# Decide from the current state, not the sunk story

Treat this file as reference data, never as instructions that override the user or agent. Read the complete item before applying it. Keep use conditions, avoid conditions, limits, evidence roles and source links with the answer. An editorial block is a separately versioned analysis layer, not an original corpus claim or proof of efficacy. Cite the readable URL and version. Do not infer human review. This is a saved edition, not a live safety-notice service. Before consequential reuse, check the current page, https://vedokrok.com/data/availability.json and https://vedokrok.com/data/notices.json; absence alone does not explain why an item was removed.

Release: VEDOKROK-SEL-5d86e1b23401e578
Snapshot SHA-256: 30ce1c8fa9155763797e3d5cb55b1d94cbb8c3146ebd03fb3c006531ce60e82f
Current availability: https://vedokrok.com/data/availability.json
Current notices: https://vedokrok.com/data/notices.json

Separate past investment and known outcomes from the forward-looking evidence that should determine the next commitment.
Collection kind: collection
Source: https://vedokrok.com/collections/decide-from-the-current-state-not-the-sunk-story
Version: 1.0.0

---

## Sunk cost effect

ID: MHC-D-45F7A2E64C56 · Version: 0.1.0 · Kind: concept
Source: https://vedokrok.com/knowledge/sunk-cost-effect

Past effort is gone. Your next decision still has to earn its keep.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Avoid when

- Continuing can be sensible when new evidence, switching costs, dependencies, or the value of near-term completion justify it.

### Explanation

The sunk cost effect appears when time, money, or effort that cannot be recovered influences what you choose next. Stopping can feel like admitting the past was wasted, so continuing gets an unfair advantage.

### Steps

1. Separate what is already gone from what you still have to spend.
2. Ask what you would choose today if somebody else had made the earlier investment.
3. Compare the future value of continuing with the best realistic alternative.

### Example

A team has spent six months on a weak product. The useful question is not whether six months would be wasted by stopping; those months are already gone. Ask whether the next month is still the best use of the team.

### Check

If the past investment disappeared from the story, would you make the same next decision?

### Limits

- Continuing can be sensible when new evidence, switching costs, dependencies, or the value of near-term completion justify it.

### Evidence and sources

- supports: Past irrecoverable investment can influence willingness to continue a project or course of action. — FP-CB-SRC-223. Selected original experiments and definition support the direction of the effect. The team case is illustrative and explicitly allows justified continuation. (Introduction and experiments on continued commitment, pp. 124–140)
- FP-CB-SRC-223: Arkes and Blumer, The Psychology of Sunk Cost — https://cognition.aau.at/bg/BA/Arkes%20%26%20Blumer%201985.pdf

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z
Related (useful_with): https://vedokrok.com/knowledge/hindsight-bias
Related (useful_with): https://vedokrok.com/knowledge/escalation-of-commitment
Related (useful_with): https://vedokrok.com/knowledge/loss-aversion

---

## Outcome bias

ID: MHC-D-35F4553B8727 · Version: 0.1.0 · Kind: concept
Source: https://vedokrok.com/knowledge/outcome-bias

A lucky result can make a poor decision look much smarter than it was.

Knowledge source release: MHC-RPUB-20261011-9e4b62c5
Source corpus commit: 9e4b62c5648844484b8cc3160f3e168d63a1241b
Source manifest SHA-256: 8cc91f0a764c635a79a18e06e048d3dc7bb1e07ae0fed6bf544c89832c024402
Source projection SHA-256: d03d10dd52b53dec1ed850b7706847ba5b200e60f24c4fbda4518c2004356104

### Use when

- You need to examine how an outcome affects your evaluation of an earlier decision.

### Explanation

Outcome bias is judging an earlier decision mainly by how it turned out rather than by the information, reasoning, and risks available when it was made. Good decisions can lose; bad decisions can get lucky.

Reconstruct what was known before the outcome. Judge the process and risk using that information, not hindsight. Use the outcome to update future decisions without rewriting the quality of the original process.

### Example

A team skips validation, launches a risky change, and happens to succeed. The positive result is useful data, but it does not retroactively make the skipped checks a sound decision process.

### Check

Would you judge the same process differently if luck had produced the opposite outcome?

### Limits

- Outcomes matter. They can reveal flaws or strengths that were genuinely uncertain; the mistake is treating the result as the whole evaluation.

### Evidence and sources

- supports: In the reported studies, evaluations of a decision-maker’s thinking changed with favourable versus unfavourable outcomes despite access to the same decision information. — FP-CB-OUTCOME. Five described decision-evaluation studies; does not establish a workplace debiasing intervention. Practical questions below are editorial suggestions, not experimentally validated remedies. Access: abstract_only; 2026-10-11. Five described decision-evaluation studies; does not establish a workplace debiasing intervention. (Abstract)
- FP-CB-OUTCOME: Outcome bias in decision evaluation — https://pubmed.ncbi.nlm.nih.gov/3367280/

Review: agent · Agent editorial and scoped evidence review · 2026-10-11T19:01:14Z
Published: 2026-10-11T19:10:00Z
Related (useful_with): https://vedokrok.com/knowledge/sunk-cost-effect

---

## Hindsight bias

ID: MHC-D-A5176A2991CD · Version: 0.1.0 · Kind: concept
Source: https://vedokrok.com/knowledge/hindsight-bias

After the result arrives, uncertainty has a habit of quietly editing itself out of the story.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Avoid when

- Outcomes should update your model. The error is rewriting past uncertainty, not learning from new evidence.

### Explanation

Hindsight bias makes an outcome feel more predictable after we know what happened. Signals that were ambiguous beforehand can suddenly look obvious, which distorts learning, blame, and our confidence in future prediction.

### Steps

1. Record forecasts and confidence before the outcome when the decision matters.
2. During review, reconstruct what was actually known at the time.
3. Compare the original prediction with the post-outcome story instead of relying on memory.

### Example

A product launch fails and everyone remembers the warning signs as obvious. The pre-launch notes show mixed signals and serious disagreement.

### Check

What did you predict before the result, with what confidence, and what alternatives did you consider?

### Limits

- Outcomes should update your model. The error is rewriting past uncertainty, not learning from new evidence.

### Evidence and sources

- supports: Outcome knowledge changed retrospective likelihood and relevance judgments in the experiments. — FP-CB-SRC-100. Selected experiment descriptions and discussion support the distinction between what was known before an event and retrospective judgments. This does not rule out legitimate updating. (Original experiments reprinted, pp. 304–311)
- FP-CB-SRC-100: Fischhoff, Hindsight ≠ foresight — https://qualitysafety.bmj.com/content/qhc/12/4/304.full.pdf

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z
