Keenbase Trading » Support and Resistance Trading Strategies

Support and Resistance Trading Strategies: The Bounce and Retest System

Every guide to support and resistance tells you to trade the strong levels, and most of them measure strength by counting touches. But a level touched four times that broke on three of them is not strong. It is busy.

The Bounce and Retest System answers that with a number. Every level carries a Reversal Chance, the share of its past touches after which price turned away, and the system trades only levels where price turned away on at least 60 percent of at least two touches.

One support level and its record: three of four touches held, so the fifth touch is traded.

From there the system has two strategies, and one question decides which one you use. If price comes back to a qualifying level and closes on the same side, you trade the bounce. If price closes through the level, you wait for it to come back and trade the retest from the other side.

Which Support and Resistance Levels Are Worth Trading

A level is worth trading when price has already respected it. That is the whole idea, and it removes the part of support and resistance that is usually left to the eye.

The levels come from structure a trader already knows. There are swing highs and lows on the monthly, weekly, daily and current chart, and the monthly, weekly and daily pivot points. There are the previous day, week and month high, low, open and close, order block midpoints, and zones where several of these sit close together.

Each level then gets two numbers. The touch count is how many times price has come back to it. The Reversal Chance is how often those touches ended with price turning away. A level with a Reversal Chance of 75 percent and 4 touches has turned price three times out of four.

Drawing and scoring all of that by hand every day is not realistic, so the system reads it from one indicator:

KT Auto Support Resistance Levels IndicatorDraws the support and resistance levels on the chart and shows the Reversal Chance and touch count on each one.See the indicator ›

The two filters that make a level tradable follow directly from those numbers:

• A Reversal Chance of 60 percent or more.
• At least 2 touches, so the record is more than one lucky turn.
• A fresh level with no touches yet is not traded, because it has no record to judge.

Levels move as new swings form, and traders sometimes read that as repainting. The system trades a level as it stood when the signal candle closed, so a later update never changes a trade that was already taken.

Bounce or Retest: The One Difference

Both strategies start at the same place, a qualifying level. They split on what the closed candle does there.

The stop loss and the target work the same way in both. What changes is only the question the candle answers: did the level hold, or did it break?

Bounce Strategy Rules

A bounce is traded on closed candles only. A wick that is still forming can look like a rejection and then close straight through the level, so the rules wait for the close.

When two levels give a signal on the same candle, take the one with the higher Reversal Chance, then the one with more touches. If a buy and a sell appear on the same candle, there is no trade: the chart is saying two different things at once.

JP225, 30 minute chart. A short bounce at the previous week open, a level with a Reversal Chance of 67 percent. Price pushed above it, could not stay there, and the rejection candle is the signal.

One Bounce, Candle by Candle

Here is one long bounce on JP225, the 30 minute chart, followed from the level to the exit.

The level: The previous week low sat at 63,472, with its zone running from 63,453.5 to 63,490.5. It had been touched twice and had turned price both times, so its Reversal Chance was 100 percent.

The candle before: The 08:00 candle closed at 63,836, well above the zone. Price was coming down to the level, not climbing up through it.

The signal candle: The 08:30 candle opened at 63,839, fell to 63,444, a few points below the bottom of the zone, and then closed at 63,704, back above the zone. That is the rejection the rules ask for: the level was tested hard and price closed on the support side.

The entry: The next candle opened, and the buy filled at 63,703.

The stop loss: The wick went lower than the zone, so the stop loss goes below the wick rather than below the zone, plus a small buffer: 63,310.50. That is 392.5 points of risk.

The target: The next level above was the daily pivot at 64,987.67, which is 1,284.67 points away, or 3.3 times the risk. Anything under 2 times the risk would have been skipped; this one passed easily.

What happened: Price moved one times the risk into profit within the first half hour, so the stop loss moved to break even. Later the same day the trade reached the daily pivot and closed at the target, for 3.3 times the risk.

JP225, 30 minute chart. The previous week low, the rejection candle that closed back above it, and the trade to the daily pivot.

