Aggregated Scoring

Competitions like ECSC 2025 that feature both Jeopardy and Attack-Defense CTFs require a method of combining the scoreboards for these distinctly different competitions without valuing one or the other more highly.

The most common approach is to merge scoreboards by normalization and average.

jeopardy: list[float] # team jeopardy scores
attackdefense: list[float] # team attack-defense scores
w = max(jeopardy) / max(attackdefense)
aggregated = [s1 + s2 * w for s1,s2 in zip(jeopardy, attackdefense)]

This has some obvious drawbacks. Most notably, the first few teams in an Attack-Defense CTF with "classic" scoring typically have significantly more points than the rest, unlike in Jeopardy. Also, when SLA is valued highly by the scoring formula, this pratically constant score added to each team devalues the Jeopardy challenges when merging.

This was the motivation for developing ATKLABv2, which can be used with the Jeopardy scoring fomula to dynamically weight the Attack-Defense challenges, ideally mapping skill to points in the same way.

Another benefit of ATKLABv2 is that the final NOP team score can be used as a baseline measure for points which did not require any effort by teams to be earned, since it can neither gain points from attack nor defense.

Thus, this score can be removed from each team's attack-defense score before merging.

nop = attackdefense[0]
w = max(jeopardy) / (max(attackdefense) - nop)
aggregated = [s1 + (s2 - nop) * w for s1,s2 in zip(jeopardy, attackdefense)]

It is highly unlikely for a playing team to earn less points than NOP using this formula.