RSI below 30. Time to buy, right?
Wrong.
I can't count how many times I bought "oversold" conditions only to watch price keep dropping. And dropping. And dropping.
"But the indicator said..."
The indicator lied. Or more accurately, I misunderstood what it was telling me.
Here's the uncomfortable truth:
Every indicator is just a mathematical transformation of price. They don't predict anything. They describe what already happened.
RSI is overbought? That means price went up a lot recently. That's it. It doesn't mean price will go down.
MACD crossed? That means the short-term average crossed the long-term average. Price already moved. The cross is confirming old news.
Moving average support? It's just an average of past prices. Price doesn't "know" where the moving average is.
I spent years treating indicators like crystal balls. They're not. They're rearview mirrors.
Okay, so if indicators don't predict, what are they good for?
"Price went up a lot" is vague. "RSI is 75" is specific.
Indicators give you numbers. Numbers you can use in rules. Rules you can automate.
This is actually useful.
When price makes a new high but RSI makes a lower high, something's off. Momentum is weakening even as price pushes higher.
Divergences don't guarantee reversals. But they're a warning sign worth noting.
A moving average smooths out the chaos. It shows you the general direction without the tick-by-tick noise.
Useful for trend identification. Not useful for timing entries.
Indicators can confirm what price action is already showing.
Price breaks resistance AND RSI breaks above 50? That's confirmation. The breakout has momentum behind it.
But notice: price action comes first. The indicator confirms.
After years of experimentation, here's what survived:
What I use it for: Divergences and extreme readings in context.
What I don't use it for: Buying oversold or selling overbought blindly.
RSI at 20 in a strong downtrend? Price can stay "oversold" for weeks. Don't catch that falling knife.
RSI divergence at a major support level? Now we're talking. That's confluence.
What I use it for: Confirming moves and identifying exhaustion.
What I don't use it for: Entry signals on its own.
Breakout on high volume? Probably real. Breakout on low volume? Probably fake.
Volume spike after extended move? Possible exhaustion. Smart money might be distributing.
What I use it for: Trend direction and dynamic support/resistance.
What I don't use it for: Precise entry timing.
Price above 200 EMA? Probably an uptrend. Look for longs. Price pulling back to 20 EMA in an uptrend? Possible entry zone.
But I don't buy just because price touched the moving average. I need price action confirmation.
Seriously. RSI, volume, and a couple of moving averages. Everything else is noise.
I used to have 10 indicators on my chart. Now I have 3. My results improved.
Let me tell you about the indicator trap.
You find an indicator. It works for a while. Then it stops working. So you add another indicator to filter out the bad signals.
Now you have two indicators. They work for a while. Then they stop working. So you add a third.
Before you know it, your chart is a mess of lines and colors, and you're more confused than when you started.
The problem isn't finding the right indicator. The problem is expecting indicators to do something they can't do.
Here's my real process:
I look at the chart. What's the trend? Where are the key levels? What's the structure?
No indicators needed for this. Just candles.
Once I have a potential setup based on price action, I check indicators for confirmation.
Price at support with bullish engulfing? Check RSI—is there divergence? Check volume—is it above average?
If indicators confirm, the setup is stronger. If they don't, I'm more cautious.
I translate my conditions into specific indicator values for dashpull.
"RSI divergence" becomes "RSI at current low is higher than RSI at previous low while price at current low is lower than previous low."
Specific. Measurable. Automatable.
You want to know my actual indicator setup? Here it is:
That's it. Four things. Clean chart. Clear thinking.
I can see:
Everything else is decoration.
Here's where indicators become powerful.
Instead of watching charts waiting for indicator signals, I define the conditions in dashpull and let the system watch.
Example setup:
Conditions:
Action: Enter long with predefined stop and target.
The system monitors 24/7. When all conditions align, it executes. I don't need to stare at indicators all day.
"RSI says buy, so I buy."
No. RSI doesn't "say" anything. It's a number. You decide what to do with that number, in context.
RSI at 30 means different things in different contexts.
In a strong uptrend, RSI 30 is a pullback. Probably a buying opportunity. In a strong downtrend, RSI 30 is just the beginning. Price can go much lower.
Context matters more than the number.
If you need 5 indicators to agree before you trade, you'll never trade. And when you do, you'll be late.
Pick 2-3 indicators. Master them. Ignore the rest.
"What if I use RSI 7 instead of RSI 14?"
Stop. You're curve fitting. The difference between RSI 7 and RSI 14 is minimal. You're optimizing noise.
Use default parameters. They work fine.
Here's a challenge for you:
Trade for one week with no indicators. Just price and volume.
You'll be forced to actually look at the chart. To see the patterns. To understand the structure.
Most traders can't do this. They're dependent on indicators to tell them what to think.
Break that dependency. Learn to read price. Then add indicators back as confirmation tools, not decision makers.
Indicators are tools. Useful tools, when used correctly.
But they're not crystal balls. They don't predict the future. They describe the past in mathematical terms.
Use them for:
Don't use them for:
dashpull lets me combine indicator conditions with price action conditions in my automated orders. The system watches for everything to align. When it does, it executes.
That's how indicators should be used. As part of a system. Not as the system itself.
Ready to build indicator-based conditional orders? Try dashpull →
Quant Trading for Retail Traders: You Don't Need a PhD (But You Need This)
Quant trading isn't just for hedge funds anymore. Here's how regular traders can use quantitative approaches without a math degree.
RSI Trading Strategy: How I Stopped Buying Every Oversold Signal (And Started Making Money)
RSI below 30 doesn't mean buy. Here's how to actually use RSI without getting destroyed by false signals.