Conjecture on monotonicity of worst-case error for spherical t-designs

Let $Q[\mathcal{X}_N]$ denote the quadrature rule associated with a spherical $t$-design consisting of $N$ points, and let $\mathrm{wce}(Q[\mathcal{X}_N])$ be its worst-case error.

Worst-case error monotonicity conjecture. For fixed $t$, if $N'>N$, then

$$\mathrm{wce}(Q[\mathcal{X}_{N'}])\leq \mathrm{wce}(Q[\mathcal{X}_N]).$$

The conjecture is motivated by numerical experiments comparing well-conditioned and efficient spherical $t$-designs. It predicts that increasing the number of points does not increase the worst-case error, but the source provides no proof or resolution.