A level that gets touched proves nothing on its own.

Support and resistance levels can work, but a high hit rate alone doesn't prove it. Dense levels get touched by chance. Draw enough lines and price will bump into one of them every time it turns.

That's the trap in the screenshots of perfect reactions people love to share. You see the turn at the line and forget the five other lines nobody mentions. You also forget the days price sliced straight through without a pause.

The honest question is narrower. Does price stop at your levels more often than at lines placed with no skill at all? If the answer is no, the levels are decoration. If the answer is yes, by a clear margin over many days, you have something worth trading.

Run this arithmetic before you trust any hit rate. Say a stock trades a 6-point range for the day. Your map has six zones, each half a point wide.

Six zones times half a point is 3 points of zone. That's 3 of 6 points, so half the day's range sits inside a level. If price turns anywhere at random, it has roughly a one in two chance of turning inside a zone.

Now make the zones a full point wide. Six points of zone now cover the whole 6-point range. Every turn of the day lands on a level, and the map scores 100 percent with zero skill.

This is why width and count matter more than any single good call. A thin map can post a lower hit rate and still be the better map.

The fix is a placebo map. For each day, build a fake map with the same number of levels and the same widths as yours. Place them at random inside the same price band.

Then score both maps with one rule written in advance. For example, count a reaction when price enters a zone and a candle closes back out the same side. Count a failure when a candle closes through it.

Draw the random map many times per day, not once. One random draw can get lucky or unlucky. Fifty draws give you a typical score for pure chance.

Your real map should beat that typical score day after day. A small edge over a couple of weeks can be noise. A steady gap over months is the signal you want.

You need a spreadsheet, your charting platform and about ten minutes after each close. Random numbers come from any spreadsheet's random function, so no coding is required. Keep one tab for your levels and one tab for the random draws.

The order matters. Writing the rules first and drawing the levels before the open are what keep the test fair. Skip either step and you'll grade your own homework with the answers in view.

Expect the first week to feel slow. After that it becomes a routine you run while reviewing the day. The numbers start to mean something once a few weeks of rows pile up. The steps below assume one instrument; run a separate sheet for each symbol you chart.

Hindsight ruins level tests faster than anything else. A line drawn after the turn will always look precise, because you placed it there knowing the answer. The only fair test uses levels that existed before price got there.

We learned this on our own work. An early check of our own level work produced a match rate that looked great. Then random lines of the same width scored close to it. The headline number was mostly density.

So pick fewer, tighter levels and keep their widths honest. Zones that swallow half the chart will always look right. Judge one instrument at a time, too. A width that's fair on an index fund can be far too wide on a slow, cheap stock. The test will flatter you there.

If your map can't beat random, change how you draw it, not how you score it. Look at where the levels come from first, before you blame the market. Lines picked by eye from distant highs tend to crowd together. Crowding is exactly what inflates hit rates.

Try precise prices instead of wide bands. Merge levels that sit too close to trade between, and drop the ones that never get tested. Then rerun the same test for another month, with the same rules and the same widths.

If the gap opens up, keep going and plan your trades from those levels. If it never opens, treat the levels as background and stop using them as entries. That call feels harsh, but it saves money.