Methodology
How Lucent calculates weapon performance — and why it matters.
01
The Problem with Simplified TTK Models
Most weapon stat tools calculate Time-to-Kill using pre-averaged damage values — a single expected damage number before shots-to-kill are calculated. It's a reasonable simplification, but averaging damage before modelling outcomes loses information that matters.
Lucent takes a different approach. Every weapon profile is evaluated through 15,000 simulated engagements with body-hit probabilities estimated from competitive play patterns, and TTK metrics are built directly from the resulting outcome distribution.
02
Expected TTK — E[TTK]
Lucent uses a simulation-based modeling approach. For each weapon profile, 15,000 statistically weighted engagements are simulated using body-part hit probabilities derived from competitive gameplay analysis.
This produces a full distribution of possible shots-to-kill outcomes rather than a single deterministic value.
From this distribution, Lucent derives several metrics:
Note — Because E[TTK] is simulation-derived, values carry a negligible Monte Carlo sampling margin that does not meaningfully affect weapon comparisons.
03
TTK Definition
Lucent defines TTK as the elapsed time between firing the first shot and target elimination.
Fire rate, open-bolt delay, and bullet velocity are incorporated where relevant to maintain consistency across weapon classes and engagement distances.
04
Body-hit probability weightings
Hit probabilities are calibrated from competitive play patterns. With estimates for close-range and long-range weapon classes.
| Region | Relative Weight |
|---|---|
| Head/Neck | Situational |
| Upper Torso | High |
| Lower Torso | Medium |
| Arms | Medium |
| Legs | Lower |
Note — Lucent models all eight body regions independently. Grouped categories are shown here for readability.
05
Distance-Weighted E[TTK]
Single-distance TTK values do not fully represent weapon performance across real matches. Lucent models expected engagement distances using calibrated log-normal distributions based on competitive Warzone gameplay patterns.
Weapons are then evaluated across these weighted engagement ranges to produce distance-adjusted E[TTK] rankings that better reflect practical in-game performance.
Separate distance models are applied for close-range and long-range weapon classes.
Conceptually, this integrates E[TTK] over the full engagement-distance distribution. In practice, the firing component (shots fired, converted via fire rate) is separated from bullet travel time before integration, since travel time can be resolved analytically from the log-normal mean — the two components are combined only in the final result, avoiding any double-counting between them.
06
Damage Per Mag & Damage Per Second
DPM and DPS use the same body-hit probability weightings as E[TTK], applied to damage rather than time. Expected damage per shot is calculated by weighting each body region's damage value by its probability of being hit.
DPM is expected damage per shot multiplied by magazine size. DPS is expected damage per shot multiplied by fire rate. Both are calculated per engagement distance, following the same damage drop-off profile used for TTK.
Weighted DPM applies the same distance-weighting model used for Weighted E[TTK], integrating DPM across realistic engagement ranges by weapon class.
07
Accuracy-Adjusted Metrics
By default, Lucent's metrics assume Theoretical accuracy — every shot fired lands. This is a baseline for comparison, not a realistic hit rate.
Pro users can apply Custom accuracy values, tested per weapon, to see accuracy-adjusted E[TTK] and damage metrics. For TTK, reduced accuracy increases the expected number of shots fired to land the required hits. For DPM and DPS, reduced accuracy directly reduces expected damage landed, since a fixed number of shots connect less often.
Accuracy adjustment is mathematically exact for E[TTK], DPM, and DPS. It is not applied to Modal, P₁₀, or P₉₀ TTK, since the transformation that holds exactly for an expected value does not hold exactly for percentiles of a distribution.
Questions or corrections?
The methodology is intentionally transparent. If you spot an error, have a suggestion, or want to discuss the approach, get in touch.