LYNCH · Master Combinatorial Backtest

3336 combinations tested (Grade × Sizing × Exit × Regime-gate). Each backtest runs 5 years, real dollars, whole shares, $50,000 position cap, SMB commissions.
Part 1 — The Winner

Best Return-per-Dollar-of-Pain (walk-forward validated)

A+B+C grade · Sizing: UVXY-mom × Reactive · Exit: Swing pure (hold 5d) · Gate: no gate

Trades
435
Net $
$+71,454
Max DD
$14,992
Return/DD
477%
Win rate
45.1%
Sharpe
0.74

Best Raw Dollar Amount (walk-forward validated)

A+B+C grade · Sizing: UVXY-mom × Reactive · Exit: Swing pure (hold 5d) · Gate: no gate

Trades
435
Net $
$+71,454
Max DD
$14,992
Return/DD
477%
Win rate
45.1%
Sharpe
0.74
Part 2 — The 10-Year-Old Report

What we did today, plain English

Imagine we had a robot that could time-travel back 5 years and try EVERY combination of LYNCH trading rules to find the best one. Today the robot tested 3,336 different rule combinations — like mixing ingredients in a kitchen to find the perfect recipe.

The 4 ingredients we mixed

  1. Which stocks to trade (Grade A only? B only? All grades?)
  2. How big to bet (Same every time? Bigger when market calm? Smaller after losses?)
  3. When to sell (Hold 5 days? Sell fast? Half-and-half?)
  4. When to skip trades entirely (Skip if VIX rising? Skip if market is dropping?)

The BIG finding

Out of 3,336 recipes, 1071 of them survived the honest test (made money in BOTH halves of the 5-year window). The winners weren't lucky — they worked consistently.

The winning recipe

The best recipe by "how smooth the ride is" (Return/Drawdown):

The best recipe by "most total dollars":

What patterns showed up in the TOP 50 recipes?

Instead of just picking ONE recipe, we asked: which ingredients appear most often in the best 50 recipes? If an ingredient shows up in the winner AND in most of the runners-up, it's a REAL edge (not luck).

Which GRADE appears in Top 50?

ValueAppears in top 50% of top 50
A+B+C50100%

If one grade dominates the top 50, it's the real edge — not the others.

Which SIZING RULE appears in Top 50?

ValueAppears in top 50% of top 50
UVXY-mom × Reactive4284%
Reactive (½ after loss)612%
VIX-mom × Reactive24%

If sizing rules matter, the winners will all use similar sizing.

Which EXIT STRATEGY appears in Top 50?

ValueAppears in top 50% of top 50
Swing pure (hold 5d)1326%
Hold 7d1122%
Hold 10d1122%
20/80 time-split612%
20/80 at 2R + BE stop36%
50/50 time-split36%
Doc 80/20 at 1R + BE stop36%

If one exit dominates, that's the way to sell. If it's mixed, exit doesn't matter much.

Which REGIME GATE appears in Top 50?

ValueAppears in top 50% of top 50
no gate1836%
VIX-mom < 1.15 only1836%
UVXY-5d < 15% only1428%

If "no gate" wins, skipping isn't worth it. If specific gates cluster, they're pulling out bad trades.

Part 3 — Full Scoreboards

Top 20 by RETURN/DRAWDOWN (walk-forward validated)

Smoothest ride per dollar of pain. If you hate drawdowns, look here.

GradeSizingExitGate nNetDD R/DDSharpeWin%Avg RWalk
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)no gate435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)VIX-mom < 1.15 only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)UVXY-5d < 15% only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 7dno gate435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 7dVIX-mom < 1.15 only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 7dUVXY-5d < 15% only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 10dno gate435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 10dVIX-mom < 1.15 only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 10dUVXY-5d < 15% only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)no gate274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)VIX-mom < 1.15 only274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)UVXY-5d < 15% only274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveHold 7dno gate274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveHold 7dVIX-mom < 1.15 only274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveHold 7dUVXY-5d < 15% only274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveHold 10dno gate274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveHold 10dVIX-mom < 1.15 only274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × ReactiveHold 10dUVXY-5d < 15% only274$+42,446$8,916476%0.5444.9%+0.11ROK
A+B+CUVXY-mom × Reactive20/80 at 2R + BE stopno gate435$+61,440$13,080470%0.7246.7%+0.14ROK
A+B+CUVXY-mom × Reactive20/80 at 2R + BE stopVIX-mom < 1.15 only435$+61,440$13,080470%0.7246.7%+0.14ROK

Top 20 by TOTAL DOLLARS (walk-forward validated)

Most raw profit. If you can stomach the DD, look here.

