October 31, 2026
Two very different kinds of markets showed up side by side in my watchlist snapshot on October 31, 2026. One is a political book with more resting depth than it trades in a day. The other is a stack of sports books that turned over several times their own order-book depth in a single session and then resolved overnight.
Neither is a trade idea. What interests me here — and what I think is the most useful takeaway in this week's polymarket analysis — is the ratio between how much money is sitting in the book and how much money moves through it. Those two numbers tell you completely different stories about the same platform.
The 2026 Balance of Power: D Senate, D House market prices Democrats taking both chambers. At the time of my snapshot it sat at 69.5% Yes / 30.5% No, with roughly $10.97M in lifetime volume, $1.27M in the prior 24 hours, and $6.16M in liquidity.
Divide a day's volume by resting liquidity and you get a crude velocity figure. Here it's about 0.21 — the book trades roughly a fifth of its own depth per session. That is an unusually patient structure. It means a participant can move size without walking the price very far, and it means the quoted number is being defended by capital rather than discovered by churn.
The price action matches. Up 2.0 points in 24 hours and 3.0 points over seven days, with Election Day on November 3. That's a book drifting, not repricing. When a market this deep moves in single digits during the final week, my read is that the flow arriving is roughly as informed as the flow already there — nobody is showing up with something the book hasn't already seen.
One structural note worth keeping in mind whenever you read joint-outcome prediction market odds: a "both chambers" book can never price above the lower of its two component probabilities. So 69.5% on the pair is an implicit statement that the weaker of the two chamber books is trading at 69.5% or higher. If you track the individual chamber markets, that relationship is a free consistency check — and the gap between the pair and the weaker leg is where correlation assumptions live.
Now the other end of the spectrum. The October 30 sports slate in my snapshot — the Houston–Dallas moneyline and its two spread lines, the Lakers–Warriors pair, Nuggets–Suns, and a Baylor–UCF college football book — carried roughly $7.93M of combined 24-hour volume on about $1.80M of combined liquidity.
That's a velocity near 4.4x: more than four dollars traded for every dollar resting. Against the midterm book's 0.21x, the two sit about twenty times apart on the same platform, in the same session.
Those games have since resolved, so there is nothing actionable in them and I'm not treating them as live. They're useful as a shape lesson. Event books with hard, imminent settlement times recycle capital constantly: market makers quote thin, traders flip in and out, and the entire lifetime volume of the market is often compressed into its final day. The Houston–Dallas moneyline alone showed $2.66M of 24-hour volume against $2.71M lifetime — essentially the whole book traded in one session.
Here's a detail I flag for anyone building their own tracking sheet. In this snapshot, Lakers vs. Warriors showed $2,051,760 of 24-hour volume against $1,905,466 of total volume. The 24-hour figure is larger than the lifetime figure, which is arithmetically impossible unless the two fields are computed over different rolling windows or refresh at different cadences. Nuggets vs. Suns showed the same pattern.
That is not a scandal — it's a reminder that any polymarket analysis built on scraped fields needs to know exactly what each field measures. When I see 24h > total, I stop treating the ratio as precise and use it only directionally. Methodology discipline beats a clean-looking number every time.
The only item from this set with a forward deadline is the balance-of-power book, with its resolution event on November 3. Three things I'll be watching in the final sessions, framed as research prompts rather than recommendations:
None of this is a trade recommendation, and trader execution on this project is currently off. It's observation and methodology — the same framework I apply to every book on the watchlist.
If you want the daily snapshots, the odds moves, and the catalyst dates as I log them, the free watchlist runs on our Telegram channel. It's open, it's free, and the journal stays transparent — come look over the numbers with fellow traders.
Velocity — daily volume divided by resting liquidity — separates books where capital is parked from books where capital is churning. A low figure (under 1x) suggests a deep, patient market where the quoted price is being defended rather than discovered. A high figure (4x or more) is typical of short-dated event markets that recycle the same capital repeatedly before settlement. Neither is better; they just require different reading.
Because a joint outcome requires every component to happen. The probability of A and B together is always less than or equal to the probability of A alone, and less than or equal to B alone. So a 69.5% price on a two-chamber sweep implies both underlying chamber markets are at 69.5% or higher. That relationship is a quick sanity check when you're comparing prediction market odds across related books.
It shouldn't, mathematically — so when you see it, the two fields are almost certainly computed over different windows or refreshed on different schedules. It's a data-pipeline artifact rather than a market signal. The practical response is to treat such ratios as directional only, and to avoid building precise conclusions on top of them.
Not necessarily. A deep book moving two or three points in a week can mean consensus, or it can mean the information that would move it simply hasn't arrived yet. Depth absorbs flow; it doesn't predict news. A thin book and a thick book can hold identical probabilities while carrying completely different sensitivity to the next headline.
No. Everything here is observation, methodology, and catalyst tracking from a timestamped snapshot. Trader execution on this project is currently off, and nothing above should be read as advice to enter or exit any position. Prediction markets carry real risk of total loss on any given contract.