Three checkpoints in this guide
Follow the full walkthrough in order, or jump directly to one of its main sections.
For account performance, start with IBKR. For trading decisions, use a journal only after its trades match IBKR's records. IBKR's statements and PortfolioAnalyst already cover holdings, returns and attribution. A separate journal earns its place when you want to ask whether a setup, trading hour or repeated sizing decision helped your closed trades. Flex Queries can provide the underlying history through a configured sync or export, but a successful connection or recognized file is not proof that every position and fee has been reconstructed correctly.
This distinction matters more than a feature list. A portfolio can rise because an open position appreciated, a currency moved or cash was added. None of those facts by itself proves that a repeatable entry or exit decision worked. Conversely, a closed-trade review cannot replace a full account statement with dividends, financing, corporate actions, deposits and unrealized positions.
Decide what you actually need to review
| Your question | Start here | Why |
|---|---|---|
| “Did my account outperform, and where did the return come from?” | IBKR statement and PortfolioAnalyst | Holdings, cash movement and open positions belong in the account view. |
| “What did I actually make on closed trades?” | IBKR realized-trade report; reconcile any journal to it | Closed-trade P&L and whole-account return are different numbers. |
| “Do I keep repeating the same trading mistake?” | A reconciled journal with your setup, plan and review notes | Broker records show what happened, but not why you made each decision. |
IBKR lets traders customize Flex reports. Two queries or files both called “IBKR Flex” can therefore describe different slices of the same trading history. Before analyzing patterns, confirm that the chosen report includes the trades and period you meant to review, and that the journal's closed results agree with IBKR's own realized-trade view.
TSB offers IBKR Flex Web Service sync as well as file import in its supported-source directory. Sync is for a configured execution-level Trade Confirmation Flex query; file upload is for recognized export structures. Neither route means every custom IBKR report or account event will work. Start with a representative period, compare a closed position and its fees with the original IBKR report, and keep that report. If the history is incomplete or the numbers do not reconcile, do not use a polished chart as evidence that a strategy works.
If you need step-by-step export and file checks, use the existing broker import guide. This page is about what to conclude after the numbers match.
Why does my IBKR P&L not match a journal?
First compare like with like. IBKR's Trades reference says commissions are netted for realized cost basis/P&L, while its mark-to-market performance presents commissions separately. Its NAV summary also separates realized result, unrealized change, fees, deposits/withdrawals and FX translations. A journal showing closed-trade net P&L will not normally equal today's account-value change.
The following example is illustrative arithmetic, not a TSB or customer result. You buy one share at $100 and sell it at $110; each fill incurs $1 commission, and we assume no other charges or currency conversion. The $10 price difference is gross trading P&L. The closed trade's net P&L is $8 after the two commissions. If the IBKR realized P&L field already includes those commissions, subtracting the $2 again in a journal would falsely produce $6. Separately, a $500 deposit raises the account balance but is not $500 of trade profit. An open position's unrealized gain can change account value without entering this closed-trade example.
For a real mismatch, inspect these in order:
- Same account and date window? A consolidated or differently timed report may include other accounts, earlier fills or a later statement correction. Match account IDs and execution IDs rather than total-row labels alone.
- Same trade state? Compare closed to closed. An open lot, partial close or reversal can create a different position boundary from a simplistic buy/sell pairing. The report's chosen detail level matters.
- Gross, realized-net or MTM? Identify which IBKR column is in view before adding or subtracting fees. The commission treatment differs between realized and mark-to-market presentation.
- Same currency? Record the trade currency and the account's base-currency view separately. IBKR's Flex reference exposes
FX Rate to Base; that is reporting context, not permission to apply an arbitrary current exchange rate to an old fill. - Non-trade account activity? Dividends, interest, financing, taxes, deposits, withdrawals, corporate actions and FX translations can affect account reporting without being a strategy's closed-trade result.
- A missing or corrected row? Check original broker records and the import preview. Do not manually invent a missing opening trade or erase a discrepancy to make totals match.
If a difference remains, show both broker and reconstructed figures with the unexplained amount and stop drawing performance conclusions from that sample. For options, assignment, exercise and expiry require their own lifecycle proof; an ordinary execution file is not automatically a complete options-ledger source.
What a trading journal can add after reconciliation
IBKR's PortfolioAnalyst reports include performance by symbol, instrument, account and long/short exposure, plus trade summaries, fees and risk measures. It is already a strong choice when those answer your question. The separate journal question is narrower: which decisions in your own process keep working or failing?
