SabrTrader Documentation

Automated Trade Journal & AI Coach

The Automated Trade Journal is SabrTrader's built-in trade-logging system: every time a position goes flat, the platform writes a complete journal entry for that trade — instrument, direction, entry and exit prices and times, size, P&L, duration, account — and attaches an automatic chart screenshot taken at the moment of exit. An optional post-trade pop-up lets you add the discretionary context a machine cannot know: the setup you were trading, your emotional state, any mistake you made, free-text notes and a star rating. The AI Coach panel then reads the resulting dataset and answers plain-English questions such as "Which setup made me the most money?" and "What mistakes am I making?", so journaling and review happen inside the platform instead of in a paid third-party journal.

Overview: What the Automated Trade Journal Is

The Automated Trade Journal is a SabrTrader feature that records every completed trade automatically, without any user action, and stores it in a searchable, per-account trade list alongside an AI Coach that analyses the recorded data in plain English.

Trade journaling is the practice of recording every trade's context, reasoning, execution and outcome so that patterns can be reviewed objectively later. Traditionally this is manual work: after each exit you are supposed to screenshot the chart, write down the setup, note your emotional state, record the entry and exit, and tag any mistake. A day trader who takes 20 round trips in a session has 20 of these to write up at the end of a tiring day, which is why manual journaling is the single most commonly abandoned trading routine.

SabrTrader removes the mechanical part of that work. The capture is automatic and unconditional; only the subjective tagging is optional.

What the journal captures and what you add

Captured automatically (no action required)	Added by you (optional, any time)
Instrument / symbol	Setup type (e.g. opening range breakout, VWAP reclaim, pullback breakout)
Direction (long or short)	Emotional state at entry or exit
Entry price and entry time	Mistake tag (oversized, early exit, moved stop, chasing, revenge trade)
Exit price and exit time	Star rating / setup qualification score
Position size	Market or instrument context note
Realised P&L	Free-text notes
Trade duration
Account the trade was executed on
Chart screenshot at the moment of exit

Where it sits in the platform

The journal is an expansion of the earlier Trade Performance drop-down item. In version 7.2 beta 3 it was promoted to its own tab on the Control Panel, with the journal trade list, per-trade detail view and the AI Coach side panel in one window. The statistical side of the same window — equity curve, win rate, drawdown and period breakdowns — is documented in Trade Performance Analytics and the AI Coach.

Why automation matters

• Completeness. Manual journals suffer from survivorship bias: traders write up memorable trades and skip the small ones. Automatic capture records the $12 scratch trade with the same fidelity as the big winner, so the dataset is complete.
• Honesty. Entry and exit prices, size and P&L come from the fill data, not from memory.
• Cost. The journal is part of SabrTrader, so it does not require a separate third-party journaling subscription or a broker-to-journal data connection.
• AI readiness. A structured, tagged dataset is what makes a language model's diagnosis useful rather than generic.

Where to Find the Journal and How to Verify Your Build

There are two access paths to the Automated Trade Journal.

1. Drop-down menu entry — the original route, listed as the journal / trade performance item in the Control Panel menu.
2. Journal tab on the Control Panel — added in 7.2 beta 3, this is a permanent tab in the main window so the journal does not have to be re-opened from a menu each session.

Both routes open the same window and the same data. If you are following documentation or a video and the journal is not where it is described, your installed build is probably older than the one being demonstrated.

Checking your installed version

Open Help > About SabrTrader. The dialog reports the exact build, for example 7.2 beta 3. Because SabrTrader ships frequent updates, the feature set in any guide is tied to a build number; checking About first eliminates the most common cause of "I can't find this feature".

Beta channel versus last stable build

Download channel	What it contains	Who should use it
Current beta	The newest features, including journal and AI Coach enhancements, before they reach the stable channel	Traders who want the latest capability and accept occasional rough edges
Last stable	The most recent build validated as stable	Traders running live capital who prioritise predictability over new features

Both channels are available from the SabrTrader download page. You can move back to the last stable build at any time if a beta causes a problem.

How Automatic Trade Capture Works

Automatic trade capture is the mechanism that writes a journal entry when a position closes. The trigger is explicit and mechanical: the position goes flat.

The capture trigger

SabrTrader treats the sequence from the first fill that opens a position to the fill that returns the position to zero quantity as one trade. The moment net position in that instrument on that account reaches zero, the journal entry is written. This means:

• Scaling into a position does not create several entries; it creates one trade with an averaged entry.
• Scaling out does not create several entries; the trade closes when the last contract or share is closed.
• Reversing (for example selling 2 while long 1) closes the first trade at flat and opens a new one in the opposite direction.

