Advertisement IC Markets

Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Discuss 1-minute to 15-minute price action setups, fading intraday momentum, key support/resistance zones, and proven short-term trading methodologies.
Post Reply
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Post by Fairman »

Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Elliott wave theory proposes that markets move in repeating patterns of five waves in the trend direction, followed by three corrective waves. It has a devoted following and a long history.

The attraction is that it offers a complete narrative of where price is in a cycle, with rules about wave lengths and overlaps and links to Fibonacci ratios.

The problem is subjectivity. Experienced analysts routinely label the same chart differently. When price does something unexpected, the count can be revised, with the earlier labeling declared a mistake. A theory that can always be adjusted after the fact is difficult to test.

Studies of its predictive value are limited, and where they exist, results are mixed at best. That does not mean traders cannot profit from it, since skill in risk management matters more than the framework. But it does mean claims of precision should be treated carefully.

A moderate use is to take the useful parts without the dogma: that trends move in impulses and corrections, that corrections often retrace a portion of the prior move, and that momentum tends to be strongest in the middle of a trend.

If you use it, define invalidation levels in advance. A wave count that has no point at which it is wrong cannot guide a stop.

Practical step: label a past chart using the rules, then ask a friend to label it independently. Compare.
It’s Fairman :geek:
Recommended broker for automated trading & scalping IC Markets
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Post by Fairman »

A practical issue with Elliott wave: two analysts can label the same chart completely differently, and both can adjust their count after price moves. If a framework can always be relabelled to fit, it can't really be wrong, which also means it isn't telling you much. If you want to use it anyway, the way to keep it honest is to write your wave count and the level that would invalidate it before the trade, and treat that level as a real stop. If price breaks it, the count was wrong, and you don't get to relabel and stay in. Used like that, it's just another way of describing structure with a clear exit.
It’s Fairman :geek:
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Post by Fairman »

To be fair to wave traders, some use it loosely as a way to think about impulse and correction, which is basically structure. That part is useful. It's the precise wave counts and predictions I'd be careful with. The hard rules of Elliott Wave, like wave 3 never being the shortest and wave 4 not overlapping wave 1, can at least tell you when a count is wrong. But many analysts simply relabel the waves when price doesn't follow, which makes the theory hard to test. Look at their invalidation level before trusting any count.
It’s Fairman :geek:
FXS Prop Desk
Posts: 1693
Joined: Wed Oct 07, 2026 9:52 am

Re: Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Post by FXS Prop Desk »

Traders who use Elliott Wave often find it most useful for defining invalidation, since the rules (such as wave 4 not overlapping wave 1) provide clear levels. Using the theory for risk management rather than prediction avoids much of the subjectivity.

Testing Elliott Wave rules objectively, for example by coding them, can also help assess their value.

Some traders also use the theory to frame scenarios rather than predictions, which keeps flexibility when price does something unexpected.
FXS Official Team | The Idea Lab & Strategy
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Post by Shadow Trader »

The overlap rule is the most usable part, though it comes with an escape hatch worth knowing. Leading and ending diagonals are allowed to overlap wave 1, so an analyst whose count breaks can simply rename the move a diagonal and carry on.

Here is how the rule looks as a trade. Say wave 1 runs from 1.0800 to 1.0860, wave 3 tops at 1.0950 and wave 4 pulls back to 1.0880, then a long there has a natural stop just under 1.0860, about 20 pips away.

If wave 5 matches wave 1 at 60 pips, the target is 1.0940, which is 3 to 1 on paper. The catch is that a diagonal relabel would let you stay in below 1.0860, so the stop only works if you decide in advance that a break there ends the trade.

On low timeframes the problem grows because the degree of each wave is open to debate. A five wave move on M1 can be wave 3 of something on M15 or the whole of a correction, and both can be argued.

Has anyone coded the rules strictly, with diagonals excluded, and compared hit rates against a plain swing structure stop? That seems the fairest test to me.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Post by Shadow Trader »

Coding the rules sounds like the clean test FXS Prop Desk suggested, but trouble starts before any wave is labelled. A program needs a definition of a swing, usually a zigzag that only counts a reversal once price moves a set distance, and that threshold quietly decides the count.

Say the same week of EURUSD M15 is run with a 10 pip and a 25 pip threshold. The smaller one might find 40 swings and a dozen valid five wave sequences, while the larger one finds 14 swings and only two, so the measured hit rate partly reflects the filter.

That does not kill the test, it just means the threshold has to be fixed before looking at results. Ideally it scales with volatility, such as a multiple of daily ATR.

On scenario framing, one practical snag is when the two counts have invalidation levels far apart. For a long from 1.0880, if the main count fails at 1.0860 and the alternate at 1.0790, holding through the first level because the alternate is still alive turns a 20 pip stop into 90.

A rule that helps is to size and stop on the primary count only. The alternate tells you what to look for next, not where to stay in.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Elliott Wave Theory: A Fair Look at an Unfalsifiable Map

Post by Shadow Trader »

Elliott wave theory proposes that markets move in repeating wave patterns, five waves in the trend direction and three against. Practitioners count waves on charts and make forecasts. Enthusiasts and sceptics both have plenty to say.

The major criticism is that the theory is flexible enough to fit almost any chart. When price does something unexpected, the count gets revised. Waves can be extended, truncated, combined or relabelled. A theory that explains every outcome predicts none of them.

Defenders argue that skilled analysts apply strict rules and that wave counts have invalidation points. Some traders find the framework useful as a way of organising their thinking, even if they don't treat it as a forecasting machine.

For scalpers on short timeframes, wave counting is especially hard. Intraday noise makes counts ambiguous, and a count that looks clear on the five-minute chart disappears at the one-minute level.

A fair evaluation would require counts published in advance, with specific price targets and timeframes, tracked over many instances.

A test of falsifiability: could a given wave count be proven wrong by price action, and by exactly how much? If the practitioner can say "this count is wrong if price closes above X", the count is a testable claim. If there's always a re-labelling available, the framework adapts to every outcome.

Defenders point out that discretionary traders in many fields, including chess players and doctors, rely on pattern recognition that's hard to formalise. That's fair, though it doesn't make the claims verifiable.

The pragmatic route is treating wave ideas as a way of structuring observation, not as a source of price targets.

Has a published, falsifiable Elliott count ever been tracked over time, and how did it do?
Post Reply