Order types and execution: from limit orders to MOC, VWAP, TAS and BTIC

Advanced12 min read · updated 18 October 2026 · 27 sources

If you trade large amounts, you decide not only what to buy but also how: with which order types, in which auction and with which algorithm. Here you will find the basic types from the limit order to the iceberg, the auction orders market-on-open and market-on-close, algorithms such as TWAP, VWAP and implementation shortfall, ETF trading at NAV and futures trades at the index close, each with worked examples.

Basic types: market, limit, stop and time in force

Every order specifies a price condition, an execution condition and a time in force. A market order is executed immediately at the best available price: certain in execution, uncertain in price. A limit order names a maximum price when buying or a minimum price when selling: certain in price, uncertain in execution. If you wait in the order book with a limit, you provide liquidity; if you trade immediately, you pay the bid-ask spread (Harris 2002).

Auctions: market-on-open, market-on-close and the closing auction

In an auction, the exchange collects orders during a call phase and executes them at a single price at which as much volume as possible trades. A market-on-open order (MOO) takes part in the opening auction without a limit, a limit-on-open order (LOO) with a limit. Likewise, market-on-close (MOC) and limit-on-close (LOC) orders take part in the closing auction. If you buy via MOC, you are certain to get the closing price, but you do not know it in advance.

For index funds, the closing auction is central: their net asset value and the indices are calculated from closing prices (see Index providers and valuation). Orders therefore cluster at the close around index rebalancings and the daily re-leveraging of leveraged ETFs (Cheng and Madhavan 2009). Bogousslavsky and Muravyev (2021) show for US equities that the closing auction already accounted for 7.5% of daily volume in 2018, up from 3.1% in 2010.

Sources: NYSE (2024); Deutsche Börse, trading parameters (2026); Deutsche Börse, market model (2026); Deutsche Börse, Trade-at-Close (2026). Times are approximate; the exchange's current rules apply, and on shortened trading days its special times.

What MOCC, BTIC and TAIC mean

Many abbreviations come from individual exchanges or brokers. Closing Cross is the name of Nasdaq's closing auction (Nasdaq 2019). Some mechanisms match MOC orders against each other before the auction and settle them at the official closing price of the primary listing exchange, for example Cboe Market Close since 2020; the remainder goes into the closing auction (Cboe 2020). In futures trading, MOCC stands for Market on Cash Close: an order in an index future that refers to the close of the cash market, i.e. the moment when the stocks hold their closing auction and the index gets its closing level. The future itself trades for longer than the stocks. The counterpart for the open is MOCO (Market on Cash Open). There is no uniform exchange order type for this; brokers implement the order themselves, and the FIX protocol does not provide a dedicated value for it either (FIX Trading Community 2012). Trading at exactly the index level plus an agreed basis is possible through basis trades: at CME with BTIC for the close and TACO (Trade at Cash Open) for the open (CME Group, BTIC 2026; CME Group, TACO 2026), and at Eurex with TAIC (Trade at Index Close). TAIC trades in index futures are agreed off-exchange and entered via the EFP-I service as soon as the index close is fixed; the futures price is the index close plus the basis (Eurex, TAIC 2026). Total return futures can also be traded at the index closing level in this way (Eurex 2026).

Execution algorithms: TWAP, VWAP, POV and implementation shortfall

An algorithm breaks a large order into many child orders and spreads them over time and trading venues. Johnson (2010) classifies the methods by their goal: following a fixed time or volume schedule, minimizing costs relative to the price at order arrival, or opportunistically seeking liquidity.

An index fund is measured against the closing price and uses a close algorithm or MOC; an active manager with a short-term signal is more likely to use implementation shortfall. A VWAP algorithm usually hits the VWAP well. But the VWAP itself can be far from the decision price if the price rises all day.

Worked examples: VWAP, implementation shortfall, market impact

Perold (1988) proposed measuring implementation against a paper portfolio that would have traded at the price at the moment of the decision, without costs. The gap to the real portfolio is called implementation shortfall; it breaks down into delay, market impact, fees and opportunity cost for the unexecuted remainder.

How large the market impact is can be estimated with an empirical rule of thumb: it grows roughly with the square root of the order size. Tóth et al. (2011) document this square-root law with almost 500,000 futures trades from June 2007 to December 2010, for order sizes between a few hundredths of a percent and a few percent of daily volume; the constant Y is of the order of one (overview in Bouchaud et al. (2018)).

ETFs: secondary market, primary market and trading at NAV

On the secondary market, investors trade ETF units on the exchange or off-exchange with market makers. On the primary market, authorized participants create or redeem units with the provider, usually in exchange for a basket of the index securities (creation and redemption). This arbitrage keeps the market price close to the iNAV (Ben-David et al. 2017); for foreign equities and illiquid securities, noticeable premiums and discounts nevertheless remain (Petajisto 2017). Large investors send a request for quote to several market makers; J.P. Morgan Asset Management (2024) distinguishes three routes:

• Risk trade: immediately at a fixed price. The market maker takes on the risk and prices it in; the total cost is known immediately.
• Trading at NAV: the price is the future net asset value of the trading day or the following day, plus or minus an agreed number of basis points. The valuation point is independent and transparent, which helps, for example, when switching between two funds. In return, the investor bears the market risk until the calculation and only learns the final price on the day after publication.
• Agency or algorithm: a broker works the order in the ETF and in the underlyings; market risk and the risk of incomplete execution remain with the investor, but less information leaks out.

