What's being scored
This scores each hook on length (shorter reads better in a feed), the presence of emotional trigger words known to prompt engagement, a readability score (the same Flesch Reading Ease formula used elsewhere on this site), whether it ends in a question, and whether it includes a number.
This is a heuristic based on common engagement patterns, not a prediction of actual performance - the platform, your audience, and the content that follows the hook all matter far more than the hook's wording alone.
Why several different signals are combined instead of just one score
Length, emotional trigger words, readability, questions, and numbers each capture a genuinely different reason a hook might work - a short, easy-to-read hook can still underperform one with more emotional pull, and vice versa. Combining several signals into one score gives a more rounded picture than optimizing for any single factor alone would.
Why emotional trigger words are scored by presence, not by how many appear
The score caps how much trigger words can contribute (up to a fixed maximum regardless of count) rather than rewarding piling on more of them - a hook stuffed with trigger words tends to read as spammy or clickbait-y past a certain point, so the scoring reflects that diminishing, even negative, return rather than treating "more trigger words" as simply better.
Why this is explicitly called a heuristic rather than a performance prediction
These structural patterns are commonly associated with engagement across many posts in aggregate, but your specific audience, platform, and what the post actually delivers after the hook all matter far more to real performance than wording alone - which is why this tool is framed as a quick structural sanity check, not a guarantee of which option will actually perform better once posted.