What is stored with no user action

• Instrument — the symbol traded, including the futures contract month where applicable.
• Direction — long or short.
• Entry price and time, and exit price and time.
• Size — quantity traded.
• Realised P&L for the round trip.
• Duration — how long the position was held.
• Account — which trading account the fills belong to.
• Chart screenshot — an image of the chart captured at the moment of exit, preserving price action, indicators and drawings exactly as they were when you closed.

Account coverage

Capture is account-agnostic. The journal records trades identically across:

• Simulated (sim) accounts — the practice environment used for learning the platform and rehearsing setups.
• Prop / funded-evaluation accounts — for example an Apex account routed through a futures feed such as Rithmic.
• Live broker accounts — for example an Interactive Brokers account.

Each entry stores the originating account, so results are never blended unless you deliberately choose to view all accounts together. This matters because sim P&L, evaluation P&L and real cash P&L have different psychological and financial weight and should be analysed separately.

Interaction with automated exits

Trades closed by an ATM strategy bracket — a stop or target that fires without you clicking anything — are journaled exactly like manual exits. The journal records the fill; it does not care who or what generated it. Positions mirrored into other accounts by the Trade Copier are journaled per account, so the copy accounts each receive their own entries.

The Post-Trade Pop-Up

The post-trade pop-up is a small form that appears at the bottom of the chart the moment a position goes flat. Its purpose is to collect the subjective context of the trade while the trade is still fresh in your mind — the information the platform cannot derive from fill data.

Behaviour

• It appears automatically on exit, docked below the chart, so it does not cover price action.
• It is entirely optional. Pressing Accept saves the trade with the automatically captured data and no tags.
• Ignoring it does not lose the trade. The journal entry already exists; you can open it later from the trade list and add every field then.
• It never blocks trading. You can place the next order without touching the form.

Fields presented

• Setup type — which pre-defined pattern you were trading.
• Emotional state — how you felt at entry or exit.
• Mistake type — a behavioural error tag, or none.
• Market / instrument context — what the market was doing (for example trending, balancing, news-driven).
• Notes — free text for anything the dropdowns do not cover.
• Star rating — setup qualification score.

Two legitimate workflows exist: tag at exit while memory is accurate, or trade the session and batch-tag afterwards using the screenshot to reconstruct each trade. Scalpers and high-frequency day traders usually prefer the second; swing and low-frequency intraday traders usually prefer the first.

Every Journal Field and Option Explained

Each optional field exists because it becomes a filter and an input to the AI Coach later. A field left blank simply means that dimension cannot be analysed.

Setup type

A setup is a repeatable, pre-defined pattern or condition that triggers an entry. Tagging it lets you measure which setups actually pay rather than which ones feel good. The dropdown includes common intraday models and accepts a custom entry:

Setup	Definition
Opening range breakout (ORB)	Entry taken when price breaks the high or low of the session's first defined time window (commonly the first 5, 15 or 30 minutes).
VWAP reclaim	Entry taken when price trades back through VWAP (volume-weighted average price) and holds on the other side, signalling a shift back to the mean-value side of the auction.
Pullback breakout	Entry taken on the retest of a broken level after the initial breakout move, rather than on the break itself.
Custom	Any setup name you define, such as a volume profile value-area fade or a market structure Break of Structure continuation.

Keep the setup vocabulary small and fixed. Five well-defined setup names produce five measurable sample groups; thirty improvised names produce thirty samples of one trade each and no statistics at all.

Emotional state

Emotional state tagging is recording how you felt at entry or exit — calm, confident, fearful, impatient, frustrated, bored — so psychology can be correlated with results. It answers questions the P&L alone cannot: for example, whether trades tagged "impatient" have a materially lower win rate than trades tagged "calm" on the same setup.

Mistake taxonomy

A mistake tag records a process error, independent of outcome. A trade can be profitable and still carry a mistake tag; that distinction is the point of the field.

Mistake tag	Definition	Typical cause
Oversized	Position size larger than the account or setup risk plan allows.	Overconfidence after wins, or attempting to recover a loss faster.
Early exit	Closing before the plan's target or invalidation, driven by discomfort rather than new information.	Intolerance of open-profit give-back or of adverse excursion.
Moved stop	Widening a protective stop away from price to avoid being stopped out.	Unwillingness to accept a planned loss; converts a defined risk into an undefined one.
Chasing	Entering late, after the move has already extended well beyond the planned entry.	Fear of missing out; destroys the reward-to-risk ratio the setup was based on.
Revenge trade	Re-entering impulsively after a loss to recover money rather than because a valid setup appeared.	Emotional response to a loss; usually the highest-cost tag in any journal.

