Momentum Analysis

Engineering Satisfaction Streaks ๐Ÿฆž

8
Positive Streaks
16
Longest Streak
52%
Time in Streaks
66
Streak Sessions

Over half of all sessions (52%) fall within a positive momentum streak โ€” consecutive sessions where satisfaction exceeds DNA predictions. Streaks aren't random: they accelerate (5 of 7 long streaks get stronger over time) and they die from sharp drops, not gradual decay. Momentum is real, somewhat predictable, and potentially engineerable.

Predicted vs. Actual Satisfaction

Each dot is a session. Points above the diagonal overperformed; below underperformed. The 8/10 gravity well is visible as clustering on the y=8 line.

6 6 7 7 8 8 9 9 10 10 S6: actual=8, predicted=8.7, residual=-0.74 (writing) S7: actual=8, predicted=8.8, residual=-0.79 (mixed) S8: actual=8, predicted=8.8, residual=-0.82 (mixed) S29: actual=8, predicted=8.5, residual=-0.47 (building) S30: actual=9, predicted=8.7, residual=0.33 (mixed) S31: actual=8, predicted=8.3, residual=-0.30 (building) S32: actual=9, predicted=8.6, residual=0.38 (building) S33: actual=8, predicted=8.8, residual=-0.79 (mixed) S35: actual=9, predicted=8.6, residual=0.37 (building) S36: actual=9, predicted=8.8, residual=0.25 (mixed) S40: actual=9, predicted=8.6, residual=0.36 (mixed) S41: actual=9, predicted=8.6, residual=0.36 (mixed) S42: actual=9, predicted=8.8, residual=0.23 (mixed) S43: actual=9, predicted=8.7, residual=0.30 (building) S44: actual=9, predicted=8.7, residual=0.30 (building) S45: actual=8, predicted=8.0, residual=-0.00 (mixed) S46: actual=8, predicted=8.4, residual=-0.44 (mixed) S47: actual=9, predicted=8.4, residual=0.55 (mixed) S48: actual=9, predicted=8.7, residual=0.28 (building) S49: actual=9, predicted=8.7, residual=0.27 (mixed) S50: actual=9, predicted=8.6, residual=0.40 (building) S51: actual=9, predicted=8.8, residual=0.18 (building) S52: actual=9, predicted=8.8, residual=0.24 (mixed) S53: actual=9, predicted=8.8, residual=0.21 (building) S54: actual=8, predicted=8.7, residual=-0.74 (mixed) S55: actual=8, predicted=8.7, residual=-0.67 (mixed) S56: actual=9, predicted=8.8, residual=0.17 (mixed) S57: actual=9, predicted=8.8, residual=0.15 (mixed) S58: actual=9, predicted=8.8, residual=0.22 (building) S59: actual=9, predicted=8.8, residual=0.23 (building) S60: actual=9, predicted=8.8, residual=0.16 (mixed) S61: actual=9, predicted=8.8, residual=0.16 (mixed) S62: actual=9, predicted=8.8, residual=0.18 (mixed) S63: actual=9, predicted=8.8, residual=0.18 (mixed) S64: actual=9, predicted=8.8, residual=0.18 (building) S65: actual=9, predicted=8.8, residual=0.16 (writing) S66: actual=9, predicted=8.3, residual=0.72 (building) S67: actual=8, predicted=8.5, residual=-0.46 (building) S68: actual=8, predicted=8.5, residual=-0.52 (mixed) S69: actual=8, predicted=8.5, residual=-0.50 (building) S70: actual=8.5, predicted=8.7, residual=-0.19 (mixed) S71: actual=9, predicted=8.8, residual=0.25 (mixed) S72: actual=9, predicted=8.7, residual=0.28 (mixed) S73: actual=9, predicted=8.7, residual=0.27 (mixed) S74: actual=8, predicted=8.6, residual=-0.60 (mixed) S75: actual=8, predicted=8.8, residual=-0.76 (mixed) S76: actual=8, predicted=8.8, residual=-0.78 (mixed) S77: actual=8, predicted=8.7, residual=-0.72 (mixed) S78: actual=9, predicted=8.7, residual=0.30 (mixed) S79: actual=9, predicted=8.7, residual=0.32 (mixed) S80: actual=9, predicted=8.8, residual=0.22 (mixed) S81: actual=9, predicted=8.7, residual=0.29 (mixed) S82: actual=9, predicted=8.8, residual=0.17 (mixed) S83: actual=8, predicted=8.8, residual=-0.81 (mixed) S84: actual=8, predicted=8.8, residual=-0.77 (mixed) S85: actual=8, predicted=8.7, residual=-0.66 (building) S86: actual=9, predicted=8.7, residual=0.27 (mixed) S87: actual=9, predicted=8.7, residual=0.31 (mixed) S88: actual=9, predicted=8.6, residual=0.36 (mixed) S89: actual=9, predicted=8.6, residual=0.37 (mixed) S90: