← Back to calculator

Lethal + Relentless: why kill chance can rise as BS gets worse

It is possible to set up a Kill Team shooting scenario where the calculator reports a higher kill chance at BS 5+ than at BS 2+, even though every other input is identical. This is not a calculator bug. It is a real consequence of how Lethal and Relentless are defined in KT2024. This note walks through the math for the specific scenario that prompted it: 4 attack dice, normal damage 3, crit damage 4, Lethal 5+, Reroll Relentless, Piercing 1, versus a 12-wound defender with a 3+ save and (optionally) Indomitus.

The two rules

Lethal X+ means an attack die scores a critical hit on an unmodified roll of X or higher, instead of the default 6+. Lethal cannot promote a die that would otherwise have failed: the calculator clamps the effective crit threshold up to the hit threshold, so for example Lethal 5+ at BS 6+ still only crits on a 6.

Relentless lets you re-roll any number of your attack dice. The calculator models this as one re-roll per failed die, where each re-rolled die is statistically identical to the original.

Per-die crit probability with Lethal 5+, Relentless

For BS 2+ through 5+, the effective crit threshold is 5 (so crit faces are {5, 6}, 2 of 6) and BS only controls the norm/fail boundary. After one allowed re-roll per fail, per-die probabilities are:

Across the BS 2+ to 5+ range shown, per-die crit probability rises monotonically as BS degrades. Each fail is a free re-roll, and every re-roll has the same 2/6 chance of landing in the fixed crit band. Worse BS just means more re-rolls feeding the same funnel. (At BS 6+ the clamp kicks in and Lethal 5+ collapses back to crit-on-6 only, so the monotonic claim does not extend past 5+.)

Average damage falls, but that is not the same as kill chance

Expected damage per die (crit dmg 4, norm dmg 3, before saves):

Averages behave intuitively: better BS, more average damage. But kill chance is a tail probability, not an average. The shape of the distribution matters more than its mean.

Why the tail flips

Against a 12-wound target with 4 attack dice and 2 defender dice (Piercing 1 strips one save die), the defender cancels roughly 1.3 hits on average, so about 2.7 hits leak. To reach 12 leaked damage you essentially need 3+ crits to leak. Norms (3 damage each) get preferentially absorbed by the defender's norm saves and rarely close a 12-damage gap on their own in this two-save-dice setup.

So kill probability is dominated by P(≥ 3 crits in 4 dice):

That 2.4× rise from BS 2+ to BS 5+ matches the inversion the calculator reports. The BS 2+ attacker throws more total hits, but most of them are norms; the BS 5+ attacker throws fewer hits, but a far higher share of them are crits, and only crit-heavy rolls breach the kill line.

Indomitus amplifies the inversion (in this setup)

Indomitus adds one defender norm save when the defender rolls two fails. A norm save can:

• cancel one norm hit, one-for-one; or
• pair with another norm save to cancel one crit hit, two-for-one.

With only 2 defender dice (Piercing 1) and no cover saves, the chance of also having a second norm save available to pair off a crit is small, so most of Indomitus' benefit lands on norm hits. In that specific setup:

• At BS 2+, the attacker generates many norm hits — Indomitus' bonus save almost always finds a norm hit to cancel, damage drops noticeably, and kill chance drops a lot.
• At BS 5+, the attacker generates zero norm hits — the only way the bonus norm save helps is by pairing with another norm save to cancel a crit, which requires the defender to roll two pre-Indom norms (just 9/36) on top of two fails, an outcome that does not exist on 2 dice. So in this scenario the bonus is a no-op.

Important caveat: with 3 defender dice (no Piercing), or with auto-normal saves from cover, the extra norm save can more often pair with another norm save to cancel a crit. In those cases Indomitus is not a pure "norm-hit eraser" and the amplification pattern above will be different. The behaviour described here is specific to the Piercing 1 / no-cover scenario.

Take-aways

1. Across BS 2+ to 5+ with Lethal 5+, per-die crit probability is independent of BS without re-rolls (the crit band is fixed). Adding Relentless makes per-die crit probability increase as BS worsens, because every extra fail becomes another shot at the same fixed crit band. At BS 6+ the clamp kicks in and this trend stops.
2. Against a high wound threshold relative to attack volume, kill chance is driven by crit-spike probability, not by expected damage. The two can move in opposite directions.
3. Indomitus and similar norm-promoting abilities compound the effect in setups where extra norm saves rarely pair off into crit cancels (e.g. the 2-defender-dice scenario above). With more defender dice or cover saves they can also help against crits.
4. The rule of thumb "better BS always means more kills" can fail whenever Relentless is combined with Lethal: Relentless converts every failure into another chance at the fixed crit band, so worse BS feeds more dice into that band. Without Relentless, per-die crit probability does not depend on BS at all, so the inversion does not appear.

← Back to calculator