Williams Alligator Tested: It Was Asleep on Only 20% of Bars, and 1 of 24 Runs Passed All Three Checks

Bill Williams built the Alligator around a story. Most of the time the market sleeps and goes sideways; then it wakes up hungry and trends, and a trader should only eat when the Alligator eats. The indicator draws three lines: a blue jaw, a red set of teeth and green lips. When they are tangled, the Alligator sleeps. When they fan out in order, it is awake and you trade in the direction of the fan.

It ships with MetaTrader 4 and 5 under Indicators, Bill Williams, next to the Awesome Oscillator. Each part of the story is a rule we can count, so I counted it.

What does the Alligator indicator measure?

The shift is the part people misread. Moving a line 8 bars to the right does not let it see the future. It means the jaw on the current bar is the value the average had 8 bars ago, so the Alligator is, by design, slower than the same averages without the shift. We look at what that costs below and in detail in Alligator indicator settings for MT4.

How was it tested?

• Data: EURUSD M5 from June 3 to August 26, 2026, 16,926 bars; gold M5 from June 16 to August 26, 13,736 bars. M15 built from the same bars.
• Six rules: enter when the Alligator wakes up (the lines become stacked); the same only on the EMA 200 side; the same only when the Awesome Oscillator agrees; the lips crossing the teeth; the close crossing the jaw; and the awakening of the same three averages without the shift.
• Exit and pass: stop 1.5 ATR, target 3 ATR, 48 bars at most, one trade at a time, the same exit as in the forex indicators list, ranked. Cost 1 pip on EURUSD and $0.30 an ounce on gold, from spread and swap costs in MT4. A run passes if it makes money after cost, in both halves (split July 15), and beats the 95th percentile of 200 random-entry runs with the same exit, as in algorithmic rule vs random entry.

How often is the Alligator asleep?

This is the first surprise. The usual line is that markets trend only 15% to 30% of the time, and the Alligator is meant to sit out the rest. On our charts it did the opposite: it called a trend on 4 bars out of 5, on both pairs and both timeframes, almost to the decimal. Three smoothed averages of different lengths are in the right order whenever price has drifted the same way for a few bars, and on a 5-minute chart that happens all day long. A typical “trend” lasted 8 bars, 40 minutes on M5. For a measure that tries to separate trend from range in a stricter way, see Choppiness Index vs ADX.

Does price follow when the Alligator wakes up?

An awakening was close to a coin toss. On EURUSD M5 price was against the signal more often than with it 12 bars later, and only gold M15 showed a lean above 53% after 48 bars, on a chart where gold rose for most of the period. More than half of all awakenings fell back to sleep or flipped within an hour on M5, which is the whipsaw that indicator lag vs whipsaw measures across 21 indicators.

Did any Alligator rule make money?

Trades in brackets. 8 of 24 runs ended above zero after cost, six of them on gold M15. That chart looks good for every trend rule because gold climbed for most of the summer: random entries there had a 95th percentile of $2.77 to $3.59 an ounce, and none of the Alligator rules beat it. The unshifted awakening, the best number in the table, made $6.97 an ounce a trade in the first half and $0.54 in the second.

The one run that passed was the lips crossing the teeth on EURUSD M15: +0.88 pips a trade in the first half, +0.52 in the second, 199 trades, well above the random band. It is a cross of a 5-bar and an 8-bar average, which is why the head-to-head with an ordinary moving average crossover in Alligator vs moving average crossover matters.

Did the AO or EMA 200 filter help?

Not in a way you could rely on. Waiting for the Awesome Oscillator to agree changed very little, because it agreed with 85% or more of awakenings anyway: both are built from averages of the same median price. The EMA 200 filter removed 38% to 41% of the signals and made the result worse on 3 of 4 charts. On EURUSD M5 it turned -0.86 pips a trade into -1.23. Filters like this help when the rule underneath has some edge to keep, which is the lesson of indicator combo vs single indicator.

What would I do with the Alligator?

• Do not read “awake” as “trending”. The lines were stacked on 80% of bars on every chart we tested, so on its own the state says little.
• Skip it on EURUSD M5. All six rules lost after the spread, and no setting in the settings test made money on that chart.
• If you use it, use the lips and teeth cross on M15 or higher, and treat it as a moving average crossover with a stop and target, not as a story about sleep and hunger.
• Do not trust a good gold M15 result in a trending month. Random entries made almost as much there.

Frequently asked questions

What is the Alligator indicator in forex?

A Bill Williams trend indicator built into MetaTrader. It draws three smoothed moving averages of the median price, 13, 8 and 5 bars, shifted 8, 5 and 3 bars to the right. When they are stacked in order the market is said to trend; when they are tangled it is said to range.

Is the Alligator indicator profitable?

Not as a stand-alone signal in this test. 1 of 24 runs of six Alligator rules made money after cost in both halves and beat random entries: the lips crossing the teeth on EURUSD M15. The classic awakening rule passed on none of the 4 charts.

How often does the Alligator sleep?

On our EURUSD and gold charts, M5 and M15, the three lines were tangled on only 19.7% to 20.5% of bars. The other 80% were counted as awake, up or down.

Does the Alligator indicator repaint?

No. The shift moves past values to the right; it does not use future bars. The lines on the current bar change until it closes, so all signals here were taken on the closed bar.

What is the best time frame for the Alligator?

In this test M15 did better than M5 on both EURUSD and gold, mainly because the 1 pip spread is a smaller part of each trade. Only one M15 rule passed all checks.