PropProof
Combine Checkup

Combine Checkup

Sample Trader · Topstep $50K Combine (SAMPLE DATA) · MNQ · 2026-05-13 to 2026-06-09 · 155 trades over 19 trading days
This combine is still alive — and this checkup is about keeping it that way. You haven't breached a rule, but the patterns below are the leaks quietly working against your target. Left unchanged, they're what fails accounts like this one.
-$1,179total net P&L
$48,821ending balance
51%win rate
+$99.08avg winner
-$118.50avg loser

Executive summary

Your closest call

The nearest this combine has come to the trailing threshold was on 2026-06-09, when your balance sat just +$209 above the liquidation line. That cushion is exactly what the habits below eat into — every leak narrows the distance between an ordinary red day and a blown account.

The evidence

Behavioral findings

Each pattern below was tested against your fills. A finding is only marked DETECTED when the numbers support it — and what's clean is named too, because knowing what you're NOT doing wrong matters.

PatternWhat your fills show
Averaging into losers
DETECTED
2 times you laddered into a position moving against you. 2 of those holds lost money, net -$750. Worst hold: 2026-06-09 at 10:06, long adds from 21,309.00 out to 21,297.75 — 12 contracts on at once, -$450. Peak concurrent exposure across the account: 12 contracts at 10:10 on 2026-06-09.
Revenge sizing
DETECTED
Next trade after a loser averaged 3.07 contracts vs 2.21 after a winner (n=70/66).
Chasing after losing streaks
DETECTED
After 2+ straight losers you re-entered in a median 1m 55s vs 6m 15s normally; those 32 chased trades net -$1,224.
Cutting winners, riding losers
DETECTED
Winners held 3m 46s, losers 12m 58s (3.4× longer). Avg winner +$99.08, avg loser -$118.50 — a 1.20:1 payoff inversion: it takes 1.2 winners to pay for one loser.
Overtrading when losing
DETECTED
Red days averaged 10.2 trades vs 6.3 on green days (9 red / 10 green). Losing days were your busiest days — activity went up exactly when edge went down.
Session fatigue
DETECTED
Across the account's life, your edge peaks at trade #9 of the day: the first 9 trades made -$549; everything after them lost +$630.
Time-of-day leak
DETECTED
Everything entered after 12:00 net -$726 (it touched 3 days). With a hard 12:00 shutoff — no other change — this account ends at $49,547 instead of $48,821.
Weekday leak
DETECTED
Wednesdays net -$1,224 across 41 trades, while Thursdays made +$690 — one weekday is quietly taxing the others.

The rules that would actually have helped

We replayed 18 candidate rules against your actual trade list — same trades, same order, each enforced mechanically (no hindsight trade-picking). Below are only the ones that moved the needle for you, ranked by impact (biggest first); a rule that wouldn't have changed your outcome isn't dressed up as if it would.

RuleBalance with rulevs your actualOutcome
No trading before 10:00
Skip every entry before 10:00 — sit out the early session. · 48 trades avoided
$50,978+$2,157improves the outcome
Cap every trade at 2 contracts
Never exceed 2 contracts. Oversized trades are scaled to 2 lots (P&L scaled proportionally — an approximation). (proportional approximation) · 0 trades avoided
$50,026+$1,205improves the outcome
Stop after 12 trades in a day
Cap the day at 12 trades — walk away once you've taken 12. · 9 trades avoided
$49,601+$780improves the outcome
All 3 rules enforced together
the rules overlap, so this is its own replay — not the sum of the rows above
$51,531+$2,710improves the outcome

Your actual ending balance was $48,821 (still active). Each ‘vs your actual’ figure is the improvement that one rule alone would have made; the bottom row is all of them enforced together.

Knowing the rules and keeping them are different problems. PropProof Guardian — in development — connects to your account and enforces rules like these mechanically, in real time: oversized positions get flattened, and a rule-breaking day ends whether you agree with it in the moment or not. Checkup customers get first access and founding pricing. To join the waitlist, reply to this report with the word GUARDIAN.

What this data cannot show

The export contains fills only. It cannot show what you saw on the chart, the trades you skipped, or whether a stop was in the market or in your head. Where the data was thin (small sample sizes are flagged with n= counts above), treat the pattern as a hypothesis to watch, not a verdict. All P&L figures are net of fees.