November 17, 2026
Most days the board is dominated by things that resolve soon β a rate decision, a race, a series. Today's two most interesting books on my watchlist both resolve on dates nobody can circle on a calendar yet. One of them moved 20.9 points in a week. The other hasn't moved at all, and is sitting on more posted liquidity than it has lifetime volume.
That combination β a far-horizon book repricing hard, and a floor-priced book that refuses to breathe β is the kind of contrast that makes a polymarket analysis worth writing down. Snapshot taken November 17, 2026.
The market labeled "Will Anthropic's market cap be less than $1.25T at market close on IPO day?" prices "Yes" at 36.4%, with "No" at 63.6%.
The movement is the story. Up 14.4 points in 24 hours and 20.9 points on the week, on $319K of 24-hour volume against $1.49M lifetime. That's roughly 21% of the book's entire history turning over in a single session β a velocity number you normally see around a scheduled catalyst, not around a company that has no confirmed listing date in the market text.
Read literally, the board now implies a roughly 64% chance that the first closing print lands at or above $1.25 trillion, down from something closer to 85% a week ago. In other words, the market has been walking its expected debut valuation down, or at least widening the distribution around it.
Two mechanical things I'd check before treating that as a signal about Anthropic itself:
The honest framing: a 20-point weekly move with no visible scheduled event is a research prompt, not a conclusion. The first thing to pull is the resolution text β specifically what happens if there is no IPO by the deadline, and whether "market cap" is defined on basic shares, diluted shares, or something else. Valuation-threshold markets live and die on that definition. Not a trade recommendation.
Now the opposite end of the board. "Will Scott Colom win the 2028 Democratic presidential nomination?" sits at 0.1% "Yes" / 99.9% "No". Day change: +0.1 points. Week change: +0.1 points. Functionally, it hasn't moved.
What's odd is the plumbing. Liquidity is listed at $1,244,518 β against $866K of total lifetime volume and $301K in the last 24 hours. A book with more posted depth than it has ever traded is unusual, and it tells you something about how these candidate fields are built.
In a large nomination field, the event is a basket of individual yes/no books that all share the same resolution date. Market makers typically quote the entire slate, so each individual candidate inherits depth from the field-level posture rather than from interest in that specific name. The result: a candidate almost nobody is betting on can still show seven figures of quoted liquidity near the floor.
The practical consequences are worth spelling out in any prediction market odds review:
I track these mostly as a calibration exercise: how does the residual behave when a field adds or drops a name? The sum of a candidate slate rarely lands exactly at 100%, and the drift in that residual is one of the more honest measures of how tightly a board is being maintained.
Put the two side by side and you get a clean lesson about depth versus conviction. The Anthropic book has strong conviction (people are actively changing their minds, fast) and shallow depth. The nomination book has deep quotes and zero conviction β nothing to discover, so nothing moves.
Neither state is better. But they fail in different ways. Thin-and-moving books punish size; deep-and-frozen books punish patience. A useful polymarket analysis habit is to label every market on your sheet as one or the other before you look at the number at all.
What I'm watching from here: whether the Anthropic band holds above the 36% area after the burst of flow settles, and whether the 24-hour volume on the nomination field keeps running at a third of its lifetime total β because that pattern usually means the slate is being restructured somewhere.
I keep a free watchlist of books like these β far-horizon conditionals, floor-priced tails, and anything where liquidity and volume disagree β in our Telegram channel. It's observation and methodology, posted transparently, with no signals and no performance claims. Come argue with the reads; fellow traders there usually catch what I miss.
It means the book implies roughly a 36% chance that Anthropic's market capitalization is below $1.25 trillion at the close of its first trading day, and about a 64% chance it finishes at or above that line. Because it's conditional on a listing happening inside the resolution window, part of that price reflects timing uncertainty rather than valuation uncertainty.
In multi-candidate fields, liquidity is often quoted across the whole slate rather than name by name. An individual longshot inherits that depth even if almost nobody is trading it specifically. High posted liquidity on a floor-priced book says more about how the event is maintained than about genuine two-sided interest.
Not on its own. On a book with $100K of liquidity, a few hundred thousand dollars of flow can move the price substantially without any new information entering the market. The useful next step is a catalyst check: look for a dated event, a resolution-text change, or a credible report that lines up with the move. If none exists, treat it as repricing.
As probability estimates with known biases. Far-horizon books carry time-value drag, floor-priced books under-price tails, and thin books overshoot. Reading them as a forecast rather than a recommendation is the whole point β the number is an input, not an instruction.
No. Everything here is research and observation on publicly visible prediction market odds, including volume, liquidity, and resolution structure. Trader execution is off, and nothing in this post should be read as advice to enter a position.