yarofeyka

just trying to understand smth in crypto

@v3v_ventures research analyst | views are my own

[yarofeyka @yarofeyka]

it's all about grinding

[yarofeyka @yarofeyka]

@HTX_Studio hell yeah

[yarofeyka @yarofeyka]

@robcrt "bro they're all done u just trippin"

[yarofeyka @yarofeyka]

never kill yourself

[zek @zekramu]

two sides constantly at war

[yarofeyka @yarofeyka]

@WatcherGuru so its basically "series usa" round

[yarofeyka @yarofeyka]

https://t.co/aPZoZAfkyy

and u're still underestimating sol

[yarofeyka @yarofeyka]

funny how premarkets are only booming now after like 4 months of actual hype

[Stacy Muur @stacy_muur]

https://t.co/xuuUM5xJKZ

[PJ @Prithvir12]

Prediction Market Liquidity Algorithms

TLDR:

- Prediction market liquidity started with simple math formulas like LMSR and constant-product AMMs.

- These worked for bootstrapping but exposed operators to big risks.

- Now platforms are shifting to adaptive algorithms and order books that let mm's adjust spreads, balance inventory, and recycle fees.

- The next wave will be cross-market netting, parlays, and reinforcement-learning bots.

1. The Problem

Every financial market has wrestled with the same question: Who takes the other side of a trade?

> In equities, specialists and market makers.
> In futures, floor traders and clearing members.
> In options, liquidity-provision algorithms tied to volatility surfaces.

> In prediction markets, the tools are newer, the risks stranger, and the algorithms are still evolving.

- Each contract needs continuous two-sided quotes. Without liquidity, the market stalls. Too much noise, and spreads widen to useless levels.

- Unlike equities or FX, prediction contracts do not map cleanly to a cash flow or hedge. The market maker is not offloading risk into some deeper pool, it is the pool.

2. The First Generation: Constant Product and LMSR

- LMSR: Robin Hanson’s cost-function model priced trades directly. A single parameter “b” set liquidity: higher b meant smoother prices, lower b meant sharper moves.

- AMMs: DeFi later used Uniswap-style constant-product pools (x·y = k) so traders could always buy outcome tokens against reserves.

Both schemes solved bootstrapping but had weaknesses:
- LMSR exposes the operator to unbounded loss.
- Constant product bleeds capital if outcomes are skewed or liquidity is shallow.

3. The Second Generation: Adaptive Market Makers

To scale, platforms began experimenting with adaptive algorithms:
- Dynamic spread adjustment: widen or tighten quotes based on order flow.
- Inventory-sensitive curves: shift odds if too much open interest piles on one side.
- Fee recycling: route trading fees back into liquidity pools, extending runway.

4. Lessons from History

- In options, Black-Scholes gave a pricing framework; volatility surfaces emerged; market makers hedged dynamically.

- In ETFs, authorized participants arbitraged NAV vs. market, keeping spreads tight.

- In FX, algorithmic market makers optimized tick-by-tick inventories.

Prediction markets are retracing these steps, but without the luxury of deep hedging instruments. Their only hedge is time diversification (many markets) and fee income.

5. Where It’s Going

The frontier is liquidity provisioning algorithms that self-hedge:
- Cross-market netting: offsetting exposure across correlated events (for example, multiple election states).

- Parlay engines: combining contracts into baskets, reducing variance.

- Reinforcement-learning LPs: bots that adjust the “b” parameter, spreads, and inventory dynamically based on realized volatility of event odds.

[rob @robcrt]

“oh sorry haha i don’t have tg or x can we connect on linkedin?”

no

[Robert Miller @bertcmiller]

You don’t have to and probably shouldn’t comment on every last thing that’s happening on CT

You can, in fact, not reply

[yarofeyka @yarofeyka]

@stacy_muur hopefully that's at least insightful enough

[yarofeyka @yarofeyka]

@stacy_muur am i doing it right?

[yarofeyka @yarofeyka]

one of the most comprehensive resources for those who want to understand derivatives. solid material

https://t.co/ZXYOsDyfH6

@MixBytes

[yarofeyka @yarofeyka]

@ckyzhu '-> % of nyc uber driver shilling on coins'

is the metric fr

[yarofeyka @yarofeyka]

ngl gonna short eth after it reaches $4,890

prove me wrong

[yarofeyka @yarofeyka]

@burckmeister that's like calling solana an ethereum L2 because you can bridge to it 😭