RSI Range Shift: Bull and Bear RSI Ranges Explained
September just closed as one of the rougher months of the year. Bonds posted their sharpest yield jump in a generation, a list of S&P 500 names finished the month deep in the red, and the market’s leadership narrowed to a handful of AI-linked stocks. That leaves a lot of charts with the same question on them: is this a pullback inside an uptrend, or the start of a downtrend? Most traders reach for RSI and look at whether it’s below 30. That’s the wrong question. The more useful one is where RSI has been bottoming and topping over the last few months, and whether that range just moved. That idea is called the RSI range shift, and this post covers how it works, how to classify the regime with plain rules, a worked example, and the mistakes that make it look better on a chart than it trades.
What is the RSI range shift?
The RSI range shift is the observation that RSI doesn’t oscillate across the full 0–100 scale evenly. In an uptrend, 14-period RSI tends to work in a band of roughly 40 to 80, and in a downtrend in a band of roughly 20 to 60. When RSI moves from one band to the other, the trend regime has likely changed.
J. Welles Wilder introduced RSI in 1978 with fixed 30 and 70 thresholds. Later practitioners, most prominently Andrew Cardwell and Constance Brown, noticed that those fixed lines fail in trending markets. In a strong uptrend, RSI rarely reaches 30 at all. The pullbacks bottom in the 40s and the stock resumes higher. In a downtrend, RSI rarely reaches 70; rallies stall in the 50s and 60s. Brown’s work describes the 40–50 zone as typical support in bull markets and 55–65 as typical resistance in bear markets. The exact numbers vary by source and by instrument, so treat them as zones, not lines.
Why it can produce an edge
The range shift is a way of measuring trend persistence with a momentum oscillator. RSI is a ratio of average gains to average losses. If buyers keep stepping in before the losses accumulate, RSI’s floor rises. That is momentum, and the time-series momentum effect, where assets that have been rising tend to keep rising over the next few months, is one of the better-documented patterns in markets. The range shift doesn’t create that effect. It gives you a cleaner read on whether it’s still present. The potential edge is in avoiding two expensive errors: buying “oversold” readings in a downtrend and selling “overbought” readings in an uptrend. More on the parent indicator in the RSI trading strategy guide.
Bull and bear RSI ranges: the reference table
A bull range and a bear range differ mainly in where RSI’s pullbacks end, not where its peaks are. The floor is the tell. A market can hit 70 in either regime, but a market in a genuine uptrend should not keep printing RSI in the 20s.
| Regime | Typical working range | Support zone | Resistance zone | What 30 means | What 70 means |
|---|---|---|---|---|---|
| Bull range | ~40–80 | 40–50 | 80+ (often stretches, not reversals) | Range break, regime warning | Normal strength |
| Bear range | ~20–60 | 20 or below (often stretches) | 55–65 | Normal weakness | Range break, regime warning |
| Transitional | 30–70 | none reliable | none reliable | Inconclusive | Inconclusive |
The practical rule is that the levels you expect RSI to bounce from depend on the regime you are in. A 42 reading is a buyable dip in a bull range and a mid-range nothing in a bear range.
How to classify the RSI regime with rules
You can turn the range shift into a mechanical classifier by tracking RSI’s extremes over a rolling window. The version below uses 14-day RSI and a 60-session lookback, roughly three months, which is long enough to cover two or three swings.
- Bull range: in the last 60 sessions, RSI has reached 70 at least once and has not closed below 40.
- Bear range: in the last 60 sessions, RSI has reached 30 at least once and has not closed above 60.
- Transitional: neither condition holds.
- Range shift: the label changes from bull to bear or from bear to bull. Many traders also flag a “warning” the first time RSI breaks the regime’s floor (below 40 in a bull range) or ceiling (above 60 in a bear range), before the full flip is confirmed.
Two design choices matter. First, use closing RSI values, not intraday spikes, so a single bad print doesn’t flip the label. Second, require both conditions in each test. A market that hasn’t touched 70 but also hasn’t fallen below 40 is drifting, not trending.