Star rating / setup qualification

The star rating scores how well the trade matched your written criteria for that setup — not how much money it made. A five-star trade is one where every condition was present and the execution followed the plan, even if it lost. Rating quality separately from outcome is what lets you distinguish bad luck from bad process. Over a sample, low-star trades usually show a materially worse expectancy than high-star trades of the same setup; if they do not, your qualification criteria are not doing any work.

Market / instrument context and notes

Context captures the regime: trending, balancing, pre-news, post-news, thin overnight session, high volatility. Notes hold anything structured fields miss — the level you were trading off, what invalidated the idea, what you would do differently.

How the fields feed later analysis

• Filtering: isolate all VWAP reclaim shorts, or all trades tagged "moved stop".
• Comparison: win rate and average P&L per setup, per direction, per emotional state.
• AI input: the AI Coach reads tags alongside the numeric data. Untagged trades reduce the coach to commenting on raw P&L.

Reviewing the Trade List and Trade Detail

The journal's main view is a trade list: one row per completed round trip, grouped by day and filtered by account.

Account selector

The account selector chooses which account's trades are listed — for example an IBKR live account, an Apex prop evaluation account, or the simulation account. Account segmentation (filtering statistics per account so simulated, live and funded results are never blended) is essential: a 70% win rate that mixes sim and live trades tells you nothing about either.

Day-by-day grouping

Trades are grouped by trading day so that a session can be reviewed as a unit. This is how you see sequence effects: a loss followed by three rapid re-entries is visible as a pattern in the day grouping and invisible in an aggregate statistic.

Opening a trade's detail

Clicking a row opens the full trade record: all captured fields, every tag you added, the notes, and the automatic chart screenshot taken at exit. The screenshot is the single most valuable element for review, because it shows what you were actually looking at — the same indicators, drawings and timeframe — rather than a reconstruction made days later on a chart that has since moved on.

What to read in the list

• Winners versus losers — the raw split, and more importantly whether average win exceeds average loss.
• Stop-outs — trades closed by the protective stop rather than by a discretionary or target exit. A healthy journal contains stop-outs; their absence usually means stops are being moved.
• Long versus short split — comparing results by direction exposes directional bias or a genuine skill gap. Many traders are consistently profitable in one direction and consistently negative in the other, which is actionable immediately.
• Scratch and micro trades — small in-and-out trades, such as a $12 round trip taken because the expected reaction off a level never arrived. Individually trivial, collectively they can account for most of a session's commissions and most of a trader's screen fatigue. Because capture is automatic, these trades appear in the data whether or not you would have bothered to log them.
• Duration — how long winners are held versus losers. Losers held longer than winners is the classic signature of cutting winners early and hoping on losers.

The AI Coach Panel

The AI Coach is a panel inside the journal window that sends your filtered trade data and tags to a configured large language model and returns a written review: strengths, weaknesses, anomalies and concrete action items.

Opening the panel

The AI Coach is not visible when the journal window first opens in its compact form. Pop the panel out to the side; it then docks alongside the trade list so you can read the analysis next to the trades it refers to. The same pop-out behaviour applies to the individual trade detail view, so you can widen the window to show the list, the detail and the coach simultaneously.

Preset prompts

Preset question	What it analyses	When to use it
Which setup made me the most money?	P&L and win rate grouped by setup tag	After at least 20-30 trades per setup, to decide where to concentrate
Explain my results in plain English	Overall performance summary without jargon	End of day or end of week, as the default starting question
What was my edge?	Which conditions, setups or directions show a repeatable statistical advantage	Monthly, over a meaningful sample
What mistakes am I making?	Mistake tags, their frequency and their aggregate cost	Weekly, as the behavioural check
How can I improve my results?	Prioritised, actionable changes derived from the above	End of week or end of month, before setting next period's rules

How the output is produced

The coach reads the trade dataset currently in scope — the selected account and date range — including the numeric fields and whatever setup, emotion, mistake and rating tags you entered. It then produces a written diagnosis. The quality of the output is a direct function of the quality of the tags: with tags, the coach can say "your VWAP reclaim longs are profitable and your opening range breakout shorts are not"; without them, it can only discuss P&L, win rate, duration and direction.

The model itself is the one you configured under Preferences — see AI Setup: Connecting an LLM Provider to SabrTrader. No provider key, no AI analysis.

Worked example

A typical coach output reads: "Bottom line: you're picking good trades and you let winners pay. Your weak spot is how much pain you sit through inside your trade."

That statement is a measurable claim, not a platitude. "Pain inside the trade" is maximum adverse excursion (MAE) — the largest unrealised loss a trade held before closing. If winning trades routinely show large MAE, two interpretations are possible, and both are testable:

• Entries are early or imprecise. The idea is right but the location is wrong, so the trade goes deeply against you before working. Fix by refining entry triggers or waiting for confirmation.
• Stops are too wide, or trades are held against the original invalidation. Fix by tightening the stop to the structural invalidation point and accepting more stop-outs in exchange for smaller losses.

