Once a journal exists, the next question is what to do with it. Eight threads on the journals board each describe one analysis, and together they make a pretty good monthly routine. Here's the combined version.
Clean data first
The pivot table thread starts with structure: one row per trade, with columns covering date, pair, session, setup, direction, R result, R cost and any flags. Its reply points at the hidden requirement, consistent spelling. One typo in a setup name and the pivot shows a separate "setup" with three trades. Dropdown lists fix that.
The first table to build is average R by setup and session. It usually shows one pairing carrying the results and another bleeding without anyone noticing.
Calendar patterns
The heat map thread puts weekdays in rows and hours in columns, with average or total R in each cell, coloured from red to green. The important reply is about sample size: a cell with two trades in it is noise with a colour on it, so group hours into three or four blocks. Friday afternoon showing red is very common, according to that thread, and an easy window to cut if yours does too.
Execution leaks
The planned R vs actual R thread adds two columns: the R you planned from entry, stop and target, and the R you got. Winners that come in under plan point to early exits or slippage. Losers bigger than 1R point to slipped stops or moved stops. A reply tags each gap with a cause, like early exit, late exit or slippage, so the leaks can be counted rather than guessed at.
The trade duration thread is a related check. Compare the average holding time of winners and losers. A common finding is losers lasting much longer than winners, which is the disposition effect showing up in your own data. The disposition thread suggests three numbers: average win against average loss in R, holding times, and how often stops get moved. Replies recommend putting stop and target on the server at entry, and only ever moving stops towards less risk.
Setup and pair quality
- Win rate against reward to risk. One point per setup on a scatter plot, with a break-even line drawn in. A reply found a favourite setup sitting right on break-even once costs were included. Another warns to label each point with its trade count, because a setup with fifteen trades can land anywhere by chance.
- Pair by pair. The pair review thread grouped results by pair and found two favourite pairs were near break-even after costs. The follow-up plan was demo with the same rules and the same journal fields, rather than dropping them on a feeling.
The missed setups thread logs setups that met the criteria but were skipped, with a reason. After a month, a pattern showed: valid setups being skipped right after a loss. A reply found a time-of-day pattern too, with far more skips in the last hour of the session. That's information a normal journal never captures.
Putting it together
None of these needs special software. A spreadsheet with consistent fields and an hour a month gets you all of them. The threads agree on the order: clean data, then setup and session, then execution leaks, then the finer cuts once there's enough sample.
Which analysis changed your trading the most once you actually ran it on your own journal?
Threads this pulls from
- Analyzing Your Trades With a Pivot Table
- Heat Maps of Your Performance by Day and Hour
- Comparing planned R to actual R on every trade, where execution leaks
- Plotting win rate against reward to risk for each setup, and what the picture showed
- Logging the trades I didn't take, my missed setups column and what it taught me
- A pair by pair review that made me drop two of my favourites
- Trade Duration: What Holding Time Says About Your Habits
- The Disposition Effect in Your Own Data
