People flex their peak WPM. Nobody flexes their consistency score, mostly because nobody knows what it's doing. It's not a vibe - it's a coefficient of variation wearing a friendlier name.
Here's the actual formula: take every WPM sample recorded during your run, find the average, then measure how far each sample strays from that average (the standard deviation), and express that spread as a percentage of the average. Subtract from 100. A run where you type at a dead-flat 70 WPM the entire time scores near 100. A run where you spike to 95 WPM for three seconds and crater to 40 WPM twice scores much lower - even if the average WPM on both runs is identical.
This is why two typists can post the same WPM and be completely different typists. One is a metronome. The other is a rollercoaster that happens to average out. Raw WPM can't tell them apart. Consistency can.
It also explains a pattern a lot of players notice but can't name: the run that felt smooth almost always scores higher on consistency than the run that felt like a personal best, even when the personal-best run has the higher WPM. That's not you misremembering - smooth is a different variable than fast, and TypeClash is one of the few places that actually puts a number on it.
If you're trying to improve, raw WPM tells you your ceiling. Consistency tells you how often you're actually near it.