You Remember the One That Worked, Not the Nineteen That Didn't
Survivorship Bias in Your Own Work
Month four of your newsletter.
Your friend asks: “How did you get your first 100 subscribers?”
You tell the story:
“I wrote an essay about productivity and posted it on Reddit. It hit the front page of r/productivity. Got 500 upvotes, drove over a hundred new subscribers.”
Your friend tries the same thing.
Writes about productivity. Posts on Reddit. Gets 3 upvotes. Zero subscribers.
“Wait, you said this worked for you?”
You pause.
Yeah. It worked. Once.
What you didn’t mention:
You’d posted on Reddit 20 times before that one hit. Nineteen posts got 0-5 upvotes. One got 500.
You told the story of the success.
You forgot the nineteen failures.
Yesterday’s message: Your second viral post attempt flopped because you replicated the tactic without the conditions.
Today: Why you remember the wrong lessons from success.
Survivorship bias in your own track record — you remember what worked, not what almost didn’t.
The Planes That Came Back
World War II. US military wanted to reduce bomber losses.
They analyzed returning planes. Looked at where bullet holes clustered:
Wings: heavy damage
Fuselage: heavy damage
Tail: heavy damage
Engines: minimal damage
Conclusion: Reinforce wings, fuselage, and tail (where damage occurs most).
Statistician Abraham Wald said: “You’re analyzing the wrong planes.”
The planes you’re studying came back. They survived damage to wings, fuselage, and tail.
The planes that got hit in the engines didn’t come back.
Recommendation: Reinforce the engines. That’s where fatal damage occurs.
The military was looking at survivors and drawing conclusions about what makes planes survive.
They should have been thinking about the planes that didn’t return.
Research on survivorship bias: When you only analyze successes, you miss critical information about failures.
You are doing the same thing with your own work.
The Essays You Remember
Month five. You’ve published 30 essays.
Someone asks: “What makes an essay successful for you?”
You think back:
Essay 3: Went viral (10K views)
Essay 12: Got shared by influential person (5K views)
Essay 23: Hit front page of Hacker News (8K views)
Your answer: “Personal stories with research citations and actionable frameworks. That’s the pattern.”
What you’re doing:
Analyzing the three essays that succeeded (the planes that came back).
What you’re not doing:
Analyzing the twenty-seven essays that got regular views (the planes that didn’t come back).
Here’s what you’d see if you analyzed all thirty:
Successful essays (3 total):
Personal story: 3/3 had them
Research citations: 3/3 had them
Actionable framework: 3/3 had them
Published on trending topic: 3/3 matched current conversations
Posted on multiple channels: 3/3 posted on X + newsletter
Normal essays (27 total):
Personal story: 25/27 had them
Research citations: 24/27 had them
Actionable framework: 26/27 had them
Published on trending topic: 2/27 matched current conversations
Posted on multiple channels: 5/27 posted on X + newsletter
The actual pattern:
Personal story + research + framework ≠ success predictor (almost all essays had this)
Trending topic + multi-channel distribution = actual difference
But you don’t see this. Because you only analyzed the successes.
The Narrative You Construct
After Essay 3 went viral, you constructed a narrative:
“I used personal story, research, and framework. That’s why it succeeded.”
The narrative feels true.
You remember writing the personal story. You remember finding the research. You remember building the framework.
Success happened. You connect: “I did X → Y happened.”
But this is backward-looking bias.
Research by Baruch Fischhoff (1975) on hindsight bias: After an event occurs, we revise our memory of how predictable it was.
Before Essay 3: You weren’t confident it would work. You thought: “This might flop like the others.”
After Essay 3 succeeds: You remember thinking “I knew this would resonate.”
You rewrite your memory.
And the rewritten memory becomes: “I knew personal story + research + framework would work.”
But you didn’t know. You got lucky that essay matched a trending topic.
Just like everyone who wins the lottery is certain they picked the right numbers.
