# FIFO Price-Time vs Pro-Rata Matching Algorithm Mechanics

> Compare pure Price-Time FIFO against Pro-Rata, TOP, LMM, and Split FIFO/Pro-Rata matching algorithms across CME SOFR, Treasury futures, and US options.

Canonical: https://terminalledgerdesk.com/reports/fifo-price-time-vs-pro-rata-allocation-matching-algorithm-mechanics
Last updated: 2026-11-14T19:41:50.396Z

**Publication:** Terminal Ledger Desk
**Category:** Matching Engine Clob Colocation Ptp
**Published:** 2026-11-14T08:20:25.915Z
**Author:** Marcus Thorne, CQF (Certificate in Quantitative Finance (CQF), Low-Latency FIX & Matching Engine Architect)
**Reviewer:** Not recorded

## Answer-first summary

While equities and equity-index futures reward sub-microsecond queue entry via Price-Time FIFO, short-term interest-rate (SOFR) and options venues allocate fills via Pro-Rata size fractions, TOP overlays, and Split FIFO/Pro-Rata hybrids.

## Key takeaways

- Pure Price-Time FIFO executes sweeps in ~185 ns via a single linear slab walk, rewarding sub-microsecond arrival at newly opened price levels.
- Split FIFO/Pro-Rata and TOP overlays sustain deep liquidity on thick 1-tick STIR (SOFR) and options books where latecomers would otherwise never fill.
- Matching engines must implement Pro-Rata divisions exclusively with integer uint128_t floor arithmetic—never IEEE-754 floats—to keep HA state replicas bit-identical.
- Quoting Pro-Rata markets requires sizing dynamic over-quote factors carefully alongside hardware mass-cancel switches to survive unexpected 10x macro sweeps.

## Related financial entities

- **CME Aurora Financial Data Center Standard** (CME-AURORA) — product: Dedicated CME Group hosting facility in Aurora, Illinois, housing the Globex matching engine and offering direct low-latency colocation to clearing members.
- **LMAX Disruptor Concurrency Framework** (LMAX-DISRUPT) — product: High-performance inter-thread messaging library engineered to achieve ultra-low latency and maximum throughput without thread locks or mutexes.
- **Unified Multi-Asset Clearing Engine Standard** (UNIFIED-MARGIN) — product: Universal clearing engine that aggregates distinct financial asset classes into a single real-time valuation, netting collateral and exposures simultaneously.

## Analysis

Not every Central Limit Order Book rewards the fastest trader to arrive at a price level. While cash equities (Nasdaq, NYSE, Cboe) and CME E-mini S&P 500 (`ES`) run pure **Price-Time `FIFO`**—where `Order #1` must fill 100% before `Order #2` receives a single lot—short-term interest-rate (`STIR`) futures like **CME Three-Month SOFR (`SR3`)** and multi-listed US equity options (`Cboe`, `ISE`, `PHLX`) run **Pro-Rata** and **Split `FIFO` / `Pro-Rata`** algorithms (`CME Algorithm K, C, T, F`). Under Pro-Rata, arriving 50 milliseconds late still captures 40% of an incoming sweep if you quote 40% of visible book size—fundamentally inverting execution strategy, over-quoting risk, and matching-engine CPU math!

## Pure Price-Time FIFO (`Algorithm F`): Speed Races & Queue-Position Alpha

In pure Price-Time `FIFO` (`First-In, First-Out`), every price level is an [intrusive doubly-linked queue](/reports/intrusive-doubly-linked-order-book-o1-cancel-radix-price-indexing) sorted strictly by exchange gateway arrival sequence. Matching an aggressive `500`-lot sweep requires zero division instructions: the engine simply drains `head` orders sequentially until `500` lots fill.

Economically, `FIFO` incentivizes **sub-microsecond latency competition** at the exact tick when an old price level clears and a new tick opens—driving demand for [FPGA cut-through pipelines](/reports/fpga-tick-to-trade-cut-through-mac-pcs-transceiver-pipeline-budgets) inside [CME Aurora](/engines/cme-aurora-colocation-standard). However, on ultra-thick 1-tick spread instruments (like 3-Month SOFR `SR3` where `150,000+` contracts sit on the best bid for hours), pure `FIFO` paralyzes liquidity turnover because newcomers sitting behind `150,000` contracts have zero hope of filling without crossing the spread.

