Ask a discretionary futures trader what's killing their results and a lot of them will blame the signal. Too many entries, half of them fail, and the good ones are buried in the bad. But the uncomfortable truth is that most "false signals" aren't the indicator misfiring. They're the setup firing exactly as defined, on bars you shouldn't be trading. Fix the definition and the noise mostly takes care of itself.
A "false signal" is usually a filtering problem
A pullback setup like a second entry has a precise structure: a higher low in an uptrend, a bar that breaks the prior bar's high, and so on. (If that structure is fuzzy for you, start with our second entry guide.) The pattern itself is binary. Either the bars line up or they don't.
The trouble is that plenty of bars line up structurally while being terrible trades. A second entry that triggers on a huge bar with 12 ticks of risk is technically a valid second entry. So is one where the signal bar closes on the wrong end of its range, or one that's the fifth pullback in a dying trend. The pattern says yes; common sense says no.
So the goal isn't a "more accurate" indicator. It's a consistent set of rules for which valid signals you skip. That's what filtering is, and it's where the edge lives.
The filters that cut noise
Most useful filters fall into three buckets. On its own each one trims a slice of bad trades; together they change the character of what shows up on your chart.
Start with bar quality. The single bar that triggers your entry tells you a lot. A long entry on a bar that closed near its low is weak, because buyers didn't hold the push. WiSE's Max High-Close % filter (and Close-Low % for shorts) hides exactly those: set it to 30 and any long signal bar whose close sits more than 30% down from the high gets dropped. Same idea with Hide Entries by color. A red bar triggering a long is a contradiction, so you can hide it outright.
Next comes risk. Every entry implies a stop behind the signal bar, and that distance is your risk in ticks. A setup that needs 9 ticks of room on an instrument where you size for 5 isn't the same trade as one that needs 4, even if both are "second entries." The Hide Entries by Max Risk (ticks) filter enforces this for you: set it to 6 and a 7-tick setup simply doesn't display. You stop cherry-picking risk in the moment and let the rule decide.
The one traders skip is structure, and it matters most. A second entry inside a strong trend is not the same as the fourth pullback after the trend has stalled. Count caps (Max count Entries and Max count HL/LH per swing) stop the late, low-odds signals from cluttering the chart. Hide 0/1 without 2 removes the early scaffolding so you only see completed setups. And a moving average filter can hide entries that fire when price is stretched too far from the mean, the ones that look great and reverse immediately.
Keep the failures on the chart
One thing worth doing the opposite of your instinct: don't hide setups that fail. When a second entry triggers and then breaks its stop, WiSE marks it failed (x), or challenged (c) if it stalls and turns before hitting target. Those marks aren't clutter. A cluster of failed second entries near a high is one of the more reliable tells that a trend is done, and if you filter them off the chart you've thrown away information you paid for in real trades.
Cutting noise means hiding the setups you won't take. It doesn't mean hiding what the market is telling you.
Apply it consistently, every session
I've watched this play out enough times to be blunt about it: the filters only work if you stop overriding them. The failure mode isn't a bad setting. It's setting a 6-tick risk cap on Monday and taking a 9-tick "obvious" trade on Thursday because it felt right. Two weeks later the chart is noisy again and the indicator gets the blame.
A sane way to dial it in:
- Trade one instrument and one bar type you already know well. For ES that's often a 2000-tick chart (see our guide on tick charts vs. time charts for why tick bars suit this style).
- Start with structure filters only (count caps, hide 0/1 without 2). Watch a week. Note which surviving signals you'd still have skipped.
- Add one bar-quality or risk filter at a time and tune it to those skips. Don't stack all five on day one and wonder what changed.
The WiSE indicator applies this whole filtering stack in real time and marks over 16 signal types with zero repaint, so what you test is what you trade. Each filter above maps to a real setting. The filtering signals reference walks through every one, and understanding the signals explains the 0/1/2/H/L marks and the challenged/failed logic. WiSE's method is built on Mack's adaptation of Al Brooks' price action work, so the definitions are consistent whether you apply them by eye or let the tool do it.
Fewer signals, taken the same way every session, beats a chart full of maybes. That's the entire point of filtering, and it's the difference between blaming the indicator and trading.
Trading futures involves substantial risk of loss and is not appropriate for all investors. Nothing here is trading advice; signals are analysis tools, not recommendations.