A level tested five times: stronger or weaker?

[Fairman]

A level tested five times: stronger or weaker?

There's a split view on this. One camp says the more often a level is tested and holds, the stronger it is. The other says every test uses up some of the orders defending it, so each test makes a break more likely.

I think both are partly right, and the useful question is how the tests look rather than how many there are.

If each test produces a clean rejection that travels a decent distance away, the level is being defended actively. Someone wants to trade there. Counting doesn't matter much in that case.

If the tests get closer together and the bounces get smaller, that's a sign of weakening. Price is coming back faster and not travelling as far away each time. That's often what it looks like before a break.

The other thing that changes over multiple tests is where the stops are. After several holds, a lot of traders will have positions leaning on that level, with stops just beyond. That pool gets bigger with each test, which makes a break more attractive for anyone wanting liquidity.

So I don't fade the fifth test just because the first four held. I look at the bounce sizes and the gaps between tests. If both are shrinking, I'd rather wait for the break and trade the retest than lean on the level again.

[Fairman]

Time between tests is something I'd add to the picture. Two tests ten minutes apart are really one event, the same order flow bumping into the level twice. Tests spread across days are more meaningful, because different participants are making separate decisions there. When people say a level was tested five times, it helps to check whether that means five separate sessions or five touches in one messy hour. The first is a strong level. The second is often just a choppy patch that happens to sit at that price.

[Fairman]

There's a related case where a level gets tested from both sides. Price breaks below a support, comes back to test it from underneath a couple of times, and fails. That level has now been tested several times, but its role has flipped. Counting tests without noting which side they came from mixes two very different stories. I mark flipped levels in a different colour, so when I count tests I'm only counting the ones that happened with the level playing the same role each time. If the retests keep failing from underneath, that tells you the old level has flipped, and I'd treat it as resistance until price proves otherwise.

[FXS Prop Desk]

Both views can be right depending on how price reacts. If each test produces a strong bounce, the level is being defended. If the bounces get smaller with each test, the defending orders are being used up and a break becomes more likely.

Looking at the size of each reaction, rather than just the number of tests, gives a clearer picture. A level with weakening bounces is often a better breakout candidate than one with strong ones.

[Fairman]

Where each test came from matters too. If price reaches the level after a sharp move, that test is more forceful than one that drifts slowly into it. Five tests that each came after a strong push, and each held, says something different from five tests where price just wandered up to the level and back. I'd look at the approach as much as the count. Strong approaches that get rejected suggest real orders at the level. Weak drifts suggest the level is holding mainly because nobody is pushing hard yet.

[Fairman]

How does the reaction size change across the tests? On the first test, price might bounce thirty pips. By the fourth or fifth, maybe only five or ten. Shrinking reactions often mean the orders defending the level are getting used up, and a break is getting closer. Growing or steady reactions suggest the level still has plenty behind it. I keep a quick note of the bounce size after each test. It's a more useful number than the test count on its own, and it's easy to measure on the chart in a few seconds.