Brent tests near-equal highs - is a double top forming?Brent spot in USDTHINKMARKETS:BRENTThinkMarketsBrent has gained a net 1,136 points from the 22 September low at 94.151 to the 12 October high at 105.510, though not in a straight line: a 785 point pullback into 30 September sits inside it. Then on Monday it stopped at 105.510, and this morning it stopped again at 105.441. Sixty-nine points apart, thirty-two hours apart. Price is 104.897 with the 08:00 UTC candle open, up 0.76% on the session. The 105.059 reading is that candle's high, not the day's. The pattern: a potential double top ●Peak one: 12 October 16:00 UTC at 105.510. ●Peak two: 14 October 00:00 UTC at 105.441, which is 0.069 lower. Near-equal rather than identical, which is enough to monitor and not enough to call a reversal. ●The trough: 102.648 on the 13 October 08:00 UTC candle, which is the neckline. ●The height: 2.862 points from peak to neckline, or 2.4 average four-hour ranges at the current 1.20 ATR. ●Completion: a four-hour close below 102.648. The measured objective is then approximately 99.79, and the 50% retracement of the September advance at 99.831 sits 4.5 points above it, not beneath it. ●Invalidation: a four-hour close above 105.51 removes the pattern. The next mapped resistance is 105.84 to 105.90 with the 15 September high at 105.874, and above that the chart is not empty: 106.146 on 14 September and 106.407 on 11 September are still there. Two things keep this provisional. The peaks are only thirty-two hours apart, which is tight for a double top even on a four-hour chart, and price is still 225 points above the neckline, about 1.9 average ranges. A double top is a shape until the neckline closes beneath it; until then this is an uptrend pausing at the same price twice. Candlestick patterns at the second peak ●13 October 20:00 UTC, shooting star, now void: a long upper wick into 105.133 with the body at the bottom. The very next candle reached 105.441 and voided it. I had called this one live by checking only the current candle instead of every candle since, which is the wrong test. ●14 October 00:00 UTC, upper-wick rejection candle: this is the candle that made peak two at 105.441, and it qualifies on both the shooting star and doji tests, which is one signal rather than two. Void above 105.441, and nothing since has reached it, so this one is still live. ●14 October 04:00 UTC, bearish engulfing, now void: it swallowed the previous candle, but the live candle's high at 105.059 has traded above its 104.726 invalidation, so the signal is dead. So one rejection signal is live and two have failed. That is the honest state of the evidence, and it is a thinner case than the three-candle cluster it looked like at first glance. The trendlines Two rising support lines matter. The primary one runs from the 22 September low at 94.151 through the 30 September low at 94.789, climbs 1.9 points per four-hour bar and is worth 95.968 at this candle. No low has traded beneath it in three weeks, and it sits far below price. The steep one runs from the 6 October low at 96.325 through the 12 October low at 100.649, climbs 20.6 points per four-hour bar and is worth 103.532 at this candle. That is the line doing the work: this morning's low at 103.701 held 17 points above it. It is not a fixed level, so its trigger has to be recomputed on each new candle. The levels Above price, 105.06 to 105.51 holds this candle's high and both peaks. Above that, 105.84 to 105.90 is the 15 September high, and beyond it the September highs at 106.146 and 106.407 remain on the chart. Below, 103.53 to 103.70 pairs the steep rising line with this morning's low and the 20 EMA at 103.605. Then the neckline band at 102.65 to 102.83, where the provisional 23.6% retracement at 102.829 sits with the trough. Then 101.17 to 102.14, holding the 50 EMA at 102.138 and the provisional 38.2% at 101.171. All four moving averages are below price and in bullish order at 103.605, 102.138, 100.725 and 98.232, which makes them potential dynamic support rather than confirmed support. RSI is 60.6 against its own average at 59.0. Momentum eased slightly at the second peak, but price did not make a higher high there, so this is not a divergence and should not be labelled one. Scenarios ●Bullish: A four-hour close above 105.51 clears both peaks and removes the double top, leaving 105.84 to 105.90 as the only mapped structure above. ●Bearish: A four-hour close below the steep line at its value for that candle, together with a sustained loss of the 103.53 to 103.70 band, is the first real deterioration. The line rises with each bar, so 103.53 is today's number and not a fixed one. A confirmed close below 102.648 completes the double top and opens the measured area near 99.79, with the 50% retracement at 99.831 just above it. A failed breakdown remains possible. ●No confirmation: While price holds above the steep line at its current value and below 105.06, the pattern stays unresolved. Closes inside 105.06 to 105.51 are a third test of the same ceiling rather than a break of it. What this tells us Two near-equal highs identify possible resistance, not a confirmed reversal. The candle evidence is thinner than it first appeared: of the three rejection candles at the second peak, two have already failed their own invalidation tests and one remains. The pattern still needs a four-hour close below 102.648, and the measured area near 99.79 sits just under the 50% retracement at 99.831. The other side is equally clear. The steep rising line sat 17 points under this morning's low, and the advance that produced a net 1,136 points in three weeks, pullbacks included, has not broken anything yet. So: does the third approach to 105.5 clear it, or does the neckline go first? Any opinions, news, research, analyses, prices or other information contained on this website is provided as general market commentary and does not constitute investment advice. ThinkMarkets will not accept liability for any loss or damage including, without limitation, to any loss of profit which may arise directly or indirectly from use of or reliance on such information.