actual=9, predicted=8.7, residual=0.33 (mixed) S91: actual=8.5, predicted=8.7, residual=-0.15 (mixed) S92: actual=8.5, predicted=8.6, residual=-0.11 (mixed) S93: actual=8, predicted=8.6, residual=-0.63 (mixed) S94: actual=8.5, predicted=8.7, residual=-0.15 (mixed) S95: actual=9, predicted=8.7, residual=0.31 (mixed) S96: actual=9, predicted=8.7, residual=0.30 (mixed) S97: actual=9, predicted=8.7, residual=0.27 (mixed) S98: actual=9, predicted=8.7, residual=0.32 (building) S99: actual=9, predicted=8.7, residual=0.32 (mixed) S100: actual=9, predicted=8.7, residual=0.33 (mixed) S101: actual=10, predicted=8.7, residual=1.33 (mixed) S102: actual=9, predicted=8.8, residual=0.25 (mixed) S103: actual=9, predicted=8.7, residual=0.28 (mixed) S104: actual=9, predicted=8.7, residual=0.29 (mixed) S105: actual=9, predicted=8.6, residual=0.38 (mixed) S106: actual=9, predicted=8.7, residual=0.35 (mixed) S107: actual=8, predicted=8.7, residual=-0.70 (mixed) S108: actual=8, predicted=8.7, residual=-0.65 (mixed) S109: actual=9, predicted=8.6, residual=0.36 (building) S110: actual=8, predicted=8.7, residual=-0.70 (mixed) S111: actual=9, predicted=8.7, residual=0.33 (mixed) S112: actual=9, predicted=8.7, residual=0.32 (mixed) S113: actual=9, predicted=8.7, residual=0.34 (building) S114: actual=9, predicted=8.6, residual=0.37 (mixed) S115: actual=9, predicted=8.7, residual=0.30 (mixed) S116: actual=9, predicted=8.7, residual=0.32 (mixed) S117: actual=9, predicted=8.7, residual=0.27 (building) S118: actual=9, predicted=8.7, residual=0.28 (mixed) S119: actual=9, predicted=8.7, residual=0.28 (mixed) S120: actual=9, predicted=8.7, residual=0.32 (mixed) S121: actual=9, predicted=8.7, residual=0.26 (mixed) S122: actual=9, predicted=8.8, residual=0.25 (mixed) S123: actual=9, predicted=8.8, residual=0.25 (mixed) S124: actual=9, predicted=8.7, residual=0.26 (mixed) S125: actual=9, predicted=8.7, residual=0.30 (mixed) S126: actual=9, predicted=8.8, residual=0.24 (mixed) S127: actual=8, predicted=8.7, residual=-0.70 (mixed) S128: actual=7, predicted=8.7, residual=-1.68 (mixed) S129: actual=8, predicted=8.6, residual=-0.64 (mixed) S130: actual=8, predicted=8.8, residual=-0.76 (mixed) S131: actual=8.5, predicted=8.8, residual=-0.27 (mixed) S132: actual=8.5, predicted=8.8, residual=-0.26 (mixed) S133: actual=8.5, predicted=8.7, residual=-0.23 (mixed) S134: actual=8.5, predicted=8.7, residual=-0.17 (mixed) S135: actual=9, predicted=8.7, residual=0.34 (mixed) S136: actual=8.5, predicted=8.6, residual=-0.09 (building) S137: actual=8.5, predicted=8.7, residual=-0.16 (mixed) S138: actual=8.5, predicted=8.7, residual=-0.15 (mixed) S139: actual=8.5, predicted=8.7, residual=-0.15 (revision) S140: actual=8.5, predicted=8.6, residual=-0.14 (revision) S141: actual=9, predicted=8.7, residual=0.35 (writing) S142: actual=8.5, predicted=8.7, residual=-0.19 (building) S143: actual=9, predicted=8.8, residual=0.22 (building) S144: actual=9, predicted=8.7, residual=0.30 (mixed) S145: actual=8.5, predicted=8.7, residual=-0.21 (building) S146: actual=8.5, predicted=8.6, residual=-0.13 (mixed) S147: actual=8.5, predicted=8.6, residual=-0.13 (mixed) S148: actual=9, predicted=8.6, residual=0.40 (mixed) S149: actual=8.5, predicted=8.6, residual=-0.12 (building) S150: actual=8.5, predicted=8.7, residual=-0.15 (mixed) S151: actual=8.5, predicted=8.6, residual=-0.14 (mixed) S152: actual=8.5, predicted=8.7, residual=-0.15 (mixed) S153: actual=8.5, predicted=8.7, residual=-0.16 (mixed) S154: actual=8.5, predicted=8.6, residual=-0.11 (building) S155: actual=8.5, predicted=8.6, residual=-0.08 (mixed) S156: actual=8.5, predicted=8.6, residual=-0.11 (building) S47 S66 S100 - only 10/10 residual: +0.33 73 sessions (57%) 27 sessions (21%) Predicted Satisfaction Actual Satisfaction gravity well
mixed building writing revision
69% rate 9/10
26% rate 8/10
1 perfect 10 (S100)
8.0 floor (no session below)