![CME Split FIFO/Pro-Rata multi-pass allocation waterfall showing TOP pioneer slice, LMM slice, FIFO split, floor Pro-Rata, and residual FIFO](/api/figure?type=flow&title=Multi-Pass%20Split%20FIFO%20%2F%20Pro-Rata%20Allocation%20Waterfall&subtitle=Deterministic%20uint128%20integer%20math%20across%20TOP%20-%3E%20LMM%20-%3E%20FIFO%25%20-%3E%20Pro-Rata%25%20-%3E%20Residual&steps=1.%20TOP%20Pioneer%3A%3APrice%20setter%20gets%20up%20to%20TOP_Cap%20%28100%25%29%7C2.%20LMM%20%2B%20FIFO%20%25%3A%3ADesignated%20MM%25%20%2B%20X%25%20queue-head%20FIFO%7C3.%20Pro-Rata%20Pass%3A%3Afloor%28Rem%20*%20q_i%20%2F%20Q_tot%29%3B%20drop%20if%20%3C%202%7C4.%20Residual%20Tail%3A%3ADistribute%201-lot%20remainders%20Size%2FFIFO&metrics=FLOAT%20USAGE%3A%3A0%25%20%28Banned%29%3A%3A100%25%20deterministic%20uint128%7CPRO-RATA%20MIN%3A%3A2%20Lots%3A%3ABlocks%201-lot%20split%20gaming%7CMATCH%20COST%20%28500%29%3A%3A1.42%20us%3A%3Avs%200.18%20us%20pure%20FIFO%20sweep%7COVER-QUOTE%20RISK%3A%3A5x%E2%80%9315x%3A%3ARequires%20instant%20SMP%2FPurge)

## Pure Pro-Rata, TOP (`Top Order`), LMM & CME Split `FIFO/Pro-Rata` Pass Cascades

To keep thick interest-rate and options books liquid, CME Globex and options exchanges execute multi-pass deterministic allocation cascades (`MatchAlgorithm` field in CME MDP 3.0 `MDInstrumentDefinition`):

1. **TOP (`Top Order Priority` Overlay)**: Before splitting lots proportionally, the **single resting order that established the new Best Bid/Offer first** receives `100%` of its size up to a venue `TOP_Cap` (e.g., `min(OrderQty, 200 lots)`), rewarding price-improving pioneers!
2. **LMM (`Lead Market Maker` Allocation)**: Designated market makers meeting continuous two-sided quoting SLAs capture a fixed contractual percentage (`e.g., 15%–40%`) ahead of general Pro-Rata.
3. **Split `FIFO` + `Pro-Rata` (`Algorithm K` — e.g., `40% FIFO / 60% Pro-Rata`)**: `X%` of the aggressor order fills `FIFO` from the queue head, and the remaining `(100 - X)%` distributes across *all* resting orders at that price via integer floor rounding: `alloc_i = floor(RemQty * restingQty_i / TotalLevelQty)`.
4. **Residual Lot Pass (`FIFO` or Largest-Order)**: Fractional rounding remainders (`1 lot` per order) go sequentially by size-then-time until `aggressorLeaves == 0`!

| Market / Product | CME / Venue Code | Allocation Waterfall | Primary Alpha / Execution Edge |
| :--- | :--- | :--- | :--- |
| **US Equities & CME E-mini (`ES`, `NQ`)** | `F` (`Pure FIFO`) | `100%` Price-Time Arrival | Sub-microsecond [Aurora](/engines/cme-aurora-colocation-standard) / Carteret queue entry |
| **CME 3-Month SOFR (`SR3`) & Eurodollar** | `K` / `C` (`Split` / `TOP+PR`) | `TOP` -> `LMM` -> `FIFO%` -> `Pro-Rata` -> `Resid` | Optimal quote sizing + instant sweep cancel |
| **CME US Treasury Futures (`ZF`, `ZN`)** | `F` or `K` (Contract-specific) | Pure `FIFO` (`ZN`/`ZB`) or Split (`ZT`) | Implied-spread + cash-futures basis speed |
| **US Equity Options ([Cboe](/platforms/colmex-pro-2-0-terminal-26), ISE, MIAX)** | `Customer-Overlay PR` | Public Customer `FIFO` -> `PMM/LMM` -> `Pro-Rata` | Mass-Quote `MMR` protection + Volatility skew |