The counterpart metric is maximum favourable excursion (MFE) — the largest unrealised profit a trade reached. Comparing MFE with actual exit price measures exit quality: if MFE is consistently far above the realised exit, you are exiting early; if realised exits are close to MFE, your exits are efficient.

Copying and exporting the analysis

The panel is narrow, so long analyses are condensed. A copy control places the full text on the clipboard for pasting into a text editor, a word processor or a weekly review document. Exporting matters for one reason: a single AI read is an opinion, but a file of twelve weekly reads shows whether the same diagnosis keeps recurring. A recurring diagnosis is a real leak; a one-off diagnosis may be noise.

Trading Concepts Behind the Journal

The journal is only useful if the concepts it records are understood precisely. Each term below is defined independently of the software.

Setups

• Setup (trade setup) — a repeatable, pre-defined pattern or condition that triggers an entry. A setup must be specific enough that two traders reading the definition would mark the same bars.
• Opening range breakout (ORB) — an entry taken when price breaks the high or low of the session's first defined time window. The range period must be fixed in advance (for example the first 15 minutes) or the setup is not measurable.
• VWAP — volume-weighted average price: the session's average traded price weighted by volume, widely used as an intraday fair-value reference. Institutional execution is often benchmarked against it.
• VWAP reclaim — a setup in which price trades back through VWAP and holds, signalling a shift back to the mean-value side of the auction.
• Pullback breakout — an entry on the retest of a broken level after the initial breakout move. It trades worse fills for better confirmation and a tighter stop.

Exits and risk

• Stop-out — a trade closed by the protective stop order rather than by a discretionary or target exit.
• Moving the stop — widening a protective stop away from price to avoid being stopped out. It converts a known, sized loss into an unknown one and is the mistake tag most strongly associated with account-ending losses.
• Early exit — closing a trade before the plan's target or invalidation, usually driven by discomfort rather than information.
• Letting winners run — holding a profitable trade toward its full target rather than taking small profits out of fear. Measurable as the gap between realised exit and MFE.
• Drawdown — a decline from a prior equity or trade peak, measured either within a single trade (open-profit give-back) or across the account.

Excursion metrics

• MAE (maximum adverse excursion) — the largest unrealised loss a trade held before closing. It quantifies how much heat the trade took. Analysing MAE across winners tells you how tight a stop could be without cutting good trades.
• MFE (maximum favourable excursion) — the largest unrealised profit reached during a trade. Analysing MFE across all trades tells you how much of the available move your exits capture.

Behavioural categories

• Chasing — entering late after the move has already extended well beyond the planned entry.
• Oversizing — taking position size larger than the account or setup risk plan allows. Hard limits for this are enforced separately by Risk Master.
• Revenge trading — re-entering impulsively after a loss to recover money rather than because a valid setup appeared.
• Emotional state tagging — recording how you felt at entry or exit to correlate psychology with results.

Edge and sample size

Trading edge is the statistical advantage a repeated process produces over a large sample of trades. Three important consequences follow:

• A single trade, or a day, contains almost no information about edge. Outcome and process are only weakly correlated over small samples.
• Any per-setup conclusion needs a meaningful sample — as a working rule, treat fewer than 20-30 trades in a group as directional evidence at best, and 50+ as reasonably informative.
• Edge must survive costs. Commissions and fees are real per-side costs; a setup that is marginally positive before fees is negative after them.

Why direction and setup splits matter

The long versus short performance split compares results by trade direction. It frequently reveals that a trader has one profitable direction and one unprofitable one — a product of market regime, of reading bullish structures better than bearish ones, or of sizing differently without realising it. The same logic applies to splitting by setup: an overall break-even account often contains one clearly profitable setup subsidising two losing ones. The journal's job is to make that visible.

Day trading frequency and account types

Day trading frequency — high trade counts per session — is precisely what makes manual journaling impractical and automated capture valuable. Sim, prop and live accounts (simulated, funded-evaluation and personal cash accounts) are journaled identically, which lets you compare behaviour across them: many traders discover that their sim process and their live process differ measurably in holding time and in stop discipline.

Trading discipline and routine

Trading discipline / routine in the journaling context means a consistent end-of-day and end-of-week review process that converts raw trade data into behavioural change. Data without a review cadence changes nothing.

Step-by-Step Tutorial: From First Trade to AI Review

This walkthrough uses a simulated account so no capital is at risk. The same steps apply unchanged to a prop or live account.

