Will NT8 close the correct position?

What will happen if these trades happen in chronological order:

  1. Bought 10 ES at 7700 for long-term, so NOT intraday position.
  2. Bought 10 ES at 7730 for intraday position.
  3. Sold intraday position 10 ES at 7785.

What happens? ???

Will the intraday sell close the LONG term position as that is the oldest position?

What happens if the long-term position is different in size from the sold position?

Normally NT8 will take first the oldest order. But I don’t want that.

Marcus, assuming both entries are in the same account and the same ES contract month, the account has one net position: long 20. A filled sell of 10 reduces that to long 10. It does not preserve separate broker positions labelled “long-term” and “intraday”.

If the first entry were 5 and the second 10, selling 10 would leave long 5. The remaining quantity and the entry/exit pairing shown in a performance report are different things.

NinjaTrader’s performance guide explains that its account-level trade matching can differ from an individual strategy’s executions. Separate ATM instances can manage separate portions of the position, but they do not create separate broker positions in that same account/instrument.

Are these manual orders with ATM set to None, two separate ATM instances, or a NinjaScript strategy? That determines which stops/targets NT manages after the partial exit. Reproduce the exact sequence in Sim101 first and check the remaining position plus working orders.

This is actually determined at exchange Level. For ES, CME uses FIFO “First in First Out” and as far as I’m aware you cannot change that.

To me that looks like not honest/correct and does not correspond anymore with what my intentions are.

I know that these FIFO rule applies, I want to open the discussion about the existing rules that can destroy customers strategies. I also make myself no illusions that the rules will be changed. All orders, long-term or intraday look the same, so the “system” does not see any difference. Only I know which ones are long-term.

I have a mathematically created strategy behind each trade. That strategy depends on the exact entry price and timing. I calculate my expected risk and outcome based on that specific entry price. If then the intraday open trade is closing the long-term position, my strategy is completely ruined. Nothing from my strategy is then relevant anymore. So I am at that moment reduced to a gambler. I want to trade, not gamble.

Conclusion: I should open a second account to trade long-term.

I am a discretional trader (based on math) with only buy at market orders.

Marcus, I see the distinction now: you want each strategy measured against its own entry price.

Using your example at 7785, before costs:

  • Match the sell to 7700: realized 10 x 85 points, remaining open 10 x 55 points.
  • Assign it to the intraday entry at 7730: realized 10 x 55 points, remaining open 10 x 85 points.

Both add up to 1,400 contract-points. After that sell, both still have ten ES contracts exposed to the next price move. The pairing changes which strategy gets the realized result; it does not by itself change the combined result or remaining quantity.

For your discretionary calculations, you can keep two virtual trade records with their original entry prices and assign the exit to the intraday record. The records must still reconcile to the account’s net quantity. That preserves your strategy attribution without treating the performance report’s pairing as a new entry decision.

An account cannot use the same remaining contracts as two independently owned positions. If independent broker positions are the requirement, the broker needs to confirm the account arrangement. That is separate from whether your mathematical strategy works.

I don’t agree on that. Just one example (but there are more problems):

Each trade has different stops, stops on long-term are further away as the leverage of these trade are lower. If the long-term is closed by an intraday trade, the stops for the remain position are not correct anymore as well as the leverage. The settings are now intraday instead of long-term!

A long-term trade is based on another set of rules then an intraday trade. So the global result will not be the same. It will be worse due to the wrong set of settings that are used. I have to trade then a long-term trade on intraday settings as these settings are valid for the remaining open position.

That is completely wrong.

Yes, if the remaining quantity is then managed with the wrong stop or exit rule, the later result can be different. My 1,400-point comparison was only the snapshot immediately after the partial fill, before costs. It was not a claim that different future exit rules produce the same result.

The point to check is whether a stop actually changes, rather than which entry the performance report pairs with the sell. A report pairing alone does not establish which working order remains. After the partial exit, check the remaining quantity, each working stop’s price and quantity, and its ATM association if you use one.