For example, suppose you suspect your late-session entries harm performance. After confirming that the selected synced or imported history contains complete closed trades and net costs, label each trade with its intended setup and planned entry window. Separate late-session from earlier trades within the same setup and instrument, then compare sample size, net P&L per unit of planned risk, win/loss distribution and the worst sequences. Inspect a few losing and winning examples with their charts or notes. A few late losses are a lead to investigate, not proof of a bad time window; if the late trades were on a different market or risk size, the comparison is confounded. The next action may be a limited rule test—such as stopping new entries after a chosen time for the next review period—rather than changing strategies wholesale.
That is the honest TSB continuation: use your own verified trade history to ask what is working, what is holding performance back and what to test next. TSB should not claim it inferred your setup or emotion from IBKR fills alone. You may need to add context yourself, and the answer must be revisited as new trades arrive. If all you need is account-level attribution, keep using PortfolioAnalyst; another subscription is not automatically useful.
Once the IBKR history is reconciled, TSB AI Coach can examine the selected account and period and point to the trades behind a pattern. For “late entries,” a useful answer shows the compared groups and their sample sizes, not a confident guess from a statement summary. Coach may instead say the evidence is insufficient. It is grounded in the trader's recorded history, not a language model separately retrained on that person. Coach requires Full Access; the free journal workspace does not make Coach free.
What a useful AI Coach answer would have to show
Consider the broad question, “What keeps going wrong in my IBKR trading?” A generic AI summary of an uploaded statement cannot responsibly know whether the problem is fees, oversized trades, one setup, one session, or a short unlucky streak. The useful workflow narrows the question against the selected, recorded history and tells the trader what evidence is available. For example, a late-session hypothesis needs closed-trade timestamps, a consistent session definition, comparable instruments and enough observations in both groups. A setup hypothesis also needs the trader's setup labels; a broker fill does not contain the reason for the entry.
| Coach question after reconciliation | Evidence to ask for in the answer | When the answer must stop short |
|---|---|---|
| “Are my late entries worse than my earlier entries?” | Selected account and dates, the compared closed-trade groups, sample sizes, net results and several referenced trades | Missing timestamps, only a handful of late entries, different instruments or risk sizes, or an unexplained partial close |
| “Am I giving back too much in commissions?” | Recorded commission coverage and a fee-drag comparison for the same account, currency and trade group | Missing fees, another currency without reliable conversion, or an IBKR realized-net field that would be double-counted |
| “Which of my setups deserves more attention?” | Setup-labelled trades, comparable risk basis, outcome distribution and examples on both sides | Setup labels absent, changed mid-period, or concentrated in one market regime |
| “Do losses get worse after I break my plan?” | The trader's recorded plan/violation context tied to actual outcomes and a stated observation window | Plan context not recorded, or a causal claim inferred from a small, self-selected sample |
The useful habit is to return to the same question as more trades and context accumulate: inspect the entries behind a pattern, make one bounded change, then compare the next set of similar trades. A worthwhile answer should say “insufficient evidence” when the sample does not justify a verdict. Even a verified connection cannot turn incomplete history or a small sample into a dependable trading verdict.
A safe first review
- Choose IBKR Flex sync or a matching file-import route in TSB. Flex sync needs an execution-level Trade Confirmation query; a file needs a recognized export structure. Keep the original IBKR report, and never send access tokens or unredacted statements through an unverified channel.
- Review a small period containing complete closed positions. Check the right account, dates, trades, currency and costs. A successful connection or accepted file is not yet proof that the result is right.
- Match a fully closed position to IBKR's own realized result. Include one partial exit if you often scale out. If the numbers disagree, keep the discrepancy and stop before drawing strategy conclusions.
- Add the trading context the broker could not know: setup, planned risk, session, rule, chart and reason for exit. Then compare one meaningful pattern over enough comparable trades to avoid a conclusion from a lucky streak.
- Write one next-session decision and a review date. The journal is useful when later trades show whether that decision helped—not simply because it can draw another equity curve.
The answer to “which journal for IBKR?” is therefore conditional. IBKR itself covers substantial performance reporting. A separate trade-review system earns its place when its verified source route and your added decision context answer a recurring question the account report did not answer. If the route or reconciliation fails, the honest result is not verified, not a polished analytics screenshot.
