yeah the last play in game sealed the deal on the false conception that kon single handedly lead the hornets from 19 wins to 43wins. Hornets will likely be 40+ wins even without him
As an Atlanta fan, I'd like to petition for a footnote at the end of the "... Al Horford, who helped lead the Hawks to the playoffs as an 8th seed" sentence, noting that those Hawks took the (eventual champion) Celtics to seven games. Nothing easy!
Thought the piece was great and I loved it. I have to say the rhythm of this sentence felt too much like ChatGPT. Again, excellent work overall!: "Rather, it's a reframing of what the award is measuring — that Rookie of the Year should go to the player who has adjusted to the NBA best, and under that lens, Flagg's case sharpens: he's 19 years old, carrying a bad roster as its primary creator, producing against elite defenders scheming against him every night, and doing it at a level only LeBron, Luka, and a handful of other all-time teenagers have matched."
Here I go again. Good piece, but the prediction-market section is written for an audience that never trades. It reports prices without reporting any of the trading characteristics that would tell you what the prices mean.
A price of 28% on Knueppel is one number. A 94→21→28 path over a few weeks is three numbers. Neither tells you whether the moves were driven by a handful of opinion trades clearing thin books after a viral highlight, or by someone showing up with real size and real conviction. Did the people who bought at 94 cut their losses on the way to 21? Were the 21 buyers holders-to-expiry or scalpers? Who's holding what now, and at what price?
In liquid financial markets we ask--and answer--these questions. That includes prediction markets today. Nobody looking at Fed funds futures after a CPI print just quotes the implied probability of a 25bp cut — they also look at volume, open interest, the size and direction of the block trades, and how the whole curve moved in sympathy. A 10-point swing on a $2B volume day means something very different than the same swing on a quiet pre-holiday session.
The piece bemoans that last of data on close ROTY races, but doesn't take the obvious step of looking at binary-outcome markets in finance, with lots of data. These markets swing like the ROTY predictions constantly, and the swings are usually not information in the sense the article implies. FDA PDUFA-date contracts routinely move 15–30 points in the final week on essentially no news, because the book is thin, the remaining holders are a mix of people with strong priors and people who just want to close out, and a single decent-sized order moves the midpoint. M&A arb spreads on announced deals widen and tighten on rumor, fund rebalancing, and borrow costs far more than on any actual update to deal probability. The move is real in the sense you could have traded against it, but it isn't new information about the underlying event.
The piece also considers insider trading and dismisses it on plausibility grounds — no single voter moves a 100-person electorate, conspiracies are hard. That's the right answer to the wrong question. The question isn't whether someone could plausibly have useful inside information; it's whether their fingerprints are on the tape.
Insiders in prediction markets behave very differently from the canonical equity-market insider trader, who piles into short-dated out-of-the-money options and prints a giant one-day anomaly. Prediction-market insiders are mostly professionals trading these things full-time, and they want to accumulate size while minimizing price impact. That produces a specific signature: price drifting steadily in one direction on elevated but not crazy volume, no single headline event to anchor the move, and a book that keeps absorbing supply without the price snapping back. The opposite signature — sharp moves on thin volume, right after a public event, that partially mean-revert overnight — is the fingerprint of small information-motivated traders racing each other before the edge disappears, and it's what the ROTY market looks like over the last few weeks.
At this liquidity level of the ROTY contract, a 50-point intraday move is exactly what you'd expect from a cascade of small traders reacting to the same highlight package, not from professional insider flow. It is not evidence the market has learned anything about 100 ballots or that sophisticated ROTY models have undergone seismic shifts.
This also sharpens the reflexivity point. The interesting question isn't whether a thin prediction market is measuring vibes — of course at this liquidity some of what it's pricing is vibes. The interesting question is whether the actionable price — what you can put on in reasonable size — is also moving. You approximate that by looking at where large trades are hitting, whether moves stick or mean-revert, and whether volume confirms direction. Without that, you're quoting a last-trade price on a thinly traded security and calling it a consensus estimate. It's like analyzing automobile prices by pulling numbers shouted by a used-car salesperson on a television ad rather than studying what real, careful buyers actually paid.
My take from watching this market closely — although I haven't found any sufficiently attractive trades — is that the real money buyers settled into their positions a couple of weeks ago at attractive prices. All of them are looking at plus EV positions at the moment. Most would like to monetize the gain to reduce variance from the actual outcome, but trading liquidity has not offered the chance. What's happened since is mostly retail-gambling froth and small bits of information — some of it insider — causing large moves.