A worked example with plain numbers
Take a stock that rallied from $95 to $128 over the summer. Its 14-day RSI swings over the last few months looked like this:
| Swing | Price | RSI | Read |
|---|---|---|---|
| Peak 1 (July) | $118 | 74 | Reaches 70, consistent with bull range |
| Pullback 1 | $111 | 44 | Holds the 40–50 support zone |
| Peak 2 (August) | $128 | 78 | New high, RSI confirms |
| Pullback 2 (early Sept) | $120 | 46 | Holds support again — bull range intact |
| Drop (late Sept) | $109 | 33 | Closes below 40 — range warning |
| Rebound | $116 | 58 | Stalls under 60 |
| Next drop | $104 | 27 | Below 30 |
Read it in order. Through August the classifier says bull range: RSI hit 70 twice and its pullbacks ended at 44 and 46. The late-September close at 33 is the first warning. The floor broke. The rebound is the decisive test. In a bull range, a rebound from a pullback usually carries RSI back toward 70. This one stalls at 58, inside the bear-range resistance zone of 55–65, and price makes a lower high at $116 versus $128. The next leg down takes RSI to 27. Run the 60-session test now: RSI has reached 30 and hasn’t closed above 60 since the break. Once the July and August peaks age out of the window, the label flips to bear range.
Note what a fixed 30/70 trader would have done: bought the 33 reading in late September as “nearly oversold,” and again at 27. The range-shift trader treats both as confirmation that the regime has changed.
A rules-based template to test
A long-side template for a proper backtest:
- Regime filter: classifier reads bull range.
- Setup: RSI pulls back into 40–50.
- Entry: RSI turns up and price closes above the prior day’s high. Enter next open.
- Stop: below the pullback’s swing low.
- Exit: RSI closes below 40 (regime warning) or a trailing stop is hit.
Sizing with numbers: on a $50,000 account risking 1% per trade, the risk budget is $500. Entry at $120 with the swing low at $114 is $6 of risk per share, so the position is $500 / $6 = 83 shares. See position sizing for the full method.
Common mistakes with the RSI range shift
The most common mistake is treating the 40 and 60 lines as precise. They are zones derived from observation, and different instruments sit in different ranges. A low-volatility utility may work in 45–70 during an uptrend. A high-beta tech name can dip to 35 and still be in a bull trend. Calibrate the zones to the instrument by checking where its RSI pullbacks actually ended during past trends you can identify on price alone.
The other failure modes:
- Using it as an entry signal. A range shift labels the regime. On its own it tells you nothing about the next three days, so it needs a separate trigger and stop.
- Confirming RSI with RSI. Pairing the range shift with RSI divergence or stochastic RSI is one piece of evidence counted twice, since they all come from the same closes. Use something independent. For an index, check whether participation agrees with market breadth indicators. For a single stock, check the 200-day moving average or a higher-high/higher-low price structure.
- Ignoring lookback sensitivity. A 20-session window flips constantly. A 120-session window flips months late. Test several windows. If the edge exists only at exactly 60, you have found noise.
- Forgetting that RSI settings change the ranges. A 9-period RSI swings wider than a 14-period RSI, so 40/60 zones calibrated on one don’t transfer to the other.
- No out-of-sample check. Range rules are easy to fit to a chart after the fact. Validate them on data you didn’t look at while building them; see out-of-sample testing.
FAQ
What is a bullish RSI range?
A bullish RSI range is when 14-period RSI works roughly between 40 and 80, with pullbacks bottoming in the 40–50 zone instead of reaching 30. It shows buyers consistently stepping in before losses build up. The range persists as long as RSI keeps holding the low 40s on pullbacks and reaching 70 on rallies.
What does it mean when RSI stays above 40?
An RSI that holds above 40 through several pullbacks is a sign of a persistent uptrend. Average losses never grow large relative to average gains. Traders using range rules treat each test of the 40–50 zone as potential support. The first close below 40 is a warning that the bull range may be ending, not an oversold signal.
Is the RSI range shift the same as RSI divergence?
No. Divergence compares RSI’s direction to price’s direction over one or two swings, such as price making a higher high while RSI makes a lower high. A range shift looks at where RSI’s floor and ceiling sit over several months. They often appear together near trend changes, but a range shift can occur without divergence and vice versa. Hidden divergence is the closest relative, since it also focuses on pullback lows within a trend.
Does the RSI range shift work on weekly charts?
Yes, and many practitioners prefer it there. Weekly RSI changes regime less often, so each shift carries more weight and produces fewer false flips. The trade-off is lag: a weekly range shift can confirm weeks after the daily chart showed the warning. A common approach is to use the weekly range for the regime label and the daily chart for timing.
Key takeaways
- RSI tends to work in roughly 40–80 during uptrends and 20–60 during downtrends; the fixed 30/70 levels mean different things in each.
- The floor is the tell. Repeated pullbacks holding 40–50 signal a bull range; rallies stalling at 55–65 signal a bear range.
- Turn it into rules with a rolling lookback: reached 70 and no close under 40 is bull, reached 30 and no close over 60 is bear.
- A range shift is a regime label. It needs a separate entry trigger, a price-based stop and position sizing.
- Calibrate zones per instrument and RSI length, confirm with something independent of RSI, and validate out of sample before trading it.