This One Tool Turned Mention Markets Into Free Money

Everyone's chasing many different plays on Poly Market, whether it's culture, crypto, finance, you name it. But

there's an entire subsection that's gaining massive steam called mention markets, where you literally are taking a position on how many times or if

someone even says a singular word. Trump

saying China, Powell saying inflation, CEO saying AI on earnings calls. All

priced like a market. And today I'm going to show you how I find an edge in those markets using my personal favorite tool, mention metrics without guessing,

without vibes only, and using actual data. If you haven't seen these,

data. If you haven't seen these, mentioned markets are questions like how many times will Jerome Powell say inflation in the next FOMC press meeting? Will Trump say China at least

meeting? Will Trump say China at least five times in this rally? How many times will Apple say AI on their next earnings call? And platforms like Poly Market,

call? And platforms like Poly Market, list these because they're the perfect blend of DGEN and data brain crossover.

Instead of betting on who wins or what a specific result will be for a release date, Rotten Tomato score, you're taking a play on how people talk. The number of times a word or phrase gets mentioned in

a speech, debate, presser, or earnings call. So where does mention metrics fit

call. So where does mention metrics fit in? Mention metrics is basically an

in? Mention metrics is basically an analytics tool built specifically for these mentioned markets. It tracks how often specific words or phrases show up in speeches, earnings [music] calls,

press briefings, debates, all of that stuff. It turns it into charts and stats

stuff. It turns it into charts and stats that you can actually use. You plug in a keyword like inflation or China. And

pick who you care about, Trump, Powell, Tesla, whatever, and it shows how many speeches mentioned it, the average mentions per speech, trends over time, and actual excerpts without the word in context. So instead of guessing, eh,

context. So instead of guessing, eh, feels like you'll be saying a lot lately, you have receipts. The way I like to use it starts from the market, not from the tool. Say you see a market that I like of how many times will

Powell say inflation on the next presser or will Trump mention China at least seven times in his upcoming rally. Each

outcome has a probability. Before you

take a position or touch anything at all, you want a feel for what the market is assuming. And as always, this video

is assuming. And as always, this video is educational only. It's not financial advice and prediction markets can be age restricted where you live. So always

respect local laws where your jurisdiction is and don't risk money you can't afford to lose. Now we go into mention metrics and build a baseline.

Let's stick with Powell and inflation as this example. Inside of mention metrics,

this example. Inside of mention metrics, you type your keyword inflation, filter by speaker, Jerome Powell or Federal Reserve speeches, and filter by category FOMC press conference or press gaggles.

You choose a time range like last 6 months or year to date. What you get back from mention metrics is the average mentions per speech, median mentions, how many speeches had at least one mention, a chart of these mentions over

time, and a list of past speeches with excerpts containing that word. That

alone already tells you if the market is capping. The question is basically, will

capping. The question is basically, will he say inflation at least three times, and your data says that Powell's averaged it 8 to 10 mentions in every presser this year? That's a completely different play than if he's only

averaged it 1.3 times. The real edge isn't just what's the average. It's did

the resume change? Mention metrics lets you look at mentions over time. Maybe

Powell barely sled inflation in early 2025. Then there were some nasty CPI

2025. Then there were some nasty CPI prints and he started hammering it every single speech. Same with Trump in China.

single speech. Same with Trump in China.

Maybe in 2024 it was every second sentence. He couldn't stop saying it.

sentence. He couldn't stop saying it.

But lately, he's been on a different rant. Or maybe the AI mentions exploded

rant. Or maybe the AI mentions exploded across earnings calls after Q1. And now

every CEO is farming that narrative, continuing to say it. So instead of just average equals five mentions, you're asking, has this been trending up or down? Are these spikes around specific

down? Are these spikes around specific topics or news? How does it correlate in the actual conversation that was going on with the transcript? And is the current environment more hypy or calmer than the historical average? If the

market's basically pricing the long-term average, but the last few speeches were way above or below that, that's where you start finding these mispricings in mention markets. The other spicy part of

mention markets. The other spicy part of mention metrics is its filters. You can

slice up mentions by speaker and look at Trump vers Biden, Powell vers Yellen, or category, earnings calls versus press briefings versus rallies versus debates and start to find where that trend is

showing. You can also look at tags, AI,

showing. You can also look at tags, AI, China, inflation, etc. That matters because people talk differently in different settings. CEOs might drop AI

different settings. CEOs might drop AI non-stop on earnings call, but barely in random interviews. Politicians might say

random interviews. Politicians might say China a ton at rallies, but not in formal speeches. and Powell might lean

formal speeches. and Powell might lean on inflation in an FOMC presser, but not in a random panel or interview. If the

market is about a very specific context, like during the post meeting press conference, you don't want to use all that context data. You want to filter it down to the category and maybe even if that speaker's recent behavior in that

exact format. Was it in a Q&A? Was it in

exact format. Was it in a Q&A? Was it in a predetermined set of public speaking?

Was it a speech that was written for them or an interview where they were posed a question? So, how do you actually turn this view into the market?

Here's a simple process I'd use if I were evaluating a mention market purely from a researching. Copy the market question exact wording, speech, event date. Look at mention metrics query that

date. Look at mention metrics query that matches that setup as closely as possible. So Powell FOMC, look at

possible. So Powell FOMC, look at previous FOMC's. Take a look at where

previous FOMC's. Take a look at where else those mentions may have come outside of FOMC's. And then look at the average and median mentions, recent trend, the last few events only, and any obvious resume change. Ask, is the

market pricing above or below what this data suggests? Is there a narrative

data suggests? Is there a narrative reason this time is different? Then

decide if you'd lean with or against the crowd or just pass completely. Most of

the edge is actually from saying no when the data in the price doesn't give you a clean story. A couple of reality checks

clean story. A couple of reality checks so no one copes in the comments later.

Historical mentions do not guarantee future behavior. People can change tone,

future behavior. People can change tone, get new talking points, or react to fresh news. Transcripts can be messy.

fresh news. Transcripts can be messy.

Different sources, different tagging, and markets can move fast right before or during the event as people react live. Mention metrics is just one edge.

live. Mention metrics is just one edge.

It's not a full-blown cheat code, but it does help you identify situations where history could repeat itself. And also,

again, this video is not financial advice. You're going to be taking an

advice. You're going to be taking an edge yourself. Everything that I've just

edge yourself. Everything that I've just outlined, how to use mention metrics, the things that I look for, and finding an edge in mention markets using this tool. Start from the market, build a

tool. Start from the market, build a datadriven baseline, look for changes, and match the context of the question.

If you want a follow-up or other products that I've been using to help identify opportunities in prediction markets, leave a comment below and let me know what it is you want to see me break down. As always, hit the

break down. As always, hit the subscribe, leave a like, and keep it easy. We'll catch you next time.

easy. We'll catch you next time.