For the intended long-term allocation, the required state would be the original long-term exit rule protecting the remaining ten contracts, with no leftover intraday exit able to close those ten unexpectedly. If the remaining orders instead apply the intraday rule, that is a real order-management problem, not just a reporting difference.

Are your stops separate manual orders, or are they attached to ATM instances? A Sim101 replay of the two entries and partial exit, with the working-order list before and after, would identify exactly where the intended assignment is lost. Please remove account identifiers from any screenshot.

This is a simplified example, for which I only use the stop and the leverage. I don’t use my actual settings, just random numbers.

  1. Long-term buy 10 contracts ES 700 at 7700. Stop at 100 points, leverage 10 .
  2. Intraday trade buy 10 contracts ES 700 at 7700. Stop at 10 points, leverage 2 5.
  3. I want to close the intraday trade, but the exchange takes my long-term position (FIFO) .

What is the result?

I have an open position with a stop at 10 points instead of 100 points. My leverage is 25 instead of 10. So if the market moves more than 10 points and hit the stop, my intentional open long-term trade will be closed with a 10 points loss instead of staying open with a 100 points stop. So my two trades are ruined by the FIFO rules.

Because of the changed leverage and stop, the risks are hugely bigger. That is the problem. Each set of intended trades has different settings and cannot be used for the other one without impacting the whole setup.

My stops are separate manual orders, they can be changed manually. Change will in general be reducing the stop to protect partial profits.

Theoretically I could change, immediately after the previous closed trade, the settings from the remaining intraday trade and make him long-term. But there are two problems:

My entry price can be substantially different. So risk/reward changes.

If the remaining position is closed before I can change the settings, hitting the actual stop (10 points), I lose 10 points and a trade that long-term was still profitable.

The manual-order detail helps. Your concern about the delay is valid: a stop can fill while you are trying to change or cancel it. I would not rely on a quick manual correction.

In your example, though, a FIFO report match does not itself replace the 100-point stop with the 10-point stop. They are separate working sell orders. A separate market sell of ten contracts and cancellation of the intended intraday exit are different operations. If the tight stop is left working, it can close the remaining ten; if both ten-contract stops remain working, the order quantities can even exceed the remaining position. That is the concrete failure to test.

For a Sim101 replay, label the two manual stop order IDs before the exit, then capture their prices, quantities and states after the ten-contract sell. The intended result is long ten, the original long-term stop still working for ten, and no active intraday exit. Also test a partial fill and a stop triggering during cancellation. A pending cancellation is not a confirmed cancellation.

The entry-price display and per-strategy P&L can still differ. For actual ES futures, however, ten contracts of the same expiry have the same $500-per-point exposure; FIFO does not change that contract value when it pairs a fill with an older entry.

NinjaTrader distinguishes closing an account/instrument position from closing a selected ATM: Operations > Order Entry > Trade Controls > Closing a Position or ATM Strategy Position
ES contract size: https://www.cmegroup.com/markets/equities/sp/sandp-futures.html

When you close the intraday allocation, do you send a separate ten-contract market sell, or use a Close control?

I am a discretionary daytrader who trades everything through DOM. So I have a very simple way of trading: I push BUY MARKET to buy x contracts and SELL MARKET to sell x contracts. I manually put my stops for each trade; executing at market price.

I never have more than 1 open position intradaday, and will have never more than 1 position long-term once I start LT trading too.

My solution is: 2 accounts. I want to keep things simple as it has less unexpected risks, and positions are always very clear.

Thanks, that answers the order-entry question: separate manual market orders and manually placed stops.

For the two-account setup you chose, the useful simulation check is the account boundary. After the intraday exit, verify that account is flat with no remaining intraday stop, while the other account still has its intended position and stop. Check the Account column in both Positions and Orders, rather than relying only on the DOM display.

A manual SELL MARKET fill and cancellation of a separately placed stop are still distinct events. Being flat does not by itself prove that the old stop is gone. NinjaTrader’s SuperDOM help treats its Close control differently: it closes the current position and cancels working orders for that instrument/account combination.

That keeps the check focused on your actual manual workflow, without adding an ATM or virtual sub-position system.