The detail that makes this a trade and not a guess is the close. At 08:30 the candle was, for a while, below the level. A trader buying the touch would have been buying a falling market. The rules only act once the candle has closed back above.

Break and Retest Strategy Rules

A broken level does not stop mattering. Traders who bought at old support are underwater once it breaks, and many of them sell when price comes back to where they bought. That is why old support so often turns into resistance, and old resistance into support.

The retest strategy trades that turn, and it asks for two things the bounce does not: a clear break first, and a retest within a short window.

The break has to close clear of the zone because a candle that closes a few points past a level often slips straight back. The 10 candle window matters for the opposite reason: a retest that arrives much later is really a new touch of a new level, and it should be judged as one.

A qualifying level that breaks cleanly and then retests within 10 candles is not a daily event. With the filters above, months can pass between retest trades on one chart, so the break and retest works best as the second strategy on a chart you already watch for bounces.

Where the Stop Loss and the Target Go

The stop loss goes beyond the far edge of the zone and beyond the signal candle’s wick, whichever is further, plus a small buffer of 0.3 times the 14 period ATR. A stop inside the wick sits exactly where the market has just shown it is willing to go.

The target is the next opposing level: the next resistance above a long, the next support below a short. Levels closer than half an ATR are passed over, since a target that close is only noise away.

If that target is less than twice the stop loss distance away, the trade is skipped. This is the rule that removes the most setups, and it is intentional. A bounce straight into the next level has nowhere to go, even when the level it bounced from is excellent.

Once a trade is one times the risk in profit, the stop loss moves to break even plus a few points. From then on the worst outcome is a small win.

Bounce Strategy Results on JP225

The bounce strategy was tested on JP225, the 30 minute chart, with the published settings, over nearly six years of price history in one continuous run.

The win rate is close to a coin toss, and that is how the system is built. The average win is about 1.6 times the average loss, because every trade needs a target at least twice as far away as its stop loss, and break even turns some would-be losses into small wins.

Account balance of the bounce strategy on JP225 M30 at 1 percent risk per trade. Original Keenbase Trading Lab data.

The curve is not a straight line, and it is better to know that before trading it. For the first three and a half years the balance mostly moved sideways, with long gaps between trades. Most of the gain came in the last two years. A strategy that waits for a proven level can go months without a signal on one chart.

When a Level Fails

A Reversal Chance of 67 percent still means one touch in three did not hold. The strategy does not avoid failed levels. It keeps them small.

JP225, 30 minute chart. A bounce at the previous week high that failed: the next candle fell through the level and the stop loss closed the trade.

Here the previous week high had become support with a Reversal Chance of 67 percent. The signal candle closed above the zone and the trade opened, then the next candle fell straight through the level and hit the stop loss within the same half hour.

Nothing in the rules was broken. The level simply belonged to the one touch in three that fails.

Three things are worth knowing before you trade it:

• Losses come in runs. In the results above the longest run was 5 losing trades in a row, and a win rate near 50 percent means a run like that can come at any time.
• The levels come from your broker’s prices. A test on another broker’s data will not match these figures trade for trade.
• Very low timeframes are a poor fit. The strongest levels come from daily and weekly structure, and a one minute chart turns them into noise.

Which Timeframe and Market to Start With

Start with JP225 on the 30 minute chart, the settings in the results above, and 1 percent risk per trade. It is the market the published result comes from, and the settings the EA starts with.

Anything beyond that is your own test. The 1 hour chart is the natural next step, since it keeps the same levels and gives fewer, slower signals.

Trade It by Hand or With the EA

Both strategies can be traded by hand with the indicator. Wait for a qualifying level and for the close, place the stop loss beyond the zone and the wick, and check that the next level is at least twice as far away.

If you would rather not watch for the close, the same rules run automatically:

KT Support Resistance EAFinds the same levels, trades the bounce and, when switched on, the break and retest, with the stop loss and target placed by the rules above.See the EA ›

The EA trades the bounce by default, and the break and retest is one input away. It includes the JP225 30 minute settings used in the results above.

Whichever way you trade it, the decision is always the same one: has this level held before, and did the candle that just closed hold it again?