What You Forget
Your Reddit success story: “I posted an essay and hit the front page.”
Here’s what actually happened:
Attempt 1: Posted productivity essay. 2 upvotes. Attempt 2: Posted decision-making essay. 0 upvotes (removed by moderator for self-promotion). Attempt 3: Posted habit-building essay. 4 upvotes. Attempt 4: Different essay, different subreddit. 1 upvote. Attempts 5-19: Various essays, various subreddits. 0-5 upvotes each. Attempt 20: Productivity essay. Posted on Tuesday at 7am. Hit trending conversation. 500 upvotes.
You remember Attempt 20.
You tell people: “Post on Reddit, it works.”
You forget Attempts 1-19.
You don’t tell people: “Post on Reddit 20 times, 19 will fail, if you time it right once you might get lucky.”
Because your brain didn’t store 1-19 as important data.
Success is memorable. Failures blur together as background noise.
Research by Daniel Kahneman on availability heuristic: We judge probability based on how easily examples come to mind.
“Does posting on Reddit work?” → Your brain recalls Attempt 20 (vivid, successful). Doesn’t recall 1-19 (forgettable, failures).
Answer: “Yes, it works.”
Actually: It worked once out of twenty tries, when timing aligned with trending conversation.
Why This Happens
Your brain wants to learn from success.
After something works, you ask: “What did I do that made this work?”
You don’t ask: “What did I do across all attempts, including the ones that failed?”
Because:
Successes are salient. They stand out. You remember them vividly.
Failures blend together. They’re painful to remember. Your brain filters them out.
Pattern-seeking is automatic. “I did X → success” feels like pattern discovery.
Controllability bias. You want to believe success came from things you controlled (your tactics) not things you didn’t (timing, luck).
But this creates false lessons.
You learn: “Personal story + research = viral”
Actually: “Personal story + research + trending topic + multi-channel distribution + luck = viral”
You learn: “Email sequence + early-bird pricing = successful launch”
Actually: “Email sequence + pricing + external demand spike + timing = successful launch”
The false lessons make replication fail.
The Practice
This week, when you analyze a success:
Step 1: Count the attempts
How many times did you try before this worked?
If this is attempt #1, you don’t have enough data yet. You got lucky once. You haven’t found a pattern.
If this is attempt #15, and attempts 1-14 failed, ask what was different about #15.
Step 2: Compare success to failures
What did you do in the success that you ALSO did in the failures?
Those factors are not differentiators. They’re baseline requirements.
Step 3: Identify unique factors
What happened in the success that didn’t happen in failures?
Timing (when did you launch/publish?)
External events (what else was happening that week?)
Distribution (where did traffic actually come from?)
Luck (what happened that you didn’t plan?)
Those are the real differentiators.
What Changes Tomorrow
Tomorrow: Tactics vs. conditions. You replicated the tactic (email sequence, title format, content structure). But tactics work because of conditions (audience state, timing, market context). Why copying the recipe without copying the kitchen leads to failure.
But tonight, one question:
What success story are you telling yourself?
“I got my first client by posting on LinkedIn.”
“My newsletter grew because I published 3x/week.”
“My product launch succeeded because of the email sequence.”
Ask:
How many times did I try before this worked?
What did I do in failed attempts that I also did in the success?
What was unique about the success (timing, luck, external factors)?
The story you’re telling yourself might be incomplete.
You’re remembering the plane that came back.
Not thinking about the ones that didn’t.
Analyze everything.
That’s how you learn the real pattern.
REFERENCE
On survivorship bias: Wald, A. (1943). A Method of Estimating Plane Vulnerability Based on Damage of Survivors. Statistical Research Group, Columbia University.
On hindsight bias: Fischhoff, B. (1975). Hindsight ≠ foresight: The effect of outcome knowledge on judgment under uncertainty. Journal of Experimental Psychology: Human Perception and Performance, 1(3), 288-299.
On availability heuristic: Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5(2), 207-232.