Held vs Not-Held Orders: Wetin Broker Discretion Mean

Wetin make held order difren from not-held order
Di difrens between held and not-held orders depend on one question: your broker get obligation to fill your order right now, or dem get space to choose the time?

Held order na instruction to execute immediately for the market price wey dey ground, and the broker must follow that immediacy. Not-held order give the broker discretion over time and price, and that discretion mean say dem no get any obligation to fill the order for one particular instant. Retail market orders na held orders by default. Block and algorithmic orders usually na not-held.

Held versus not-held orders: the main difference

"Held" na short form for "held to the market." The instruction dey clear: take wetin dey ground, right now. The broker no fit wait for better print, and e no fit leave part of the order make e wait for the quote to settle. Immediacy na the main thing. The price na wetin the market show when the order reach.

"Not held" change the deliverable. Time and price come turn to wetin the broker go judge. One "not-held" order fit rest for one hour, or e fit finish for thirty seconds when liquidity show face. Wetin the customer lose dey clear: no moment dey wey the broker suppose don fill the order. If the fill land after the price don move, that one no mean say the broker break the instruction.

Na that one be the difference for accountability, and na why "held" order pass just fast market order. Speed na the outcome. The obligation na the main thing wey dem dey buy.

How much immediacy dey cost?

If you drop buy order wey you wan make dem execute sharp-sharp, you go cross the bid-ask spread as you arrive. The spread na the gap between the best price wey buyer dey show for public and the best price wey seller dey show for public, and to cross am na the fee wey you pay to go first. That gap dey move throughout the day. The panel wey dey down so show the average of the quoted spread for two big-cap names in basis points of the midpoint, where one basis point na one hundredth of one percent, across half-hour blocks of one normal trading session.

SELECT
    formatDateTime(
        toStartOfInterval(toTimeZone(sip_timestamp, 'America/New_York'), INTERVAL 30 MINUTE),
        '%H:%i')                                       AS et_time,
    round(avgIf(toFloat64(ask_price - bid_price)
                / toFloat64((ask_price + bid_price) / 2) * 10000,
                ticker = 'AAPL'), 2)                   AS aapl_spread_bps,
    round(avgIf(toFloat64(ask_price - bid_price)
                / toFloat64((ask_price + bid_price) / 2) * 10000,
                ticker = 'KO'), 2)                     AS ko_spread_bps
FROM global_markets.cache_stocks_quotes
WHERE ticker IN ('AAPL', 'KO')
  AND sip_timestamp >= '2026-06-17 13:00:00'
  AND sip_timestamp <  '2026-06-17 20:30:00'
  AND bid_price > 0
  AND ask_price > bid_price
  AND toFloat64(ask_price - bid_price) / toFloat64(bid_price) < 0.02
GROUP BY et_time
HAVING countIf(ticker = 'AAPL') > 0
   AND countIf(ticker = 'KO') > 0
ORDER BY et_time

At 09:00 ET, before the market open, the AAPL quote average bin dey 6.57 bps wide and KO average bin dey 21.21 bps. By 12:00 the AAPL figure reach 0.89 bps, against 1.42 bps for KO. The last block for the panel, 16:00, print 6.72 bps for AAPL. Any held order go pay the spread of the minute wey e arrive, and nobody for the broker side get permission to wait make price better. If price ceiling matter pass time guarantee, that trade na matter of market order vs limit order.

Wetin be not-held order?

Not-held order na buy or sell instruction wey no get time limit for execution. For institutional ticket, dem dey mark am as NH, wey dey stay beside the size and the ticker. The customer dey tell the trading desk make dem work the order: dem go choose the venue, the pace, and use dem own judgment for price.

Two things dey happen. The advantage na say, if dem work the order, e fit wait for natural liquidity instead of paying the offer immediately, and e fit spread the size instead of pushing the price with everything at once. The disadvantage na say the customer no get guarantee say dem go fill the order at any point in time. If the market run, the order go chase am, and the desk no fail for wetin dem ask dem do. Discretion dey work for both sides as the trading session dey go.

Why big order no fit be one trade

Discretion get meaning only when the order too big to execute as one print. The tape dey consist of plenty small pieces and small-small big ones, and that distribution of size na why people dey "work" order for market.

SELECT
    print_size,
    round(100 * prints / sum(prints) OVER (), 2) AS prints_pct,
    round(100 * shares / sum(shares) OVER (), 2) AS shares_pct
FROM
(
    SELECT
        multiIf(size < 100,  'under 100',
                size < 500,  '100 to 499',
                size < 1000, '500 to 999',
                size < 5000, '1,000 to 4,999',
                             '5,000 and up') AS print_size,
        count()                              AS prints,
        sum(size)                            AS shares
    FROM global_markets.stocks_trades
    WHERE ticker = 'AAPL'
      AND sip_timestamp >= '2026-06-17 04:00:00'
      AND sip_timestamp <  '2026-06-18 04:00:00'
      AND size > 0
    GROUP BY print_size
)
ORDER BY multiIf(print_size = 'under 100', 1,
                 print_size = '100 to 499', 2,
                 print_size = '500 to 999', 3,
                 print_size = '1,000 to 4,999', 4,
                 5)

Prints of under 100 shares na 88.99% of every AAPL trade for that day and 22.84% of the shares wey change hand. The 5,000 and up bucket dey work the other way: 0.02% of the prints, and 48.77% of the shares. Small portion of big prints dey carry heavy part of the volume, while the total count of trades na the smallest bucket dey dominate, even for one of the most liquid stocks wey dey exist.

