Volume concentration in event-contract ladders

Most event-contract markets aren't one market — they're a ladder. "Mariners win 80+ games this season" is one rung. "Mariners win 75+" and "Mariners win 85+" are adjacent rungs. The pricing on the popular threshold is what gets discussed. The pricing on the unpopular thresholds is almost entirely noise.

The example

From our quote data the morning of May 27, 2026: the Seattle Mariners "wins this season" ladder on Kalshi looked like this:

• 70+ wins: 96.0% · 18 contracts traded in 24h
• 75+ wins: 90.0% · 9 contracts traded in 24h
• 80+ wins: 70.0% · 200 contracts traded in 24h
• 85+ wins: 28.0% · 14 contracts traded in 24h
• 90+ wins: 8.0% · 6 contracts traded in 24h

The 80+ rung carries roughly 80% of the ladder's total volume — 200 contracts versus 47 combined for the other four rungs. This isn't market efficiency choosing where to trade; it's reflection of where the public marketing was done. Kalshi advertised the 80+ market on social, news outlets cited 80+ wins as the "will the Mariners be good" threshold, and traders followed the headline.

Why this matters

Price discovery in any market requires actual trades. When a market has 200 contracts trading per day, every bid/ask quote competes against real money. The mid-price reflects what marginal buyers and sellers actually believe.

When a market has 9 contracts trading per day, the mid-price reflects whatever stale limit orders one or two retail traders left in the book. A new buyer can move the implied probability 5-10 percentage points by hitting the ask once. The price is there but the information contentisn't.

Concretely: the 75+ wins rung shows 90% implied probability. That looks tight — 6 percentage points off the 80+ rung which is at 70%. But that 90% is based on 9 contracts. The real-information probability of 75+ wins, given the 80+ rung's 70% pricing, is probably more like 85-92%. The exact answer depends on the joint distribution, and the 75+ rung's posted 90% is consistent with that range — but it could just as easily be off by 3-5 percentage points. The number you see is more noise than signal.

The HiveMind insight

We've been reposting alerts on the loud rung ("80+ wins at 70%, +6.5pp in 24h on $361k volume") and the HiveMind account has been calling these out — correctly. The biggest move in this ladder isn't the 80+ rung. It's the 2+ wins direction (when the favorite increases their lead, the 90+ rung repricing matters more than the 80+ which has effectively settled). The alert that surfaces the loud rung misses where the actual price discovery is happening.

What HiveMind is identifying — and what our updated alert system now captures — is that the SHARPER signal is on the rungs where the price is genuinely uncertain (15% to 85% implied) and the volume is meaningful (≥250 contracts/24h). The boring rung at 90+% is settled. The exciting rung at 8% is noise. The middle rung carrying meaningful volume is where the actual market discovery is happening.

How to read a ladder

When you encounter an event-contract ladder, do this:

1. Sort by implied probability descending
2. Drop any rung in [0%, 18%] or [82%, 100%] — settled or settling, no signal
3. Of the remaining rungs, identify the one with the most 24h volume
4. That's the price-discovery rung. Treat the others as illustrative, not pricing.
5. Look for monotonicity violations (T20 priced higher than T10) — these are market-maker mistakes, occasionally arbable

Our updated alert system follows this exact algorithm. Movement alerts now report the rung with the largest absolute information move, not the rung with the largest percentage move. They're different things, and the difference is where the analytical edge lives.

Cross-platform implications

The volume-concentration pattern is even more pronounced on Polymarket than on Kalshi because Polymarket users skew toward fewer-but-larger positions. A typical Polymarket sports ladder might have one rung carrying 95% of the volume and four rungs carrying combined <5%. Those four rungs are essentially decoration.

For sportsbooks, the concentration is different in shape but similar in implication. DraftKings player props have ~20 alternate lines per stat per game, but 80% of the action is on the headline line. The alternates exist for shopping opportunities (matched betting, low-hold arbs) but their prices are not informationally efficient and should be read as quote-shopping data, not as "the market thinks X."

Practical takeaway

Three rules when looking at any event-contract or alternate-line ladder:

• Volume below ~50 contracts per 24h = read the price as illustrative, not authoritative.
• Adjacent-rung prices that look weirdly tight (1-3pp apart) on thin rungs are usually just stale limit orders, not actual market consensus.
• The rung with the most volume IS the consensus price for the underlying thesis. The other rungs are derivatives.

When you see SportsBookISH alerts highlighting a particular rung as the "biggest move," we've already filtered for the rung where the volume justifies treating the move as signal. You can trust the call. The alternative rungs in the same ladder may be moving in the same direction, but those moves are mechanical — not new information.