## Exchange Matching Algorithm Matrix Across Global Asset Classes

Comparing venue-specific allocation waterfalls and resulting strategic equilibrium across US equities, CME futures, and US options.

```formula
Q_fifo_pass = floor( Q_agg * Pct_FIFO )
Q_pr_rem = Q_agg - Q_top - Q_fifo_pass
Alloc_PR(i) = ( floor( (Q_pr_rem * q_i) / Q_level_rem ) >= 2 ) ? floor( (Q_pr_rem * q_i) / Q_level_rem ) : 0
Q_residual_lots = Q_pr_rem - Sum_{i=1..N}( Alloc_PR(i) )

Worked Split 40% FIFO / 60% Pro-Rata Fill (No TOP, Q_agg = 500 lots, Resting = [A: 1000, B: 2000, C: 2000]):
 Pass 1 (40% FIFO = 200 lots) : Order A fills 200 lots -> Remaining resting: [A: 800, B: 2000, C: 2000] (Total = 4,800)
 Pass 2 (60% PR = 300 lots) : A gets floor(300*800/4800)=50; B gets floor(300*2000/4800)=125; C gets 125 (Sum = 300, Residual = 0)
 Final Total Fills : Order A = 250 lots (50%), Order B = 125 lots (25%), Order C = 125 lots (25%).
```

## Fixed-Point 128-Bit Integer Division & Over-Quoting Risk Controls

Inside a deterministic matching core (see our [Deterministic Matching Engines & Colocation Guide](/reports/deterministic-matching-engine-clob-colocation-ptp-clock-sync-guide)), **IEEE-754 floating-point math (`double`) is strictly forbidden** for Pro-Rata splits because minor rounding differences between Intel and AMD FPU pipelines or compiler flags could cause primary and backup state machines to diverge by 1 lot! Instead, matching engines compute exact integer quotients using `uint128_t` (`(__uint128_t)remQty * orderQty / totalLevelQty`).

For trading desks, Pro-Rata creates a dangerous **over-quoting trap**: to win `100` lots of an expected `1,000`-lot sweep on a `9,000`-lot book, a trader quotes `1,000` lots (`10%` book share). What if a macro shock unleashes a **`50,000`-lot sweep** that fills all `1,000` lots instantly—10x desired risk?! Institutional EMS gateways wire [CFTC](/standards/cftc)-mandated mass-cancel purges and size Pro-Rata tail exposure in the [Terminal Margin VaR & ES Calculator](/calculators/terminal-margin-var-es-calculator).

![Matching engine execution latency for a 200-order price-level sweep across Pure FIFO, TOP+FIFO, Split K, and Options Customer/LMM Pro-Rata](/api/figure?type=bars&title=Matching%20Core%20CPU%20Cost%20per%20200-Order%20Price-Level%20Sweep&subtitle=Single-core%20matching%20time%20as%20allocation%20pass%20complexity%20increases%20%28nanoseconds%29&steps=Pure%20Price-Time%20FIFO%20%28ES%29%3A%3A185%20ns%20%20%20Single%20linear%20slab%20walk%7CFIFO%20with%20LMM%20%2F%20TOP%20%28F%2FT%29%3A%3A340%20ns%20%20%20Pioneer%20tag%20%2B%20linear%20walk%7CCME%20Split%20FIFO%2FPro-Rata%20%28K%29%3A%3A1%2C180%20ns%20%20%203-pass%20uint128%20%2B%20residual%7COptions%20Customer%20%2B%20LMM%20%2B%20PR%3A%3A1%2C940%20ns%20%20%20Priority%20tier%20%2B%20PMM%20overlay&metrics=SOFR%20ALGORITHM%3A%3ASplit%20%2F%20PR%3A%3ADeep%20100k%2B%20lot%20STIR%20books%7COPTIONS%20PRIORITY%3A%3ACust%20First%3A%3APublic%20customer%20FIFO%20ahead%7CTOP%20REWARD%3A%3AFirst%20Setter%3A%3AIncentivizes%20narrowing%20BBO%7CRESIDUAL%20SORT%3A%3AQty%20-%3E%20Time%3A%3ADefeats%20micro-order%20clones)