Put 400,000 share order against that distribution. No single counterparty dey stand there dey wait. The parent order go turn to hundreds or thousands of child orders, and the decision of when to release each one na exactly the discretion wey not-held instruction dey give. Some of those children dey hide their size, and na that one be the work wey iceberg orders dey do.

Which benchmark dem dey use take measure worked order?

Held order no need benchmark. Dem fill am as e reach, so the arrival price na the score. Worked order must get yardstick wey dem agree before time. The two wey dem dey use pass na arrival midpoint and the session VWAP. Trading desk wey dey measure against the close dey work towards the closing auction, na there MOC and MOO cutoff times come dey matter.

How far the arrival and VWAP fit stay from each other? The trace wey dey down so measure the distance between AAPL daily VWAP and that morning opening print over 9 sessions for June 2026. VWAP na the average price of the day wey dem weight base on how many shares trade for each one.

SELECT
    toString(date)                                                AS session_date,
    round(10000 * (toFloat64(max(vwap)) - toFloat64(max(open)))
          / toFloat64(max(open)), 1)                              AS vwap_minus_open_bps
FROM global_markets.stocks_daily_aggs
WHERE ticker = 'AAPL'
  AND date >= '2026-06-15'
  AND date <= '2026-06-26'
  AND open > 0
  AND vwap > 0
GROUP BY date
ORDER BY date

On 2026-06-15 the gap measure 64.3 bps. On 2026-06-26 e measure 266.3 bps. Each of those distances na space wey worked order get, for one side or the other, and nobody know the sign when the opening bell ring. The space no be the same for every name too.

SELECT
    ticker,
    round(avg(gap_bps), 1) AS avg_abs_gap_bps,
    round(max(gap_bps), 1) AS widest_gap_bps
FROM
(
    SELECT
        ticker,
        date,
        abs(10000 * (toFloat64(max(vwap)) - toFloat64(max(open)))
            / toFloat64(max(open))) AS gap_bps
    FROM global_markets.stocks_daily_aggs
    WHERE ticker IN ('AAPL', 'MSFT', 'NVDA', 'SPY', 'KO', 'JNJ')
      AND date >= '2026-06-01'
      AND date <= '2026-06-30'
      AND open > 0
      AND vwap > 0
    GROUP BY ticker, date
)
GROUP BY ticker
ORDER BY avg_abs_gap_bps DESC

Across June 2026, MSFT average 119.6 bps between the day VWAP and the open, with the widest single session of 396 bps. For the other side of the panel, SPY average 43.6 bps on the same measure. Two identical orders for those two names give desk very different amount of rope.

Where each instruction dey show up

• Retail market order dey held in practice: e dey go out for immediate execution, and the confirmation dey happen for the same minute.
• Marketable limit order dey sit close to held. E dey cap the price and e no dey give any timing discretion.
• Block wey dem hand over to broker trading desk typically no-held, with the NH marker for the ticket.
• Algorithmic order na not-held by construction. The schedule na the discretion, wey dem write as code.

Time in force dey answer different question: how long the order go stay alive, wey time in force cover for full. Who decide the moment of execution na the held question. For options, working an order dey closer to normal pass exceptional, and the reasons why resting order no dey fill dey laid out for why options orders don't get filled.

How discretion take connect to best execution

Broker wey dey handle customer order get duty to use reasonable diligence to find the best terms wey dey available. If customer give 'not-held' instruction, e no mean say that duty don stop. E only change how dem dey measure the duty; instead of the price wey dey screen for the exact time wey order reach, dem go look at the quality of judgment wey the broker use throughout the time wey dem work the order.

This difference dey show for public execution statistics. Standardized market center reports no dey include 'not-held' orders for inside their sample, because those reports assume say the goal na immediate execution, and 'worked' order no ever claim say na immediate execution e want. Rule 605 and 606 execution reports dey cover wetin those disclosures include. The quality of one 'worked' order dey inside the benchmark reporting wey the trading desk dey do, and na the customer get to ask for am.

Di question wey you suppose ask your broker

Two questions go settle am. Dem dey work my order, or dem dey send am out for immediate execution? And if dem dey work am, which benchmark dem dey use measure the fill, and I go fit see that measurement? One trading desk wey dey work orders get answer ready. One broker wey dey route everything for immediate execution get simpler answer, wey still dey important to know. The reader wey get two hundred share order and the reader wey get two hundred thousand share order want different products, and the words held and not-held na how the industry take dey differentiate dem.

FAQ

Wetin "not-held" mean for order?

Not-held order give broker power to decide the time and the price wey dem go take execute the trade. Dem ask the broker make e use im sense work the order over a period, instead of make e just fill am sharp-sharp. The customer come accept say nobody guarantee say the trade go fill for one particular price.

Retail market order na held order?

For reality, yes. Dem dey route am make e execute immediately against the market price wey dey ground, and the broker get the duty to do am sharp-sharp. Most retail platforms no dey print the word "held" or "not-held", because almost everything wey dem dey accept na "held" basis dem dey take handle am.

Not-held order dey give better price?

Sometimes, but nobody guarantee am. To work an order over time fit make you earn the spread instead of you to pay am, and e fit reduce the impact if you dump all the order for market one time. That same power wey the broker get mean say the fill fit land far from the price wey you see when you start, especially if market move while dem dey work the order.

Wetin be the difference between not-held order and limit order?

Limit order dey control price and nothing else: e go fill at the limit price or better, anytime the market reach that level. Not-held order no dey control price or timing from the customer side. Both of dem dey with the broker, inside whatever agreement wey una reach.

Every panel wey dey up so carry the SQL wey produce am. Open one, put any ticker wey you dey follow, and ask the same question for your own names for the Strasmore terminal.