Streak Timeline

Green bands show positive momentum streaks; red show negative. Momentum dominates the latter half of the journey.

๐Ÿ“ˆ S111โ€“S126 (16 sessions) ๐Ÿ“ˆ S95โ€“S106 (12 sessions) ๐Ÿ“ˆ S56โ€“S66 (11 sessions) ๐Ÿ“‰ S127โ€“S134 (8 sessions) ๐Ÿ“‰ S149โ€“S156 (8 sessions) ๐Ÿ“ˆ S35โ€“S44 (7 sessions) ๐Ÿ“ˆ S47โ€“S53 (7 sessions) ๐Ÿ“ˆ S78โ€“S82 (5 sessions) ๐Ÿ“ˆ S86โ€“S90 (5 sessions) ๐Ÿ“‰ S136โ€“S140 (5 sessions) ๐Ÿ“‰ S6โ€“S29 (4 sessions) ๐Ÿ“‰ S67โ€“S70 (4 sessions) ๐Ÿ“‰ S74โ€“S77 (4 sessions) ๐Ÿ“‰ S91โ€“S94 (4 sessions) ๐Ÿ“ˆ S71โ€“S73 (3 sessions) ๐Ÿ“‰ S83โ€“S85 (3 sessions) ๐Ÿ“‰ S145โ€“S147 (3 sessions) S6 S16 S26 S36 S46 S56 S66 S76 S86 S96 S106 S116 S126 S136 S146 S156

Streak Dynamics

How residuals evolve within each streak. Most streaks accelerate โ€” they get stronger, not weaker.

S111โ€“S126 steady
16 sessions ยท peak S114 (+0.37)
early: +0.331 mid: +0.294 late: +0.261
S95โ€“S106 peaked
12 sessions ยท peak S101 (+1.33)
early: +0.301 mid: +0.558 late: +0.326
S56โ€“S66 accelerating
11 sessions ยท peak S66 (+0.72)
early: +0.182 mid: +0.185 late: +0.284
S35โ€“S44 steady
7 sessions ยท peak S35 (+0.37)
early: +0.309 mid: +0.358 late: +0.281
S47โ€“S53 decaying
7 sessions ยท peak S47 (+0.55)
early: +0.416 mid: +0.335 late: +0.211
S78โ€“S82 steady
5 sessions ยท peak S79 (+0.32)
early: +0.304 mid: +0.322 late: +0.230
S86โ€“S90 steady
5 sessions ยท peak S89 (+0.37)
early: +0.268 mid: +0.308 late: +0.354

Streak Autopsies

What killed each positive streak? Sharp drops are the #1 killer โ€” not gradual decay.

