RSI Beyond Overbought and Oversold

[Fairman]

RSI Beyond Overbought and Oversold

Many beginners learn RSI as a simple rule: above 70 means overbought and sell, below 30 means oversold and buy. In trending markets, that rule often loses money, because price can stay overbought for a long time.

diagram.png (37.29 KiB) Viewed 81 times

The Relative Strength Index (RSI) measures momentum on a scale from 0 to 100. It can offer more insight when used differently.

Better ways to use it:

1. Trend confirmation. In strong uptrends, RSI often stays between 40 and 90, with dips toward 40 to 50 acting as support. In downtrends, it tends to hold between 10 and 60.

2. Divergence. If price makes a higher high but RSI makes a lower high, momentum may be weakening. This works best at key levels and after liquidity sweeps.

3. Range conditions. In sideways markets, overbought and oversold readings can be more useful for mean reversion.

4. Momentum shifts. A move above 50 or below 50 can hint at a change in control.

How to apply it:

- Identify the market condition first
- Use RSI as a confirmation tool, not a trigger on its own
- Combine with structure and liquidity
- Test any rule you plan to use

Warning: indicators describe the past. If your analysis relies entirely on RSI, you are always a step behind.

Use it to support decisions, not to make them.

[PropScalpDesk]

In strong uptrends, RSI often stays between 40 and 90, with dips toward 40 to 50 acting as support.

The version usually credited to Andrew Cardwell and Constance Brown is roughly 40 to 80 in bull phases and 20 to 60 in bear phases. Readings of 90 or 10 do happen on a 14-period RSI, but rarely enough that I wouldn't build a range around them. The useful idea is the floor: in an uptrend the dips stop around 40 to 50 instead of reaching 30.

On point 4, the 50 line crosses constantly in chop. On M5 EURUSD in Asia it can flip a dozen times while price goes nowhere, so I only count a cross after RSI has spent hours on the other side and then holds on the first pullback.

[Fairman]

Readings of 90 or 10 do happen on a 14-period RSI, but rarely enough that I wouldn't build a range around them.

fair correction on the 90, that was me being sloppy. 40 to 80 is what I actually see on H1 in a proper trend, a 90 print is rare enough to ignore. And yeah the 50 cross in Asia is useless, I've stopped even looking at it before London opens.

[LondonNewsTrader]

In trending markets, that rule often loses money, because price can stay overbought for a long time.

Your 40 to 50 support zone for uptrends is a useful idea, but it's worth checking what timeframe and period people are using before relying on it. On a 14-period RSI on the 1H chart, a single release candle can push the reading from 55 to 25 in one bar, and then the zone gets "broken" by an event rather than by a change in trend. After a strong US number, EURUSD's hourly RSI can sit under 30 for several hours while the dollar reprices.

So I'd treat RSI readings in the hours around the biggest data as a special case. The momentum is real, but it reflects new information rather than overbought or oversold conditions in the usual sense, and the ranges you describe usually reassert themselves once the market has settled.

Your point about identifying the market condition first is the key to all of it.

[PropScalpDesk]

Good catch on owning the 90. Cardwell/Brown ranges are the ones that show up in my H1 logs too — 40–80 with-trend, and a 90 is an outlier, not a target zone.

Asia 50-crosses I stopped marking years ago. They fire constantly in a flat book and never changed a London decision for me. If the H1 floor holds after a pullback into the 40s in a bull phase, that is the RSI information I will actually use.