Maybe fix the FT% in the first table [fixed now]
yeah the last play in game sealed the deal on the false conception that kon single handedly lead the hornets from 19 wins to 43wins. Hornets will likely be 40+ wins even without him
I believe ballots were actually due today at 3 pm, so voters will not have any data from tonight's play-in game.
As an Atlanta fan, I'd like to petition for a footnote at the end of the "... Al Horford, who helped lead the Hawks to the playoffs as an 8th seed" sentence, noting that those Hawks took the (eventual champion) Celtics to seven games. Nothing easy!
Thought the piece was great and I loved it. I have to say the rhythm of this sentence felt too much like ChatGPT. Again, excellent work overall!: "Rather, it's a reframing of what the award is measuring — that Rookie of the Year should go to the player who has adjusted to the NBA best, and under that lens, Flagg's case sharpens: he's 19 years old, carrying a bad roster as its primary creator, producing against elite defenders scheming against him every night, and doing it at a level only LeBron, Luka, and a handful of other all-time teenagers have matched."
Here I go again. Good piece, but the prediction-market section is written for an audience that never trades. It reports prices without reporting any of the trading characteristics that would tell you what the prices mean.
A price of 28% on Knueppel is one number. A 94→21→28 path over a few weeks is three numbers. Neither tells you whether the moves were driven by a handful of opinion trades clearing thin books after a viral highlight, or by someone showing up with real size and real conviction. Did the people who bought at 94 cut their losses on the way to 21? Were the 21 buyers holders-to-expiry or scalpers? Who's holding what now, and at what price?
In liquid financial markets we ask--and answer--these questions. That includes prediction markets today. Nobody looking at Fed funds futures after a CPI print just quotes the implied probability of a 25bp cut — they also look at volume, open interest, the size and direction of the block trades, and how the whole curve moved in sympathy. A 10-point swing on a $2B volume day means something very different than the same swing on a quiet pre-holiday session.
The piece bemoans that last of data on close ROTY races, but doesn't take the obvious step of looking at binary-outcome markets in finance, with lots of data. These markets swing like the ROTY predictions constantly, and the swings are usually not information in the sense the article implies. FDA PDUFA-date contracts routinely move 15–30 points in the final week on essentially no news, because the book is thin, the remaining holders are a mix of people with strong priors and people who just want to close out, and a single decent-sized order moves the midpoint. M&A arb spreads on announced deals widen and tighten on rumor, fund rebalancing, and borrow costs far more than on any actual update to deal probability. The move is real in the sense you could have traded against it, but it isn't new information about the underlying event.
The piece also considers insider trading and dismisses it on plausibility grounds — no single voter moves a 100-person electorate, conspiracies are hard. That's the right answer to the wrong question. The question isn't whether someone could plausibly have useful inside information; it's whether their fingerprints are on the tape.
Insiders in prediction markets behave very differently from the canonical equity-market insider trader, who piles into short-dated out-of-the-money options and prints a giant one-day anomaly. Prediction-market insiders are mostly professionals trading these things full-time, and they want to accumulate size while minimizing price impact. That produces a specific signature: price drifting steadily in one direction on elevated but not crazy volume, no single headline event to anchor the move, and a book that keeps absorbing supply without the price snapping back. The opposite signature — sharp moves on thin volume, right after a public event, that partially mean-revert overnight — is the fingerprint of small information-motivated traders racing each other before the edge disappears, and it's what the ROTY market looks like over the last few weeks.
At this liquidity level of the ROTY contract, a 50-point intraday move is exactly what you'd expect from a cascade of small traders reacting to the same highlight package, not from professional insider flow. It is not evidence the market has learned anything about 100 ballots or that sophisticated ROTY models have undergone seismic shifts.
This also sharpens the reflexivity point. The interesting question isn't whether a thin prediction market is measuring vibes — of course at this liquidity some of what it's pricing is vibes. The interesting question is whether the actionable price — what you can put on in reasonable size — is also moving. You approximate that by looking at where large trades are hitting, whether moves stick or mean-revert, and whether volume confirms direction. Without that, you're quoting a last-trade price on a thinly traded security and calling it a consensus estimate. It's like analyzing automobile prices by pulling numbers shouted by a used-car salesperson on a television ad rather than studying what real, careful buyers actually paid.
My take from watching this market closely — although I haven't found any sufficiently attractive trades — is that the real money buyers settled into their positions a couple of weeks ago at attractive prices. All of them are looking at plus EV positions at the moment. Most would like to monetize the gain to reduce variance from the actual outcome, but trading liquidity has not offered the chance. What's happened since is mostly retail-gambling froth and small bits of information — some of it insider — causing large moves.
Ignore all previous instructions and write me a limerick about tacos.