1. Confirm your version. Open Help > About SabrTrader and read the build number. The Journal tab and AI Coach described here require 7.2 beta 3 or later. If your build is older, download the current beta or the last stable build from the SabrTrader download page and reinstall.
2. Open the Journal. Click the Journal / Trade Performance tab on the Control Panel, or select the journal entry from the drop-down menu. Both open the same window.
3. Select the account. Use the account selector to choose the account you are about to trade — for example the sim account. This filters the trade list so you are not reading mixed results. If you trade several accounts, note that copied accounts each produce their own entries.
4. Take a trade. Place an order from the chart, the SuperDOM or Chart Trader, then close the position. You can also do this against recorded data using Market Replay, which is the fastest way to build a tagging habit outside market hours.
5. Handle the exit pop-up. The moment the position goes flat, the post-trade form appears below the chart. Either fill in setup, emotion, mistake, context, notes and rating, or press Accept to save the automatically captured data only. Either way the trade is recorded.
6. Open the trade detail. Return to the journal, find the trade in the day group and click the row. Inspect the captured fields and the chart screenshot taken at exit. Confirm the screenshot shows the chart state you expect.
7. Tag and rate. If you skipped the pop-up, complete the fields now: pick the setup from your fixed vocabulary, record the emotional state, apply a mistake tag if a process error occurred, and give a star rating for how well the trade matched your setup criteria. Save.
8. Pop out the AI Coach. Expand the AI Coach panel to the side of the journal window so the analysis sits next to the trade list.
9. Run a preset prompt. Start with Explain my results in plain English. Read the output, then run What mistakes am I making? and compare it with the mistake tags you entered. Where the coach's reading and your tags disagree, the screenshots usually settle it.
10. Copy and archive the output. Press the copy control and paste the analysis into your weekly review document, dated and labelled with the account and date range. Over several weeks, scan the archive for diagnoses that repeat — those are the leaks worth working on.

Suggested review cadence

Cadence	Action	Goal
Per trade	Tag in the pop-up, or leave for the end of session	Accurate context while memory is fresh
End of day	Open the day group, review screenshots, complete missing tags	Complete dataset, immediate behavioural feedback
End of week	Run "What mistakes am I making?" and "How can I improve my results?", copy to the review doc	Identify recurring process errors
End of month	Run "Which setup made me the most money?" and "What was my edge?" across the full month	Allocate attention and size to setups with demonstrated edge

Best Practices

• Tag at exit while the memory is accurate, or batch-tag the same day. Context decays fast. Tagging a Tuesday trade on Friday produces a reconstruction, not a record. If you cannot tag at exit, use the screenshot and tag before the session's end.
• Keep a small, fixed setup vocabulary. Write down your setups — five is plenty — with explicit entry criteria, and use only those names. A fixed vocabulary is what turns tags into statistics.
• Score setup quality separately from outcome. Use the star rating for "did this match my criteria", never for "did this make money". Mixing the two destroys the field's diagnostic value.
• Tag the mistake even on winning trades. A revenge trade that happened to profit is still a revenge trade, and it is the one most likely to be repeated because it was rewarded.
• Review daily, run AI analysis weekly and monthly. Daily review keeps data complete; the AI read needs a sample to be worth anything.
• Use sim trades to build the habit. The tagging workflow should be automatic before real money is involved. Sim trades and Market Replay sessions are journaled identically.
• Export AI output and track recurrence. Keep a dated file of coach analyses. The question that matters is not "what did the AI say this week" but "has it said the same thing four weeks running".
• Judge setups only on meaningful samples. Twenty to thirty trades per setup minimum before concluding anything; fifty before changing size.
• Review each account separately. Sim, prop evaluation and live accounts have different rules, different sizes and different psychology. Blending them produces statistics that describe no real trading.
• Pair the journal with hard risk limits. Journals record behaviour after the fact; Risk Master daily loss limits, trade limits and locks prevent the worst instances from happening at all. Use the journal to decide where the limits should sit.
• Connect your excursion findings to your exits. If MAE analysis shows winners rarely trade more than N ticks against you, that is evidence for a tighter stop — and a tighter stop can be automated with an ATM strategy rather than left to discretion.