GradeSizingExitGate nNetDD R/DDSharpeWin%Avg RWalk
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)no gate435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)VIX-mom < 1.15 only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)UVXY-5d < 15% only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 7dno gate435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 7dVIX-mom < 1.15 only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 7dUVXY-5d < 15% only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 10dno gate435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 10dVIX-mom < 1.15 only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveHold 10dUVXY-5d < 15% only435$+71,454$14,992477%0.7445.1%+0.26ROK
A+B+CUVXY-momSwing pure (hold 5d)no gate435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momSwing pure (hold 5d)VIX-mom < 1.15 only435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momSwing pure (hold 5d)UVXY-5d < 15% only435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momHold 7dno gate435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momHold 7dVIX-mom < 1.15 only435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momHold 7dUVXY-5d < 15% only435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momHold 10dno gate435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momHold 10dVIX-mom < 1.15 only435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-momHold 10dUVXY-5d < 15% only435$+66,083$21,787303%0.7445.1%+0.26ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)no gate570$+65,083$22,884284%0.5142.5%+0.16ROK
A+B+CUVXY-mom × ReactiveSwing pure (hold 5d)VIX-mom < 1.15 only570$+65,083$22,884284%0.5142.5%+0.16ROK

Top 15 by SHARPE RATIO (risk-adjusted return)

Best consistency. High Sharpe = fewer surprises.

GradeSizingExitGate nNetDD R/DDSharpeWin%Avg RWalk
A+B+CUVXY-mom × Reactive20/80 at 2R + BE stopSPY > 20-DMA only38$+16,353$4,576357%1.7550.0%+0.50ROK
A+B+CUVXY-mom20/80 at 2R + BE stopSPY > 20-DMA only38$+17,327$6,085285%1.7550.0%+0.50ROK
A+B+CReactive (½ after loss)20/80 at 2R + BE stopSPY > 20-DMA only38$+11,424$3,617316%1.7550.0%+0.50ROK
A+B+CVIX-mom × Reactive20/80 at 2R + BE stopSPY > 20-DMA only38$+11,424$3,617316%1.7550.0%+0.50ROK
A+B+CFlat20/80 at 2R + BE stopSPY > 20-DMA only38$+11,420$4,685244%1.7550.0%+0.50ROK
A+B+CVIX-mom20/80 at 2R + BE stopSPY > 20-DMA only38$+11,420$4,685244%1.7550.0%+0.50ROK
A+B+CFlatSwing pure (hold 5d)SPY > 20-DMA only38$+11,842$5,398219%1.6147.4%+0.52ROK
A+B+CFlatHold 7dSPY > 20-DMA only38$+11,842$5,398219%1.6147.4%+0.52ROK
A+B+CFlatHold 10dSPY > 20-DMA only38$+11,842$5,398219%1.6147.4%+0.52ROK
A+B+CVIX-momSwing pure (hold 5d)SPY > 20-DMA only38$+11,842$5,398219%1.6147.4%+0.52ROK
A+B+CVIX-momHold 7dSPY > 20-DMA only38$+11,842$5,398219%1.6147.4%+0.52ROK
A+B+CVIX-momHold 10dSPY > 20-DMA only38$+11,842$5,398219%1.6147.4%+0.52ROK
A+B+CUVXY-momSwing pure (hold 5d)SPY > 20-DMA only38$+17,869$6,752265%1.6147.4%+0.52ROK
A+B+CUVXY-momHold 7dSPY > 20-DMA only38$+17,869$6,752265%1.6147.4%+0.52ROK
A+B+CUVXY-momHold 10dSPY > 20-DMA only38$+17,869$6,752265%1.6147.4%+0.52ROK

Bottom 10 — the WORST recipes (don't do these)

Included so you can see what fails. Notice which patterns to AVOID.

GradeSizingExitGate nNetDD R/DDSharpeWin%Avg RWalk
C onlyFlatDay-trade (all out day 1)no gate203$-5,762$10,933-53%-0.4540.4%-0.04Rno
C onlyFlatDay-trade (all out day 1)VIX-mom < 1.15 only203$-5,762$10,933-53%-0.4540.4%-0.04Rno
C onlyVIX-momDay-trade (all out day 1)no gate203$-5,762$10,933-53%-0.4540.4%-0.04Rno
C onlyVIX-momDay-trade (all out day 1)VIX-mom < 1.15 only203$-5,762$10,933-53%-0.4540.4%-0.04Rno
C onlyFlatDay-trade (all out day 1)UVXY-5d < 15% only199$-5,595$10,766-52%-0.4440.2%-0.04Rno
C onlyVIX-momDay-trade (all out day 1)UVXY-5d < 15% only199$-5,595$10,766-52%-0.4440.2%-0.04Rno
A+B+CFlatDay-trade (all out day 1)SPY > 20-DMA only38$-1,916$4,090-47%0.0844.7%+0.01Rno
A+B+CVIX-momDay-trade (all out day 1)SPY > 20-DMA only38$-1,916$4,090-47%0.0844.7%+0.01Rno
C onlyFlatDay-trade (all out day 1)UVXY-5d < 15% only213$-3,888$8,381-46%-0.4843.7%-0.04Rno
C onlyVIX-momDay-trade (all out day 1)UVXY-5d < 15% only213$-3,888$8,381-46%-0.4843.7%-0.04Rno