The A+/A/B/C setup-grading framework, popularized by Lance Breitstein, says: don't size every trade the same. A+ setups (your best conditions present) get full size; A, B, and C setups get progressively smaller allocations. The journal's job is to track which grades you took and how each grade actually performed — then the AI Coach surfaces whether your grading is accurate. Most traders' biggest expectancy leak is sizing C setups like A+ setups.
- Lance Breitstein popularized the modern A+/A/B/C grading framework — credit where due.
- A+ = full size; A = 80%; B = 50%; C = 25% or skip. Ratios are starting points, not gospel.
- Define grades from your own data — top winners give A+ criteria, top losers give C warning signs.
- Tag every trade with a grade. Ungraded trades are usually the impulsive ones.
- The AI Coach surfaces whether your grading is accurate or whether you're overgrading.
- Recalibrate grade definitions every 30–60 days based on what the data shows.
Where the framework comes from
The clearest, most accessible explanation of A+/A/B/C setup grading in the public market comes from Lance Breitstein, an equities prop trader who teaches the framework on YouTube and through his courses. Earlier roots exist in the SMB Capital prop-firm tradition (Mike Bellafiore's "One Good Trade" and "The PlayBook" cover similar ground), and Brett Steenbarger's trading-psychology work touches on the discipline of size-by-quality.
The core insight is simple: not all valid setups deserve the same position size. A 1.5R risk on a marginal setup is a 1.5R risk on a marginal setup. Over hundreds of trades, treating mediocre setups like great ones is the single most common reason competent traders fail to grow accounts.
What each grade means
A+ setups
Your highest-conviction conditions present. Time of day is right (your best session). Instrument is your strongest (the one your A+ trades cluster in). Context is right (the broader market is doing what your setup needs). Indicator alignment is clean (no conflicting signals). Mental state is right (sleep was good, no distractions, no recent emotional trade). When all of these line up, you're looking at an A+. Full size — your maximum risk-per-trade.
A setups
Strong, but missing one or two A+ conditions. Maybe the time-of-day is right but the context is mixed. Maybe the alignment is clean but you slept poorly. ~80% of A+ size. Still worth taking; just respect that conviction is slightly lower.
B setups
Technically valid — the setup's core rule is satisfied — but several A+ conditions are missing. The trade has positive expectancy historically, just not as much as A+. ~50% of A+ size. Take with explicit awareness that the win rate or R-multiple expectation is lower than headline numbers suggest.
C setups
The setup is barely valid. You can find a reason to take it but you're reaching. Most C-grade trades shouldn't be taken at all. When taken, they get ~25% size or less. The C grade exists primarily so that the trade gets logged honestly — if you take a marginal trade, log it as C rather than upgrading it to B post-hoc.
How to define your grades
The grade definitions are personal — they come from your own trade history, not from a generic template. The two-pass exercise:
Pass 1: top winners → A+ criteria
Pull your top 10 winning trades over the last 90 days. For each, write down what was true at the moment of entry — time of day, instrument, indicator alignment, market context, your mental state, day of week, position in your weekly cycle, anything else you can identify. Look for patterns across the 10. The recurring features are your A+ conditions.
Pass 2: top losers → C warnings
Pull your top 10 losers. Do the same exercise. The features that recur in your worst trades are your C warning signs. If "took the trade after lunch" appears in 6 of 10 losing trades, "post-lunch" is a C-grade signal.
From those two passes, you have a working definition. The middle grades (A, B) are degrees between the extremes. Refine over time.
A starting size table
| Grade | Size (multiple of base) | Take it? |
|---|---|---|
| A+ | 100% (e.g., 1.5R risk) | Always |
| A | ~80% | Yes |
| B | ~50% | With awareness |
| C | ~25% or skip | Usually skip |
The exact ratios matter less than the discipline of having any size differentiation. A trader who sizes uniformly across grades is making a hidden bet that all valid setups have equal expectancy — which is almost never true.
How the AI Coach checks your grading
Once every trade carries a grade tag (A+, A, B, C), the AI Coach can answer questions that would take hours to compute manually:
- "What's my win rate by grade?" — A+ should beat A, A should beat B, etc. If A+ doesn't lead, your grading is broken.
- "What's my expectancy in R by grade?" — same logic. A+ should have the highest R-expectancy.
- "What's the distribution of grades I take?" — if 60% of your trades are A+, you're probably overgrading.