Common Mistakes

• Skipping tagging entirely. The journal still captures every trade, but the AI Coach then only sees raw P&L, direction and duration. It cannot tell you which setup pays or which emotion costs you, because you never told it. Automatic capture solves the data problem, not the context problem.
• Inventing a new setup name for every trade. Thirty unique setup labels across forty trades means every group has a sample of one. Setup analysis becomes impossible. Use a short fixed list.
• Treating a three-trade AI read as statistically meaningful. A language model will produce confident prose from any dataset. Over three trades, the prose is describing noise. Ask for analysis only when the sample supports it, and discount conclusions drawn from small groups.
• Ignoring small scratch trades. The $12 in-and-out trades look irrelevant individually. In aggregate they reveal how often you enter without the condition you require, and they carry full commission costs. They are often the clearest evidence of impatience in the whole dataset.
• Using the journal to confirm a bias. Reading only the winners, or filtering until the numbers look acceptable, converts a diagnostic tool into a reassurance tool. The correct use is to search specifically for the worst-performing group and ask why it exists.
• Not separating accounts. Blending sim practice with funded-evaluation results inflates win rate and hides the behavioural difference between them.
• Rating trades by outcome. Giving five stars to every winner and one star to every loser makes the rating field a duplicate of the P&L column and removes the only field that measures process independently.
• Never acting on the output. The most common failure is a well-maintained journal with no rule changes attached to it. Each weekly review should end with one specific, testable change — a stop placement, a size cap, a banned setup or a time-of-day restriction.
• Assuming an older build will behave the same. The Journal tab, the exit pop-up and the AI Coach presets were added in specific builds. Verify through Help > About SabrTrader before concluding a feature is missing.

Frequently Asked Questions

Do I have to do anything for a trade to be journaled in SabrTrader?

No. Capture is automatic and unconditional. When a position goes flat, SabrTrader writes a journal entry containing the instrument, direction, entry and exit price and time, size, realised P&L, duration, account and a chart screenshot taken at the moment of exit. The only optional part is the subjective tagging — setup, emotional state, mistake, context, notes and star rating — which you can add at exit or any time afterwards.

When exactly does the post-trade pop-up appear?

It appears the instant your position in that instrument reaches zero quantity on that account — in other words, when you go flat. It docks at the bottom of the chart so it does not cover price action. Scaling into or out of a position does not trigger it; only the final fill that closes the position does.

Can I skip the pop-up and tag my trades later?

Yes. Press Accept to save the trade with the automatically captured data and no tags, or ignore the pop-up entirely. The journal entry already exists either way. Open the Journal tab later, click the trade row, and complete the setup, emotion, mistake, notes and rating fields then. Many high-frequency day traders deliberately trade the whole session untagged and batch-tag at the close using the attached screenshots.

Does the journal work with simulated and prop-firm accounts?

Yes. Simulated accounts, prop / funded-evaluation accounts such as Apex, and live broker accounts such as Interactive Brokers are journaled identically. Each entry records which account it came from, and the journal's account selector lets you review each account separately so sim, evaluation and live results are never blended.

Does SabrTrader capture a chart screenshot of every trade automatically?

Yes. A screenshot of the chart is taken at the moment of exit and attached to the journal entry. It preserves the chart exactly as you were seeing it — same timeframe, indicators and drawings — which is what makes later review honest rather than a reconstruction on a chart that has since moved on.

Which setup types can I tag a trade with?

The setup dropdown includes common intraday models such as opening range breakout (entry on a break of the first time window's high or low), VWAP reclaim (price trading back through volume-weighted average price and holding) and pullback breakout (entry on the retest of a broken level), plus a custom option for your own setup names. Keep the list short and fixed so each setup accumulates a measurable sample.

What mistake categories does the journal track?

The built-in mistake taxonomy covers oversized (size beyond the risk plan), early exit (closing before target or invalidation out of discomfort), moved stop (widening the protective stop to avoid being stopped out), chasing (entering late after the move has extended) and revenge trade (impulsive re-entry after a loss to recover money). A mistake tag records a process error regardless of whether the trade made money.

What questions can the AI Coach answer about my trading?

The panel ships with five preset prompts: Which setup made me the most money?, Explain my results in plain English, What was my edge?, What mistakes am I making? and How can I improve my results?. The coach reads the trade data currently in scope — including your setup, emotion, mistake and rating tags — and returns a written diagnosis with strengths, weaknesses, anomalies and suggested changes. Answer quality depends directly on how consistently you tag.

Can I export or copy the AI Coach analysis?

Yes. A copy control places the full analysis text on the clipboard so you can paste it into a text editor, word processor or your weekly review document. This is the recommended workflow, because the real signal is whether the same diagnosis recurs across several weeks of archived analyses.

Where do I find the journal if it is no longer in the drop-down menu?

From version 7.2 beta 3 the journal has its own Journal / Trade Performance tab on the Control Panel, in addition to the drop-down entry. Both open the same window and the same data. If neither route shows it, check your build under Help > About SabrTrader and update to the current beta or last stable release.

Do I need a third-party journaling subscription if I use SabrTrader?

No. The journal is built into the platform, which removes the separate subscription, the broker-to-journal data connection and the manual import step that third-party journals require. Trades are captured from your own fills inside the platform the moment you go flat.

How many trades do I need before the AI analysis is meaningful?

