Where does Auction Market Theory come from?

It was popularised alongside Market Profile through J. Peter Steidlmayer's work at the Chicago Board of Trade in the 1980s. The underlying idea, a continuous double auction, is older and belongs to economics. There is no single canonical text, which is why the versions you meet differ.

Is Auction Market Theory testable?

Not as a whole. Individual operationalisations are: define acceptance as a fixed number of brackets held above a level, and you have something countable on your own data. The frame only becomes falsifiable once you replace its adjectives with thresholds.

Does it apply outside futures?

It applies wherever a central limit order book matches buyers and sellers continuously, which covers equities and most crypto venues. It fits badly where prices are quoted by dealers rather than auctioned, and where liquidity is split across venues that never see each other.

Glossary

Auction Market Theory

Also called: AMT · double auction theory

Auction Market Theory holds that a market is a two-way auction searching for prices where business gets done. Price advertises, volume responds. When the search finds acceptance, the market builds value; when it finds rejection, it moves on. It is a reading frame, not a set of entry signals.

The double auction

Two auctions run at once. Buyers compete with buyers, sellers with sellers, and price is the mechanism that advertises the imbalance. Price rises to find sellers and falls to find buyers, and it keeps moving in one direction until enough of the other side responds.

That response is the observable part. On a footprint chart or in the depth of market, you are watching whether an advertised price is being answered or ignored.

Acceptance and rejection

Where the auction finds two-sided business it lingers: time accumulates, volume builds, a distribution forms. That region is value. Where it finds nothing it leaves quickly, and the trace is thin — few prints, little time, a pinch in the volume profile.

Acceptance and rejection are therefore statements about how much business was done at a price, not about direction. A market can accept a price on the way down exactly as readily as on the way up.

What the frame actually buys you

It converts a chart into three questions with observable answers. Who is initiating? Is the move being answered with volume, or ignored? Where did business stop? Market Profile and volume profiles exist to answer them. The theory answers nothing itself — it tells you what to look at.

A worked example

In a synthetic ES session, price holds between 5 312.00 and 5 320.00 for four hours, then trades to 5 331.00 in eleven minutes on 14 200 contracts. Over the next two hours it does 96 000 contracts between 5 326.00 and 5 331.00 and never trades back below 5 322.00.

The eleven minutes advertised a higher price. The two hours were the answer. The asymmetry is the point: the move cost 14 200 contracts, the acceptance took nearly seven times that.

The trap

The theory explains every outcome, so it forbids none. A rally that holds was acceptance; the identical rally that fails was rejection. Both readings are always available, and which one you reach for is decided by what happened afterwards. Narrated backwards, Auction Market Theory is never wrong — which is precisely what makes it useless as a signal.

The fix is to force it to commit. "If this is acceptance, price holds above 5 326.00 for the next two brackets" can fail, and you can count how often it does. "The market is seeking value" cannot fail, so it teaches you nothing, however true it sounds.

Frequently asked

Where does Auction Market Theory come from?: It was popularised alongside Market Profile through J. Peter Steidlmayer's work at the Chicago Board of Trade in the 1980s. The underlying idea, a continuous double auction, is older and belongs to economics. There is no single canonical text, which is why the versions you meet differ.
Is Auction Market Theory testable?: Not as a whole. Individual operationalisations are: define acceptance as a fixed number of brackets held above a level, and you have something countable on your own data. The frame only becomes falsifiable once you replace its adjectives with thresholds.
Does it apply outside futures?: It applies wherever a central limit order book matches buyers and sellers continuously, which covers equities and most crypto venues. It fits badly where prices are quoted by dealers rather than auctioned, and where liquidity is split across venues that never see each other.