counts only on the Wednesday it happens.
Each Wednesday the stocks are ranked on the signal. The study reports how often each fifth of that ranking beat the median stock, and the rank correlation (IC) between the signal and the return that followed.
The t-statistic uses non-overlapping dates only. A result is called proven when |t| is at least 3, it held in at least 60% of years and it is large enough to matter; with 1,470 tests across the studies, a looser bar would pass dozens by luck.
Full data and methods3Results
Oversold and overbought readings — the next week
How often a stock showing each reading beat the median stock on the same day over the next five trading sessions. The dashed line is chance (50). A filled square is a result that held up across the years (t-statistic at least 3); hollow means no measurable edge or only a faint one.
Crosses, breakouts and trend readings — the next week
Signals built to follow a move — MACD crossing up, a new 10-day high — were followed by slightly weaker weeks, not stronger ones. Golden and death crosses are rare and showed no reliable edge at any horizon.
How the RSI tilt changes with the horizon
The tilts are clearest over about a week — the only horizon where both held up across the years — and are gone by three months. Overbought readings already lag over the next session; oversold ones take about a week to show.
Across the whole range, not only at the extremes
Split every US stock into fifths by each indicator, every week. Over the next week the highest fifth beat the median stock less often than the lowest fifth for every oscillator here — the same reversal as the zones above, across the whole range. Three-month momentum, by contrast, showed no measurable edge.
How big is it?
Small. A stock with RSI(14) below 30 ended the next week 0.71 points above the median stock on average — a tilt across hundreds of names, and only a nudge for any one of them. Trading costs and the next day's gap can take all of it.
Short-term reversal: the relation underneath these readings4Robustness
Table 3 repeats the rank correlation for both indicators over the next week. The higher the RSI(14), the weaker the week that followed: -0.016 (t -5.0) across all stocks, with the same sign in 11 of 11 sectors and 3 of 3 decades, but short of the bar among S&P 500 members on the date (-0.011, t -2.8). The MACD histogram, usually read as a sign of upward momentum, leaned the same way: -0.018 (t -6.9) across all stocks and -0.013 (t -3.9) among S&P 500 members, with the same sign in 11 of 11 sectors and 3 of 3 decades. Both were strongest in the first decade and much smaller since; in the latest decade the RSI tilt can no longer be told from chance.
Rank IC and its t-statistic on non-overlapping dates.
| Sample | IC | t | Wednesdays |
|---|---|---|---|
| RSI(14), next week | |||
| All stocks, whole period | -0.0162 | -5.0 | 1,633 |
| By decade: 1995–2004 | -0.0331 | -6.5 | 516 |
| By decade: 2005–2014 | -0.0110 | -2.2 | 518 |
| By decade: 2015– | -0.0062 | -1.0 | 599 |
| S&P 500 members on the date | -0.0113 | -2.8 | 1,581 |
| Sectors with the overall sign | 11 / 11 | ||
| MACD histogram, next week | |||
| All stocks, whole period | -0.0180 | -6.9 | 1,633 |
| By decade: 1995–2004 | -0.0300 | -6.8 | 516 |
| By decade: 2005–2014 | -0.0152 | -3.7 | 518 |
| By decade: 2015– | -0.0101 | -2.1 | 599 |
| S&P 500 members on the date | -0.0134 | -3.9 | 1,581 |
| Sectors with the overall sign | 11 / 11 |
5Limitations
- Each indicator is read at its textbook setting and thresholds; other settings were not searched, since every added setting is one more test that could pass by luck.
- Crosses and breakouts count only on the Wednesday they happen, so the sample holds fewer of them than a daily count would.
- The MACD histogram is measured in price units, as charting screens print it, not scaled by the share price, so ranking it across stocks mixes the strength of a move with the level of the price; its result is read with that in mind.
- Before costs. Averages exclude trading costs, taxes and market impact.
- Same-close timing. Returns start at the close the signal is computed from; a real trade would start later.
- Survivors among smaller companies. The non-index names were chosen from companies listed when the universe was built; no delisting returns are added.
6Conclusion
The common indicators carried one small, consistent tilt, and it was short-run reversal. Stocks with RSI(14) below 30 did slightly better over the next week, and stocks above 70 or above the upper Bollinger band slightly worse: the direction the oversold and overbought readings suggest. A breakout to a new 10-day high, read as the start of a move, went the other way: it was followed by a slightly weaker week, not a stronger one. A MACD crossing up and a golden cross showed no reliable edge. The tilts fade by three months, have shrunk since the first decade and, for any one stock, are a nudge that is small next to trading costs.
References
- Brock, W., Lakonishok, J. and LeBaron, B. (1992). Simple Technical Trading Rules and the Stochastic Properties of Stock Returns. Journal of Finance 47(5).
- Park, C.-H. and Irwin, S. H. (2007). What Do We Know About the Profitability of Technical Analysis? Journal of Economic Surveys 21(4).
- Jegadeesh, N. (1990). Evidence of Predictable Behavior of Security Returns. Journal of Finance 45(3).
- Benjamini, Y. and Hochberg, Y. (1995). Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society, Series B 57(1).
- Grinold, R. C. and Kahn, R. N. (2000). Active Portfolio Management, 2nd ed. McGraw-Hill.
- Harvey, C. R., Liu, Y. and Zhu, H. (2016). … and the Cross-Section of Expected Returns. Review of Financial Studies 29(1).
- Shumway, T. (1997). The Delisting Bias in CRSP Data. Journal of Finance 52(1).
Appendix A. Questions readers ask
- Does RSI work?
- Measured on US stocks since 1995, RSI(14) below 30 was followed by a slightly better week — 51.6% of such stocks beat the median stock — and above 70 by a slightly worse one (48.5%). It is a small mean-reversion tilt, not a forecast, and it fades by three months.
- Does a MACD crossover predict returns?
- Not in this data. On the day MACD crossed above its signal line, 49.3% of stocks beat the median stock over the next week — a faint lag, not an edge.
- Do golden crosses work?
- After a golden cross (the 50-day average crossing above the 200-day), 49.7% of stocks beat the median stock over the next month; after a death cross, 49%. Neither is a measurable edge.
- Do breakouts to a new high keep going?
- Not over the next week: a close at a new 10-day high was followed by 48.7% of stocks beating the median stock — slightly worse than chance.
- What about Bollinger bands?
- A close above the upper band was followed by 48.6% of stocks beating the median stock over the next week; a close below the lower band by 50.9%. Both point the same way as RSI: short-run reversal.
Data availability and citation
Every figure in this paper is quoted from one frozen snapshot, published as JSON with the sample description, the robustness results and the test counts. The same snapshot feeds the research console, so the two cannot disagree. The paper and the snapshot are licensed under CC BY 4.0: share and adapt them with credit to Opulence Alpha and a link to this page.
Mehmet Saruhan (2026). Do RSI, MACD and other indicators predict returns? Opulence Alpha Research, Opulence Alpha Studies, 25 Sep 2026. https://opulencealpha.ai/studies/technical-indicators
Research, not advice. A measured tendency across hundreds of stocks is a nudge for any one of them, never a forecast of its price.
More evidence
All studiesWhere this is read on the S&P 500 company pages
The company pages show each S&P 500 company's readings from the latest close; a ranking lists every company by one measure.