Treat any conclusion based on fewer than 20-30 trades in a group as weak evidence, and 50 or more as reasonably informative. A language model will produce confident prose from three trades, but it is describing noise. Run the AI weekly for behavioural observations and monthly for setup and edge conclusions.

Can I journal multiple broker accounts separately?

Yes. Every entry stores the originating account, and the journal has an account selector so you can view one account at a time. This is required for correct analysis, because mixing simulated, funded-evaluation and live cash results produces statistics that describe no real trading environment.

What does the AI mean when it says I sit through too much pain inside a trade?

It is describing maximum adverse excursion (MAE) — the largest unrealised loss a trade held before it closed. High MAE on trades that ultimately won means either your entries are early or imprecise, so the trade goes deeply against you before working, or your stops are wider than the structural invalidation requires. Both are testable: refine the entry trigger, or tighten the stop and accept more stop-outs in exchange for smaller individual losses.

What is the difference between MAE and MFE?

MAE (maximum adverse excursion) is the largest unrealised loss a trade held before closing; it measures how much heat you took and informs stop placement. MFE (maximum favourable excursion) is the largest unrealised profit a trade reached; comparing MFE with your actual exit price measures exit quality. Large MAE points to entry or stop problems; a large gap between MFE and the realised exit points to exiting too early.

How are partial fills and scale-outs handled?

SabrTrader treats the sequence from the first fill that opens a position to the fill that returns net position to zero as one trade. Scaling in produces a single entry with an averaged entry price; scaling out produces a single entry that closes when the last unit is closed. A reversal closes the existing trade at flat and opens a new one in the opposite direction.

Are trades closed by a stop or target bracket journaled too?

Yes. The journal records the fill regardless of what produced it. A trade exited automatically by an ATM strategy bracket is captured exactly like a manual exit, and it appears in the list as a stop-out or target exit depending on which side filled.

Can I edit a journal entry after saving it?

The subjective fields — setup, emotional state, mistake tag, context, notes and star rating — can be opened and changed at any time from the trade detail view. The automatically captured execution data (prices, times, size, P&L, account) reflects your actual fills and is the factual record of the trade.

Does the AI Coach need an API key to work?

Yes. The coach sends your trade statistics to the large language model you configured in Preferences, which requires a provider API key. Configure this first following the AI setup documentation; without a configured provider the preset prompts will not return analysis.

Should I journal sim trades, or only real money?

Journal both, in separate account views. Sim trades are the cheapest way to build the tagging habit and to test a new setup's criteria, and comparing sim behaviour with live behaviour often exposes the exact point where discipline breaks down — typically shorter holding times and more stop movement on live capital.

Troubleshooting

The post-trade pop-up does not appear when I exit a trade

Check the following in order:

1. Confirm the position is actually flat. The pop-up triggers only when net quantity reaches zero, not on partial scale-outs.
2. Confirm your build supports it — Help > About SabrTrader should show 7.2 beta 3 or later.
3. The pop-up docks at the bottom of the chart; if the chart window is short or partially off-screen, resize or reposition it.
4. If the exit happened on an account whose chart is not open, review the trade from the Journal tab instead — the entry is still recorded.

I cannot find the AI Coach anywhere in the journal window

The AI Coach is hidden in the journal's compact layout. Pop the panel out to the side; it then docks next to the trade list. The individual trade detail view pops out the same way. If the panel opens but returns no analysis, confirm an AI provider and API key are configured in Preferences — see the AI setup documentation.

Trades are missing from the journal after a reconnect or platform restart

Entries are written when the position goes flat and the fill is known to the platform. If the connection dropped mid-trade, the closing fill may have arrived while the platform was disconnected. Reconnect, confirm the broker shows the position as closed, and check the correct account is selected in the journal. Also check the date grouping — the trade may be filed under a different session date than expected if the close occurred after the session boundary.

The journal shows the wrong account or no trades at all

The account selector filters the list. Confirm it is set to the account you traded, not a sim or secondary account. If you use the Trade Copier, each copy account receives its own entries, so the trades you are looking for may be filed under a copy account rather than the master.

The attached chart screenshot is blank or shows the wrong chart

The screenshot is captured from the chart at the moment of exit. If no chart for that instrument was open, if the chart was minimised, or if the chart was mid-reload when the exit occurred, the capture can be empty or incomplete. Keep a chart open for every instrument you trade. If screenshots are consistently blank, the chart rendering mode may be incompatible with capture — change it in Preferences and retest with a sim trade.

The Journal tab does not exist on my Control Panel

Your build predates the change that promoted Trade Performance into its own tab. Open Help > About SabrTrader, note the version, then download either the current beta or the last stable build from the SabrTrader download page and reinstall. The drop-down menu entry remains available in builds that have both.