## Minimum Pro-Rata Thresholds (`MinLot = 2`) & Tail-Split Gaming Prevention

If CME rounded fractional Pro-Rata shares *nearest* (`>= 0.5 -> 1`), a trader could split `1,000` lots into `1,000` separate `1`-lot orders to steal rounding residuals! Exchanges defeat order-splitting gaming via two strict rules:
- **Strict `floor` Truncation + Minimum `2-Lot` Threshold**: On CME Pro-Rata books, any computed `floor` share `< 2 lots` (`0` or `1`) receives **zero** lots during the Pro-Rata pass (`MinAlloc = 2`), deferring tiny quotes to the residual FIFO tail!
- **Size-Then-Time Residual Tie-Breaking**: Residual lots allocate largest-resting-order-first, rewarding genuine size commitment on [FIX API and iLink 3](/software/fix-api) sessions. Explore more order-book mechanics in the [Matching Engine, Colocation & PTP Topic Hub](/topics/matching-engine-clob-colocation-ptp).

## Frequently asked questions

### What is a TOP (Top Order Priority) overlay in CME Globex Pro-Rata matching?

When a trader places the very first order that improves the market to establish a brand-new Best Bid or Best Offer, CME stamps that order with Top Order status—guaranteeing it fills 100% (up to a contract-defined cap/min) before general Pro-Rata size splitting begins.

### Does modifying order quantity upward lose TOP status or FIFO priority on Pro-Rata books?

Yes. Increasing order quantity cancels TOP pioneer status and resets the order timestamp to the tail of the FIFO residual queue, whereas decreasing order quantity preserves existing time priority (subject to minimum TOP size retention rules).

### Why do US equity options exchanges prioritize Public Customer orders ahead of Pro-Rata market makers?

Under SEC-approved options exchange rules on Cboe, Nasdaq ISE/PHLX, and MIAX, non-professional Public Customer limit orders resting at the NBBO execute in Price-Time FIFO ahead of both Lead Market Maker participation guarantees and Pro-Rata professional quotes.

### How do trading algorithms manage over-quoting risk on CME SOFR Pro-Rata books?

Algorithms monitor real-time cumulative sweep volume across correlated expiries via CME MDP 3.0 Feed A/B; the microsecond sweep volume crosses a preset threshold or adjacent contract ticks break, an iLink 3 OrderMassActionRequest (35=q) purges remaining inflated quotes.


## Source URLs

- https://www.cmegroup.com/confluence/display/EPICSANDBOX/CME+Globex+Matching+Algorithms
- https://www.ecfr.gov/current/title-17/chapter-I/part-38
- https://cdn.cboe.com/resources/regulation/rule_book/C1_Exchange_Rule_Book.pdf

## Citation

- APA: Marcus Thorne, CQF. (2026). FIFO Price-Time vs Pro-Rata Matching Algorithms: CME Split, LMM & TOP Mechanics. Terminal Ledger Desk. https://terminalledgerdesk.com/reports/fifo-price-time-vs-pro-rata-allocation-matching-algorithm-mechanics
- Canonical article: [FIFO Price-Time vs Pro-Rata Matching Algorithms: CME Split, LMM & TOP Mechanics](https://terminalledgerdesk.com/reports/fifo-price-time-vs-pro-rata-allocation-matching-algorithm-mechanics)
- Site research index: [Terminal Ledger Desk](https://terminalledgerdesk.com/llms-full.txt)