The fundamentals are covered in ETF basics, and the special features of active funds in Active ETFs.

Futures: trade at settlement, TMAC and BTIC

With trade at settlement (TAS), you buy or sell a future during the session at an agreed differential of a few ticks to the settlement price, which is only fixed at the close. CME offers TAS for agricultural, energy, metals, crypto and Treasury futures (CME Group, TAS 2026); for E-mini equity index futures, the counterpart is called TMAC (Trade Marker at Close). It is traded as a differential to a fixing derived from the volume-weighted futures price in a 30-second window at the close at 16:00 New York time (CME Group, TMAC 2023).

Even closer to index funds is BTIC (Basis Trade at Index Close): only the basis is agreed, i.e. the spread between the future and the index; the futures price is determined after the official index close as the closing level plus the basis (CME Group, BTIC 2026). This gives a fund exposure at exactly the closing price used to calculate its NAV. CME explicitly cites cash equitization (investing cash inflows immediately via futures) and index rebalancings as use cases. What futures and swaps cost as a substitute for equities is explained in Basket swaps, total return swaps or futures.

Transaction cost analysis: what an order really costs

Transaction cost analysis (TCA) compares the price achieved with a benchmark and breaks down the difference. Commissions, fees and taxes are visible; half the spread, market impact, delay and opportunity cost are invisible. Berkowitz et al. (1988) introduced the day's VWAP as a benchmark because it is less biased than individual prices such as the closing price. Across more than 14,000 orders on the NYSE, commission and market impact cost 23 basis points on average, of which 18 were commission and 5 execution.

Each benchmark suits a purpose: the decision price for active strategies, the VWAP for orders that are not urgent, the closing price for index funds. And each can be gamed: if you supply a large part of the volume yourself, you hit the VWAP automatically. Good TCA therefore measures against several benchmarks.

Advanced: Kyle's lambda and optimal execution

Why prices react to order flow is explained by Kyle (1985): a market maker sees only the sum of informed and uninformed orders and sets the price linearly in this flow. The factor λ (Kyle's lambda) is the market impact per unit, and its reciprocal is market depth. In the basic model with one trading round, λ = σ_v / (2 σ_u): lambda rises with uncertainty about the value and falls with the volume of uninformed orders.

Almgren and Chriss (2001) formulate execution as a trade-off: trading fast costs market impact, trading slowly brings price risk. With linear market impact, the efficient frontier of strategies can be determined in closed form. A risk-neutral trader (κ_R toward zero) unwinds evenly, like TWAP; the more risk-averse, the more they trade at the start, similar to an implementation shortfall algorithm. This also yields a liquidity-adjusted value at risk that includes the costs of liquidation. Two caveats: the square-root rule is not linear, so the closed-form result is only an approximation; and a rigid schedule can be detected, so that others can trade ahead of it (front-running in the broader sense).

What the research shows

Perold (1988) made clear that the most expensive costs are often invisible: missed opportunities and delayed implementation. Berkowitz et al. (1988) found that higher commissions did not bring correspondingly lower execution costs, and that some asset managers had persistently high or low costs, whereas brokers did not. Kyle (1985) and Glosten and Milgrom (1985) explain market impact and the spread through asymmetric information. For the square-root law, Tóth et al. (2011) provide an explanation based on the structure of liquidity. On the closing auction, Bogousslavsky and Muravyev (2021) show that its growing share is linked to index funds and ETFs; deviations in closing prices reverse shortly after the close and overnight, and liquidity at the open has deteriorated.

Limitations and common mistakes

• Market order in thin markets: outside the main trading hours, it can be executed far from the last price; a limit caps this.
• Stop when prices gap: a stop does not protect against an opening far below the stop price; a stop-limit order then goes unfilled.
• VWAP as success: hitting the VWAP does not mean you traded well. In a strong trend, faster would have been better.
• Forgetting opportunity cost: an algorithm with a tight price limit looks cheap if the remainder is never bought and the price runs away.
• MOC as free execution: large imbalances also move the closing price, and after the deadline orders often can no longer be canceled.

How Portfolio Lab calculates

Portfolio Lab does not simulate execution. The backtest uses closing prices of completed trading days; the app only takes the current day's price 30 minutes after the end of the regular session. Rebalancing takes place at the closing price of the last trading day of a period, as with a market-on-close order without market impact. You enter trading costs as with a broker: purchases are made half a bid-ask spread above and sales half a spread below the closing price, plus an order fee consisting of a fixed amount plus a percentage, subject to a minimum fee. The app does not model market impact that grows with order size, partial executions or intraday prices. For large portfolios or tight markets, you are better off using a generous spread (backtest methodology).

This article explains a method; it is not investment or tax advice. Worked examples are simplified and rounded. All terms in the glossary.