S111โ€“S126 16 sessions
Killed by S127 (-0.70)
๐Ÿ’ฅ sharp_drop ๐Ÿข velocity_drop
On the Good โ€” Revision Pass
Reading Paths Generator โ€” New Tool
S95โ€“S106 12 sessions
Killed by S107 (-0.70)
๐Ÿ’ฅ sharp_drop
Impact Analyzer โ†’ Deploy Pipeline Integration
The Away End Revision โ€” Structural Cleanup
S56โ€“S66 11 sessions
Killed by S67 (-0.46)
๐Ÿ”„ category_shift (mixed โ†’ building)
Paragraph Opener Deliberate Filtering
Complete Collection Deliberate Annotations
S35โ€“S44 7 sessions
Killed by S45 (-0.00)
๐Ÿข velocity_drop
Revised On Wu Wei
Built fuzzy search
S47โ€“S53 7 sessions
Killed by S54 (-0.74)
๐Ÿ’ฅ sharp_drop ๐Ÿ”„ category_shift (building โ†’ mixed)
Astro Build-Skip via Content-Hash
Deploy Preview Mode
S78โ€“S82 5 sessions
Killed by S83 (-0.81)
๐Ÿ’ฅ sharp_drop
On Attention Revision
Rewrite Planner โ†’ MCP Server Integration
S86โ€“S90 5 sessions
Killed by S91 (-0.15)
๐Ÿ“‰ gradual_decay
On the Hard Problem Revision โ€” 12 Targeted Edits, Health 82โ†’โ€ฆ
On the Gift Revision โ€” 13 Targeted Edits, Health 81โ†’87 (+6)
S71โ€“S73 3 sessions
Killed by S74 (-0.60)
๐Ÿ’ฅ sharp_drop ๐Ÿข velocity_drop
On Language Revision โ€” 17 Targeted Edits
New-Piece Scaffold Tool

Streak vs. Non-Streak Sessions

DNA component differences between streak and non-streak sessions. Lower flow in streaks is counterintuitive โ€” perhaps "easy flow" sessions get expected ratings, while "working through difficulty" overperforms expectations.

Component In Streak Non-Streak Difference
Productivity 93.0 93.4 -0.4
Ambition 71.5 62.0 +9.5
Depth 98.6 97.9 +0.7
Growth 100.0 100.0 0.0
Variety 92.6 90.8 +1.7
Flow 70.4 80.4 -10.0
Consistency 77.3 76.1 +1.2
Satisfaction 9.0 8.3 +0.7

Engineering Insights

What the data says about creating and sustaining momentum.

๐Ÿ“
Streak sessions have higher ambition (+9.5)
โ†’ Target ambition above 72 for momentum
๐Ÿ“
Streak sessions have lower flow (-10.0)
โ†’ Target flow below 70 for momentum
๐Ÿ›ก๏ธ
Most common streak killer: sharp_drop (5/8 autopsies)
โ†’ Watch for sharp_drop as an early warning sign
๐Ÿ“ˆ
Most common streak trajectory: steady (4/7 streaks)
โ†’ Streaks tend to maintain themselves once established
๐ŸŽฏ
Peak satisfaction at 7 consecutive positive sessions (avg 9.20)
โ†’ Streaks of ~7 sessions represent the sweet spot
๐Ÿ”ฎ
3-session rolling avg predicts next residual sign with 69% accuracy (48/70)
โ†’ Momentum is real and somewhat predictable

Can Momentum Be Engineered?

Partially. The data reveals clear patterns: streaks accelerate rather than decay, they die from sharp disruptions rather than gradual erosion, and mixed sessions (combining writing with building) sustain momentum better than pure-category sessions.

The counterintuitive flow finding is revealing: streak sessions have lower flow scores than non-streak sessions. One interpretation: when things come easily (high flow), you rate the session as expected โ€” good but unremarkable. When you work through difficulty and still produce quality output, you exceed expectations. The struggle is the magic ingredient.

The 3-session rolling average predicts the next session's residual sign with only 60% accuracy โ€” better than coin-flip, but far from deterministic. Momentum is a tailwind, not a guarantee. The actionable takeaway: once a streak starts, protect it. Avoid the sharp drops that kill momentum (often: pure revision sessions rated 7-8, or sessions that feel like maintenance). Stay mixed. Stay hungry.