← AlgoRishi Blog · algorishi.com

Support and resistance: how levels get drawn and why most breaks fail

16 November 2026 · AlgoRishi

You drew a line across the top of yesterday's range. At 11:40 IST price pushed through it, you bought the call, and within nine minutes price was back under the line and your premium was smaller. Nothing was wrong with your line. The problem is what you assumed a break means.

20units in one SENSEX option lot
100points between SENSEX strikes
ThursdaySENSEX weekly expiry
15:40IST F&O close

What a level actually is

A support or resistance level is not a prediction. It is a place on the price scale where a lot of orders have been left sitting — buy orders below, sell orders above — and where, in the recent past, the tape visibly slowed down.

Most of the levels any honest system uses are drawn from facts that were already settled before the session opened:

A level is a question the market asks: is anyone still willing to defend this price? A break is not the answer. A break is the asking.

The auto-rickshaw queue

Picture the auto stand outside a railway station. Fares are supposed to be ₹80 to the bus depot. One driver asks ₹120.

Someone pays ₹120 once. Has the fare changed to ₹120? Not yet. If the next ten passengers refuse and walk to the prepaid counter, the ₹120 was a one-off and the fare snaps back to ₹80. If the next ten all pay ₹120, the fare has genuinely moved.

A price level works the same way. The first trade through it proves only that one buyer was willing. Whether the level has actually moved depends on what happens after, and that information does not exist at the moment of the break — which is exactly the moment most people click buy.

Why most breaks fail

1. Levels are where two-sided business lives

The reason price paused there before is that sellers were active. Those sellers do not vanish because the line got touched. A first poke through resistance often runs straight into the people who were waiting for a slightly better price to sell at.

2. Everyone drew the same line

Yesterday's high is not a secret. Protective stop-loss orders cluster just beyond obvious levels, and triggering them creates a burst of buying that has nothing to do with anyone wanting to own the index higher. Once that burst is exhausted, there is no one left to buy, and price slides back through the line it just cleared.

3. Your definition of a break was never written down

A wick one point through the level, a five-minute candle closing above it, and a one-minute candle closing above it are three completely different events. If you decide which one counts while it is happening, you will pick whichever one lets you enter. That is not a break condition; it is a permission slip.

4. The option in between

This is the part that does the real damage, and it is specific to buying options.

Suppose, as an illustration, SENSEX is near 84,950 and the level you care about is 85,000. You buy the 85,000 call at a premium of ₹180. One lot is 20 units, so that position costs ₹3,600.

Price breaks the level, reaches 85,060, stalls, and comes back to 84,980 in twelve minutes. But you crossed the bid-ask spread to get in, time passed, and the move was never big enough to pay for either. If the premium is ₹150 when you give up, that is ₹3,000 back — ₹600 gone on a break that technically worked.

Now do that on a Thursday, when the weekly contract expires. Premium on a near-strike option drains fastest on its last day, so the move has to be bigger and faster to clear the same hurdle.

The same level, asked twice the level poke through, then back under closes above, and holds time →
Both events cross the same line. Only the second one is a break by any written definition — and you can only tell them apart after the candle has closed.

How to measure this for yourself

Do not take anyone's word for this, ours included. Pick one level type — say yesterday's high on SENSEX — and keep a log for twenty sessions with four columns: the date, the level, how far price travelled past it before coming back, and whether a one-minute candle ever closed beyond it.

After twenty sessions you will have your own answer to the only question that matters: how often does a break of this specific line travel far enough to cover the spread and the time you paid for? The counting is the edge, if there is one.

Write your break definition in one sentence before the market opens, with a number in it. "A one-minute close more than X points beyond the level" is testable. "A convincing break" is not.
This is educational material, not a recommendation to trade anything. SEBI's studies on individual equity derivatives traders have repeatedly found that the large majority lose money over a full year, and a level on a chart does not change that arithmetic.

Levels are genuinely useful. They tell you where the market is likely to make a decision. What they do not tell you is which way the decision will go, and any method that quietly assumes otherwise is going to spend a lot of money finding out.

Disclosure: AlgoRishi publishes signal tracking on real market prices using virtual capital, for information and education only. Nothing here is investment advice or a recommendation to trade. Futures & options trading involves substantial risk of loss. Past performance is not indicative of future results. AlgoRishi is not a SEBI-registered investment adviser or research analyst.