September 8, 2026 · 8 min read
TL;DR: In 382 of this bank's 1,948 single-answer questions, at least two options open with the same two words. The key is one of those matched options 250 times, about 65% against a 50% expectation. The pair is not sloppy writing. It is the item writer telling you exactly where the discrimination lives.
Most exam-technique advice is about spotting weak options. This one is about spotting a shape, and it is the strongest structural signal we have measured in this bank that is not a defect.
Read the four options of any situational question and check the opening words. Roughly one time in five, two of them will match.
Here is one, on a subcontracted mobile application that fails the company's user-interface standards. Those standards were never in the statement of work, and the subcontractor will do the extra work if paid immediately:
Two options open Initiate a change. That is not an accident of drafting. The writer has decided that the question is about what the change request is for, and has removed every other variable so you cannot dodge it.
The first twin routes the request through change control and reads the contract before committing. The second commits to the payment milestone the subcontractor asked for, and calls that the change. The action word is identical. The entire question is in what follows it.
Yes, about 65% of the time, against the 50% chance would predict. The counts across all 1,948 single-answer questions:
| Measure | Count | Share |
|---|---|---|
| Single-answer questions in the bank | 1,948 | - |
| Questions where 2+ options share their first two words | 382 | 19.6% |
| Of those, the key is one of the matched options | 250 | 65.4% |
Chance would put it at 50%: two twins out of four options. Sixty-five percent is a solid edge, and it is a better-behaved edge than the answer-length tell, because it does not depend on how carefully someone balanced the options. It comes from how the question was designed.
The logic is straightforward. If a writer wants to test whether you know that a contract change goes through change control before pricing is agreed, the cleanest construction is two options that both initiate a change and differ only in what they do next. The unmatched options are there to fill the page. They are usually the extremes: negotiate the price directly, or terminate.
Common trap: "Two options look almost identical, so they must both be wrong, since only one can be right." That reasoning discards the answer 65% of the time it is applied here. Near-identical options are the opposite of a warning sign. They are the writer narrowing the question down for you. The pair on the subcontractor item differs by one idea, whether you read the contract before committing to terms, and that idea is the entire point of the question. This bank names the clause that separates the twins rather than just marking the loser wrong.
Four steps, and the third is the one that matters.
Step four is why the technique is worth having. A four-option situational question is a broad judgement call and feels like one. A twinned pair collapses it into a single yes-or-no about sequence, and sequence is the thing this exam tests most.
Two cases, both worth knowing.
The twins are both wrong. This happens in 132 of the 382, almost exactly a third. So check what the shared phrase is doing before trusting the pair, using the verb ladder: if the twins open with an investigation verb the key is very likely among them, and if they open with a handoff verb the twinning may exist to make the handoff feel like the obvious lane. Handoff-opening twins are too rare in this bank to put a number on, so treat that half as reasoning rather than measurement.
Three or four options share an opening. This is a different construction, usually a definitional question where every option is a noun phrase of the same type. The twinning signal carries no information there, because there is no unmatched pair to contrast against.
Neither case turns the technique off. Both are reminders that it is a way of finding where the question lives, not a way of answering it. The bank does the second part.
What is a twinned decoy on the PMP exam? Two answer options that open with the same words and diverge afterwards, such as two options both beginning Initiate a change to, or both beginning Review the timeline with. The pair is deliberately constructed so the discrimination sits in the difference, not the action.
How often does the correct answer sit inside the matched pair? In this bank, 382 of 1,948 single-answer questions contain at least two options sharing their first two words. In 250 of those 382, about 65 percent, the key is one of the twins. That is well above the 50 percent you would expect by chance.
Why would item writers build options this way? Because it is the cleanest way to test one idea. Holding the action constant and varying only the object or the timing forces the candidate to discriminate on the thing being tested, instead of picking a plausible-sounding verb.
How do I use this in the exam room? Scan the four opening phrases before reading the options fully. If two match, read those two against each other first and name the single word that differs. That word is almost always what the question is about.
PMP Practice's 2,141 questions are re-certified against PMBOK 8 and the July 2026 ECO, with every wrong answer explained, not just marked wrong, including which clause separated it from its twin. Start the free 20-question sample - no card, no signup required to try it.
What is a twinned decoy on the PMP exam?
Two answer options that open with the same words and diverge afterwards, such as two options both beginning Initiate a change to, or both beginning Review the timeline with. The pair is deliberately constructed so the discrimination sits in the difference, not the action.
How often does the correct answer sit inside the matched pair?
In this bank, 382 of 1,948 single-answer questions contain at least two options sharing their first two words. In 250 of those 382, about 65 percent, the key is one of the twins. That is well above the 50 percent you would expect by chance.
Why would item writers build options this way?
Because it is the cleanest way to test one idea. Holding the action constant and varying only the object or the timing forces the candidate to discriminate on the thing being tested, instead of picking a plausible-sounding verb.
How do I use this in the exam room?
Scan the four opening phrases before reading the options fully. If two match, read those two against each other first and name the single word that differs. That word is almost always what the question is about.