MasscrestMasscrest
Start trial

Why Masscrest · the price-impact argument

“There is always a buyer and a seller.”

Standard economic theory

Every trade has two sides. The sides don't cancel, because markets are inelastic and price impact turns on four things.

1

Two sides, not the same.

Price impact turns on where the trade comes from, whether the market saw it coming, how large it is against what can be absorbed, and whether it forces the next print.

Origin

A dealer hedging and a fund building a position are not the same trade.

Predictability

Flow the market can see coming is already in the price.

Size

What matters is size against what the market can absorb.

Propagation

Some prints force the next one: dealer hedges, margin calls, stop-outs.

2

The positioning data landscape is behind.

Every source a desk reaches for today falls short on coverage, relevance, accuracy or clarity.

Source
13F filings
Nature
Long equity holdings for >$100M managers.
Gap that matters
Quarterly data, 45 days lag, low quality.
Source
Short interest
Nature
Aggregate short position.
Gap that matters
No investor segmentation, partial.
Source
Fund flows
Nature
Fund flows and allocation.
Gap that matters
AUM of active funds has declined, not speculative nature, partial.
Source
CFTC COT
Nature
Non-commercial positioning in futures.
Gap that matters
Futures only. Weekly, lack of clear categorisation.
Source
Infer trade side
Nature
Infer trade side from bid/ask spread or similar methods.
Gap that matters
Unreliable, a large share of option trades are misclassified.
Source
Put/call ratio
Nature
Track put and call volumes.
Gap that matters
Naive, investors sell options extremely often.
Source
Dealer-gamma estimates
Nature
Estimate dealer gamma with call-put OI.
Gap that matters
Also naive, relies on unrealistic assumptions.

3

Options aren't only a hedging wrapper.

Informed positioning shows up in options first, and the print propagates non-linearly into the underlying.

  • Institutional speculation

    Options are a speculative instrument on institutional books, not only a hedging wrapper. Informed positioning often takes place in the option market.

  • Granular

    Greek strategies and trades in specific tenors or strikes reveal what investors expect and how they behave.

  • Volumes

    Large turnover, with a stable increase in volumes from all investor cohorts.

  • Amplification

    Large implied leverage, and dealer hedging that propagates a single option print non-linearly into the underlying.

Exhibit 1
Monthly gross notional delta traded, US equity options · US$BN · by investor type
InstitutionalRetail
01,0002,0003,0004,0005,0006,0002020202120222023202420252026
Source: Masscrest.Hover for monthly figures

Over 99% of option trades take place against market makers, which allows a clear interpretation of the side intention.

Retailself-directedInstitutionalAsset managers, prop. tradersMarket makerclears ~99% of printsbuybuysellsellBuy − Sell =Buy − Sell =Buy − Sell =Net retailNet market makerNet institutional

4

Call-to-put is misleading.

A worked example on NVDA. Raw call and put volumes say one thing; who bought and who sold says another.

Exhibit 2a
NVDA · call to put ratio
21D sum of call minus put contracts traded, mn shares
-1,00001,0002,0003,0004,0002020202120222023202420252026
Exhibit 2b
NVDA · net trading imbalance
Investors (Buy call − sell call) − (Buy put − sell put) vs market makers · 21D, mn shares
-300-200-10001002002020202120222023202420252026
Correlation
−0.48

Between raw call/put volume and net call/put positioning.

High call volume doesn't mean bullish positioning.

5

What good looks like.

A positioning dataset earns its price when it clears six tests.

Coverage

Complete along at least one dimension: asset class, product, region or investor type.

Timely

Daily or intraday, delivered at the same cadence.

Relevance

Flows big enough to move price.

Clear segmentation

Investor classes that genuinely behave differently.

Accuracy

Benchmarked against independent data.

Signal

Proven to improve decisions, or to generate alpha.

6

Why supply is scarce.

Demand has grown with computing capacity and in-house data teams. The constraint is on the supply side, and it is structural: everyone who holds the data has a reason not to sell it.

The adverse selection problem

If a dataset is for sale, its owner could not monetise it directly. If it is not, the owner is trading on it. Either way, the data most worth having is the least likely to reach you.

Who holds the data, and why it stays private
Asset managers
Disclosing where they are invested dilutes the informational edge they are paid to have.
Dealers
Their business is improving execution and attracting client flow; selling client data would carry an asymmetric risk.
Clearing houses
Bound by what their members will tolerate being published.
Anyone with predictive data
Has an obvious alternative use for it: trading their own book.

7

Why we are different.

Masscrest data comes out of classification algorithms, proprietary features, and relationships with third-party providers.

No book against our clients

We do not trade, execute or clear. The conflict at the centre of the supply problem is not there.

The core business is alternative data, not trading.

The focus is providing accurate, contextualised data rather than trading on it ourselves.

Not built on unrealistic assumptions or modelled data.

The data are built to be accurate, not to be proxies.

Start trialSee the product