Monthly stats and the public journal

The monthly review is where the numbers get computed. Six statistics, each with a definition you use every time so that months can be compared, plus one decision: whether anything on the plan is a candidate for change. And there is a question of whether to keep the journal to yourself.

The six numbers

Computed from the journal rows for the month, and cumulatively since the current plan version began.

Average R and expectancy are the same number when trades are logged in R; both are listed because the risk management course defines expectancy from win rate and average win/loss, and computing it both ways is a check that the journal adds up.

A few notes on each:

Win rate is reported and then ignored for decisions. It is the number that will feel most important and it is the least useful.

Average R is the number. Positive over 50 or more trades, with a plan-follow score mostly at 3, is a working plan. Positive over 15 trades is a coin flip that came up heads.

Average loser should be near −1R. If it is −1.3R, stops are being widened or slipped; check the tags. If it is −0.6R, trades are being closed before the stop; either invalidation is doing its job (good, check the exit reasons) or early exits are (bad, check the tags).

Max drawdown in R, compared against the streak arithmetic from Module 2. A −6R drawdown at a 45% win rate is ordinary. A −15R one is either a very bad month or a plan-following collapse, and the plan-follow scores will say which.

Mistake frequency is the number to drive down first. A 26% mistake rate on a 0.2R plan is probably costing more than the plan makes. Fixing it is free.

Key idea: Average R and mistake frequency are the two numbers that drive decisions. Win rate is reported and ignored.

The monthly sheet

MONTH       September      Plan v1.1 (since 1 Aug)     Trades this month: 41
                            Cumulative trades on v1.1: 83

            This month   Cumulative v1.1   Expected (from journal front page)
Win rate      47%           44%              40 to 50%
Avg R        +0.21R        +0.18R            ~0.2R
Avg win/loss  1.7 / -0.9    1.8 / -1.0
Max DD       -4.2R         -6.1R
Mistake %     26%           31%

A-grade only: 29 trades, +0.34R avg.   B/C: 12 trades, -0.11R avg.
Top mistake tags: early-exit (6), late-entry (3), missed-time-stop (2)
Execution gap (planned - realised, month): 4.1R
Breakers: two-loss x3, daily stop x1, weekly stop x0

CANDIDATES FOR CHANGE (with trade counts)
  1. Early-exit tag: 6 this month, 14 cumulative. Execution, not plan. Action: ...
  2. Filter at 0.4%: trades on 0.4-0.5% days are 11 for -0.3R avg. Watch; 
     revisit at 30+ trades in that band.

DECISION    No plan changes. Execution action on early exits. v1.1 continues.

The comparison to the expected numbers, written down before the plan started, is what keeps you from being surprised by a normal month. If the month is inside the expected range, nothing needs to happen, even if it felt bad.

The candidates list

Every idea for a rule change goes on the list with the date, the reason, and the number of trades behind it. Module 5 is about what happens next. For now: the list is not the plan, and being on the list changes nothing about how you trade tomorrow.

The public journal

The trading journals forum is where members post their monthly sheet, or a weekly summary, in public. Consider it, for three reasons.

First, accountability. A plan-follow score you will post is a plan-follow score you will earn. Knowing that the early-exit count will be seen makes the early exit slightly harder to take, and slightly is enough over a hundred trades.

Second, pattern recognition from outside. Other traders will see things in your sheet that you cannot, because they are not inside it. "Your B-grade trades are all in the last hour" is the kind of comment that takes a stranger five seconds and you three months.

Third, it is the only honest record of the process most people never see. A public journal is not a P&L brag thread. Post in R, not dollars, with plan versions and mistake counts rather than screenshots of a broker balance. A journal that posts −3R months alongside +6R ones is worth reading. One that posts only the good weeks is not, and everyone can tell.

What to post: the monthly sheet as above, with the numbers in R, the plan version, and the one decision. What not to post: dollar amounts, account size, or any prediction.

Try it: Compute the six numbers for your last full month of trading, from whatever records you have. If you cannot compute one of them, note which field was missing; that is a journal fix. Then draft a monthly sheet in the format above, and decide whether you would be willing to post it as-is. If not, ask what you would want to change first: the numbers, or the honesty of the fields behind them.

Recap

• Six monthly numbers: win rate, average R, expectancy, average winner/loser, max drawdown, mistake frequency. Same definitions every month.
• Average R and mistake frequency drive decisions. Win rate is reported and ignored.
• Average loser far from −1R means stop widening (below) or early exits or working invalidation (above). The tags say which.
• Compare the month to the expected numbers written down in advance; a month inside the range needs no action, however it felt.
• Post the monthly sheet in R to the public journals forum for accountability and outside eyes. No dollars, no predictions.