- "Which conditions present in my recent A+ trades match my historical definition?" — checks for definition drift.
- "Are there any A+ trades where I noted I felt forced?" — checks for emotional miscategorization.
The most common trader mistake with this framework: every trade becomes an A+ in real time. The mental move is "this looks great, must be A+" — and after a few weeks, you're back to uniform sizing under the guise of grade discipline. The fix: have an explicit checklist of A+ conditions written down, and only mark A+ when every condition is met. If even one is missing, it's an A. If two are missing, it's a B.
Recalibrating your grades
Every 30–60 days, re-run the two-pass exercise. Your top winners 60 days from now will share features that may differ from the features you used to define A+ today. The market changes; you change. The grade definitions should drift with you — slowly, deliberately. A grade definition that hasn't been touched in a year is probably stale.
How this combines with the playbook
The grades live inside the broader playbook. Each named setup in your playbook ("Asia-session BTC long off 4H demand", for example) has its own A+/A/B/C criteria. The grade is not "is this a good setup in general" — it's "for this specific named setup, are the A+ conditions for this setup present?" Different setups have different A+ checklists.
Grade-tracked journaling with AI grade-quality analysis.
Tag every trade A+/A/B/C and let the AI Coach surface whether your grading is honest. $19.95/mo. 14-day money-back guarantee.
Frequently asked questions
Where did the A+/A/B/C setup grading come from?
The clearest popular explanation comes from Lance Breitstein, an equities prop trader who teaches the framework on YouTube and in his courses. The core idea: not all setups deserve the same position size. A+ setups (your best, highest-conviction conditions present) get full size. A and B setups get smaller size. C setups (technically valid but conviction-weak) get minimum size or get skipped. The journal's job is to track which grades you took and how each grade performed over time.
How do I define what an A+ setup is for me?
Start with your existing trade history. Look at your top-10 winning trades and ask: what did they have in common? Time of day, instrument, indicator alignment, volume context, your mental state — list the recurring features. Those are the A+ conditions. Then look at your top-10 losing trades and list the conditions present — those are the C-grade warning signs. The grades are personal; they're defined by your own data, not by a generic checklist.
How much should I size differently by grade?
A common starting framework: A+ = full size (your maximum risk-per-trade, e.g., 1.5R), A = 80% size, B = 50% size, C = 25% size or skip. The exact ratios are less important than the discipline of having any size differentiation at all. Most traders take every trade at the same size, which is the silent killer of long-term expectancy — losses on bad setups are the same magnitude as wins on great ones.
How does the AI Coach use grade data?
Once you tag each trade with a grade (A+, A, B, C), the AI Coach can answer questions you can't answer manually: "what's my win rate by grade?", "what's my expectancy by grade?", "are my A+ trades actually better than my A trades or am I miscategorizing?", "what's the time-of-day distribution of my A+ vs C trades?" Without grade tagging, every trade looks the same to the AI.
What if my A+ trades don't outperform my A or B?
Then your grading definition is off — and that's useful information. Either your A+ criteria don't actually predict outcomes (you've picked features that feel important but aren't statistically), or you're miscategorizing in the moment (calling something A+ when it's really A). The AI Coach's job is to surface this gap. Recalibrate your grade definitions every 30–60 days based on what your data actually shows.
Should every trade be graded, or only some?
Every trade. The discipline of grading forces the question "is this actually a good setup or am I bored?" before entering. Even C grades. The C-grade is meaningful — "I'm taking this knowing it's C" is more honest than skipping the grade box. Trades that show up ungraded are usually the ones that were taken impulsively.
How does grading interact with risk management?
Grading is risk management — sizing differently by grade is exactly how you manage variance. The traditional rule "never risk more than X% per trade" sets the ceiling for A+. The grade structure tells you what to take below the ceiling. A trader risking 1.5R on every trade including C setups is doing worse risk management than a trader who takes 1.5R on A+ and 0.4R on C.
Where can I learn more about setup grading?
Lance Breitstein's public material (YouTube, blog, course) is the most accessible. Mike Bellafiore's books ("One Good Trade", "The PlayBook") cover related ground from the SMB Capital perspective. The framework is also covered in Brett Steenbarger's trading psychology books. The mechanics translate across asset classes — equities prop traders developed the discipline, but crypto and futures traders apply the same logic.