Three checkpoints in this guide
Follow the full walkthrough in order, or jump directly to one of its main sections.
Start from your account's actual transaction history, not a chart of the account value or an order list with no fill details. Then ask whether the chosen period includes both the opening and closing activity for the position you want to evaluate. A trading journal should separate trades from money movements, reconstruct closed positions with the correct contract multiplier, reconcile fees and only then compare setups. If the source file is incomplete, a detailed AI answer can still be false.
That distinction matters for tastytrade because its own API transaction reference describes account transactions as a paginated ledger that includes trade-related rows and cash movements, position transfers, dividends, interest and other events. A CSV export or accountant workbook may present them in a different layout, but the semantic problem is the same: a cash deposit is not a winning strategy, and a credit interest line is not a filled option sale. Do not calculate a trade win rate by dividing profitable ledger rows by all ledger rows.
tastytrade also has a real options backtesting tool that simulates how a selected strategy would have performed on historical market data. It can compare options structures and exit conditions, but those modeled trades are not the fills, costs and discretionary choices you actually made. A journal answers the different question: did your own repeated decisions survive their recorded costs and exceptions? Use the backtest to form a hypothesis; use reconciled personal history to check how you executed it. The two tools answer different questions.
Choose the right history before judging a strategy
Use the transaction history for the account and period you want to review, including the opening and closing activity. TSB supports specific tastytrade file exports, but a file being accepted is only the first check. In the preview, confirm that a deposit is not counted as a winning trade, an open option is not shown as closed, and one familiar position has the same contract size and after-fee result as the broker's record. If those checks fail, the performance conclusion is not verified; choose a fuller export or resolve the discrepancy before comparing setups.
| In the record | Journal treatment | Reason |
|---|---|---|
| A filled trade with its opening or closing action, contract and quantity | Possible part of a position; pair it with the other side | Can construct a supported closed position |
| Deposit, withdrawal or interest | Cash/account reconciliation only | Changes balance but not market-trade P&L |
| Opening leg without a matching close in selected range | Open or incomplete, not a closed winner | Prevents invented exit or basis |
| Option event without complete expiry/exercise/assignment economics | Explicitly limited until supporting cash/stock events reconcile | A missing stock leg can reverse the claimed result |
| Same symbol in a different account or expiry | Separate identity | Prevents unrelated fills from being paired |
Before import, check the account and date range. If an opening trade predates your download, widen the history before treating the later close as a complete strategy. Export the transaction history available for that account and period; menu labels can change. Confirm that the downloaded file contains the relevant open and close rather than relying on an order summary alone.
Option premium, multiplier and fee math
Option quotes and cash amounts are easy to confuse. Check the actual contract and quantity against the broker's result before trusting a journal total; an option with a different strike or expiry is a different position. TSB's supported files are not a promise that every option lifecycle or file layout can be reconstructed without that check.
Here is a hypothetical long option. Buy two contracts at a quoted $0.50 and sell both at $0.75, each with a 100-share multiplier. Cash outlay is $100, proceeds $150, so gross P&L is +$50. If attributable commissions and fees total $2.45, net is +$47.55. Those costs are illustrative, not a tastytrade fee quote. The useful check is simple: does this one trade show the same contract quantity and after-fee dollars in the journal and the broker's record?
Check one simple closed stock position and one option contract against the source report, including all fee columns. If the broker's number is gross but the journal's is net, explain the bridge. If a net figure already includes commissions, do not subtract them again. Futures introduce another instrument multiplier and possibly different commission treatment; one symbol label is not enough to infer dollar P&L. A recorded Value or Amount on one ledger row is a cash flow, not automatically the gain of the entire position.
Short option lifecycles need care too. Selling an option to open is not a stock sale; buying it back may close the short option. TSB has tested one file variant that preserves those actions and the usual equity-option multiplier, but that does not prove every spread, roll, exercise or assignment is complete. A roll closes one risk and opens another. If a report combines them into a single profit number without showing the new position, your “best setup” table may be misleading. Exercise and assignment can change underlying shares and basis; reconcile the corresponding cash and stock activity or mark that lifecycle not verified.
Why the account P&L can disagree with a journal
A tastytrade balance can change from closed trades, unrealized marks, deposits, transfers, fees and other ledger events. A journal's closed-trade P&L is intentionally a narrower number. If they do not match, do not assume the journal is wrong and add a fake adjustment trade. Reconcile in layers: account and currency, date range, open versus closed positions, transaction types, contract identities, gross/net fees, then cash/nontrade entries. The broker's official account history and statement are the economic control.
Suppose a week ends $800 higher but includes a $1,000 deposit and $200 net closed-trade loss. The account value rose, but the trading result did not. This is hypothetical and ignores open-position marks; a real reconciliation would name them too. The case shows why a deposit cannot be shown as a winning setup. Conversely, a trader may have positive closed trades while an unclosed futures position loses value. A file of historical closed trade rows does not prove the whole account is profitable.
Only after the record reconciles should the trader add context: setup version, intended risk, why a position was rolled, what was planned versus improvised, and which size changes were deliberate. Compare the same instrument type and contract economics; a one-lot equity-option premium sale and a futures trade should not share an unexplained “average P&L per trade” denominator. Segment the selected account and period before interpreting sessions, fees or behavior.
TSB AI Coach can take the review beyond a spreadsheet: it can compare repeated setups, fees, sessions, exits and sizing in the trader's selected history. It needs enough comparable trades and the fields required for the question. For example, it cannot diagnose fee drag when the file has no fees, or rank setups when none are labeled. It explains the calculated evidence and its limits; it does not retrain itself on the file. Ask: “Across my closed options trades, which setup still looks positive after costs and the largest winner, and which trades support that?” If an opening trade, option event or fee is missing, the answer should say what remains unverified.
For example, imagine hypothetical imported history with twenty reconciled, same-account closed option positions. Twelve with complete cost fields show +$180 gross and $220 in attributable commissions and fees: -$40 net for those twelve. A high gross win rate would hide the drag. The other eight have no fee fields, so -$40 cannot be called the after-cost result for all twenty.
An honest review separates the twelve cost-observed positions from the eight unknowns, points to the supporting records and asks for the account statement before recommending a strategy change. Deposits and interest belong in neither trade count. A useful AI answer can be detailed without pretending that missing evidence exists.
Use a supported file, then ask a better question
TSB's supported-source page lists tastytrade file exports. File support does not mean an automatic account connection. If you have a matching transaction file, select the right account and date range, inspect the import preview for skipped cash movements, then compare one familiar stock trade and one option position with the broker's report. After those checks, ask Coach which repeated setup survives its costs. The value comes from comparable personal evidence—not from declaring every mixed file complete.
Quick answer
Can tastytrade history be analyzed in a journal? Yes, when a supported export matches your account history. tastytrade's strategy backtest models possible historical trades; a journal reviews the transactions you actually made. Keep deposits out of results, reconcile one familiar position and its costs, and exclude open or incomplete option lifecycles from closed-trade conclusions. File support does not prove automatic connection or every option event. Evidence checked Sep 29, 2026.
