Calibrated loss estimates

Figure 01 / 12

The evaluated population

The cohort contains 17,300 mature decisions, of which 606 have the defined synthetic outcome: Loss-bearing event. The outcome is known by construction here. In production, label uncertainty and selection must be recorded separately.

Figure data and text version

Outcome	Count
Loss-bearing event	606
Other labeled outcomes	16,694

Ranking separates higher-risk cases from lower-risk cases. Calibration checks whether stated probabilities agree with observed outcome frequencies in comparable groups.

Validate probability calibration on held-out data and keep severity and exposure assumptions explicit.

All amounts, rates, capacity limits, and outcomes in this case are synthetic. The three conditions are separate assumptions for comparison. A better result in the response condition is not measured proof that the proposed control causes that improvement. The figures expose the calculation and its limits; a real deployment needs its own evidence.

Read the result

The rule flags 721 of 17,300 mature decisions. Of those flags, 521 meet the synthetic target, giving 72.26% precision. It misses 85 target events. Under the stated cost assumptions, residual loss and operating friction total $36,184. The important result is the connection between the population, action, capacity, and outcome—not one isolated score.

Figure 02 / 12

Four outcomes of the rule

The rule flags 521 synthetic positives and 200 negatives. It misses 85 positives. A flagged item is a decision to intervene; it is not proof of fraud, a legal prohibition, or any other real-world conclusion.

Figure data and text version

Known outcome	Flagged	Not flagged
Loss-bearing event	521	85
Other outcome	200	16,494

Figure 03 / 12

Three rates with different denominators

Precision is 72.26%, recall is 85.97%, and the false-positive rate is 1.2%. Changing the denominator changes the meaning. This record keeps each numerator attached to the population from which it came.

Figure data and text version

Metric	Numerator	Denominator	Result
Precision	521	721	72.26%
Recall	521	606	85.97%
False-positive rate	200	16,694	1.2%

Figure 04 / 12

Precision changes with prevalence

This sensitivity plot holds recall at 86% and false-positive rate at 1.2%, then changes prevalence. It is an algebraic comparison, not a forecast. Even unchanged detection quality can produce a very different review queue when the base rate changes. Horizontal positions are the labeled observations or scenarios; equal spacing does not imply equal numerical increments.

Figure data and text version

Assumed prevalence	Precision %
0.1%	6.69
0.5%	26.48
1%	41.99
2%	59.39
5%	79.04
10%	88.84

Figure 05 / 12

The threshold trade-off

Six illustrative score bands use a stated pair of detection rates. Lower sensitivity can reduce false alarms but miss more target events. These points do not come from a trained model and do not establish the best operating threshold. Horizontal positions are the labeled observations or scenarios; equal spacing does not imply equal numerical increments.

Figure data and text version

Score band	True positives	False positives
Band 1	594	2,504
Band 2	570	1,336
Band 3	521	584
Band 4	436	200
Band 5	303	67
Band 6	152	17

Figure 06 / 12

A transparent loss-and-friction calculation

At $380 severity per missed synthetic positive, residual loss is $32,300. Review costs $2,884; lost contribution on false alarms is $1,000. The calculation assumes intervention prevents all flagged-positive loss and each false alarm loses the stated contribution. Relax those assumptions before applying it to a real policy.

Figure data and text version

Cost component	USD
Missed-positive loss	32,300
Review cost	2,884
False-alarm contribution	1,000

Figure 07 / 12

Observed outcomes mature over time

The final synthetic positive count is 606. Earlier observations reveal only a stated fraction. Comparing a day-1 cohort with a day-30 cohort would confuse label age with control quality. This curve models observation delay only; it does not change the final outcome. Horizontal positions are the labeled observations or scenarios; equal spacing does not imply equal numerical increments.

Figure data and text version

Days after event	Observed positives
1	109
3	212
7	364
14	497
30	606

Figure 08 / 12

Review demand and available capacity

The flag count is 721. The comparison capacity is an illustrative 692 reviews per cohort window. A mathematical rule can be coherent while its resulting workload exceeds the operating team’s capacity. Capacity is not permission to ignore an applicable mandatory control.

Figure data and text version

Queue measure	Items
Flagged for review	721
Available capacity	692
Excess demand	29

Figure 09 / 12

A feature is an observation with provenance

This evidence contract supports calibrate probabilities before using dollars. A value needs its event time, arrival time, scope, and source. Keeping unavailable evidence distinct from a measured zero prevents an outage from becoming a falsely reassuring feature.

Figure data and text version

Field	Example	Meaning
entity_ref	Calibrated loss estimates	Subject of this case
event_time	2026-09-18T09:00:00Z	When the event occurred
received_time	2026-09-18T09:00:02Z	When the system learned it
signal_status	repaired	Evidence quality, not an outcome
label_definition	Loss-bearing event	The target used in these calculations

Figure 10 / 12

Missing evidence changes the observed population

The cells show an explicitly constructed completeness profile for three signal groups. The stress condition removes more history and device evidence. Missingness does not prove the target outcome; it changes what the decision process knows.

Figure data and text version

Signal group	Available	Missing
Identity evidence	17,214	86
Activity history	17,040	260
Context signal	16,781	519

Figure 11 / 12

Evidence, score, and action remain separate

The policy can use calibrate probabilities before using dollars only within its approved scope. The action record must retain which evidence was available, which model or rule ran, and which action was actually applied. The final action can differ from the score recommendation when a separate constraint applies.

Figure data and text version

Stage	Record
Observe	Calibrated loss estimates: evidence as of the decision time
Evaluate	Rule flags 721 of 17,300 mature decisions
Apply	Record action, reason, owner, and expiry
Reconcile	Join the action to later outcomes without overwriting history

Figure 12 / 12

What the result cannot establish

Observed classifications do not reveal every counterfactual. The synthetic labels make arithmetic possible, but production decline data is selected by prior policy. Keep measured outcomes, assumed prevention, and unknown alternatives separate when reporting impact.

Figure data and text version

Claim	Evidence in this case	Limit
Detected target	521 known synthetic positives flagged	Production labels may be delayed or wrong
Prevented loss	Assumed 197,980 USD	Requires an intervention-effect assumption
Customer impact	200 synthetic negatives flagged	Not every flag causes abandonment
Unobserved alternative	Outcome without the action	Needs a valid evaluation design

Connect the result to the system

Validate probability calibration on held-out data and keep severity and exposure assumptions explicit.

Check the population, evidence, permitted action, and actual effect together. A balanced calculation can still use the wrong population; a successful response can still leave an unknown financial outcome. The case’s numerical result applies only to its stated assumptions.