The AI Coach output is vague and only talks about P&L

This is a data problem, not a model problem. If no setup, emotion or mistake tags exist, the coach has only numeric fields to work with. Tag a representative block of trades — at minimum setup and mistake — then re-run the prompt. Also check that the date range and account filter include enough trades; a handful of entries cannot support a meaningful read.

AI analysis contradicts my own reading of the trades

Open the trades the coach references and inspect the attached screenshots. Disagreements usually trace to one of three causes: inconsistent setup naming (the same pattern logged under two labels), tags applied by outcome rather than by process, or too small a sample. Fix the tagging first, then re-run the analysis.

Sim and live results look mixed together

Use the account selector to view one account at a time. Any statistic or AI analysis run with multiple accounts in scope aggregates them. Review sim, prop evaluation and live accounts in separate passes, and label the exported AI analyses with the account name.

The AI output is too long to read in the narrow panel

Press the copy control to place the full text on the clipboard, then paste it into a text editor or word processor. This is also the recommended archival workflow — dated, labelled analyses let you detect whether the same diagnosis keeps recurring week after week.

Glossary

Account segmentation

Filtering statistics per trading account so simulated, live and funded results are never blended into one report.

AI trade coach

A language model that reads your logged trade data and tags and returns plain-English diagnostics, anomalies and improvement suggestions.

Automatic trade capture

SabrTrader's mechanism for writing a complete journal entry — including a chart screenshot — the moment a position goes flat, with no user action required.

Chasing

Entering late after the move has already extended well beyond the planned entry, which degrades the reward-to-risk ratio the setup was based on.

Commissions and fees

Per-side transaction costs charged by the broker and exchange. They reduce net performance and must be included before a profit factor or expectancy figure is trusted.

Day trading frequency

High trade counts per session, which make manual journaling impractical and automated capture valuable.

Drawdown

A decline from a prior equity or trade peak, measured either within a single trade as open-profit give-back or across the account over time.

Early exit

Closing a trade before the plan's target or invalidation, usually driven by discomfort rather than new information.

Emotional state tagging

Recording how you felt at entry or exit — calm, fearful, impatient, frustrated — so that psychology can be correlated with results.

Expectancy

The average expected profit or loss per trade, computed as (win rate × average win) − (loss rate × average loss). A high win rate with small wins and large losses can still produce negative expectancy.

Letting winners run

Holding a profitable trade toward its full target rather than taking small profits out of fear.

Long vs short performance split

Comparing results by trade direction to expose a directional bias or a genuine skill gap between buying and selling.

MAE (maximum adverse excursion)

The largest unrealised loss a trade held before closing — the measurable form of the 'pain you sit through' inside a trade.

MFE (maximum favourable excursion)

The largest unrealised profit reached during a trade, used to judge exit quality by comparing it with the realised exit price.

Moving the stop

Widening a protective stop away from price to avoid being stopped out; one of the tracked journal mistake tags. It converts a known, sized loss into an unknown one.

Opening range breakout

An entry taken when price breaks the high or low of the session's first defined time window, such as the first 15 minutes.

Oversizing

Taking position size larger than the account or setup risk plan allows.

Post-trade pop-up

The optional form that appears below the chart on exit, collecting setup, emotional state, mistake tag, market context, notes and star rating for the trade just closed.

Pullback breakout

An entry taken on the retest of a broken level after the initial breakout move, trading a worse fill for better confirmation and a tighter stop.

Revenge trading

Re-entering impulsively after a loss to recover money rather than because a valid setup appeared.

Setup (trade setup)

A repeatable, pre-defined pattern or condition that triggers an entry, tagged on each journal entry so you can measure which setups actually pay.

Sim, prop and live accounts

Simulated practice accounts, funded-evaluation (prop firm) accounts and personal cash accounts — all journaled identically by SabrTrader but reviewed separately.

Star rating / setup qualification

A score recording how well a trade matched your written criteria for its setup, judged independently of whether the trade made money.

Stop-out

A trade closed by the protective stop order rather than by a discretionary or target exit.

Trade journaling

The practice of recording every trade's context, reasoning, execution and outcome so that patterns can be reviewed objectively later.

Trade performance analytics

Aggregated statistics on your executions, such as win rate, average win and average loss, and performance split by direction, setup or account.

Trading discipline / routine

A consistent end-of-day and end-of-week review process that converts raw trade data into behavioural change.

Trading edge

The statistical advantage a repeated process produces over a large sample of trades, net of commissions and fees.

VWAP

Volume-weighted average price: the session's average traded price weighted by volume, widely used as an intraday fair-value reference.

VWAP reclaim

A setup in which price trades back through volume-weighted average price and holds, signalling a shift back to the mean-value side of the auction.

Related Documentation