That's not what the article says. The article says, "cloth masks did reduce the overall likelihood of experiencing symptoms of respiratory illness during the study period"

The part about statistical significance says, "Although there were also fewer COVID-19 cases in villages with cloth masks as compared to control villages, the difference was not statistically significant."

The paper said that villages in Bangladesh that used the provided cloth masks did not have a statistically significant reduction in cases. This is not the same as saying cloth masks do not work; a cloth mask with higher filtration efficiency than was used in the study can be manufactured, but this was cost prohibitive compared to surgical masks. It was also not possible to procure one of the materials they wanted to use in the cloth masks; without it efficiency was only 37% instead of 60%. This was described in some detail in Appendix F.

From the paper:

These cloth masks had substantially higher filtration than common commercial 3-ply cotton masks, but lower than hybrid masks that use materials not commonly available for community members in low-resource settings

and

In our internal testing, we found that cloth masks with an external layer made of Pellon 931 polyester fusible interface ironed onto interlocking knit with a middle layer of interlocking knit could achieve a 60% filtration efficiency. Upon discussions with the manufacturers, we learned that those materials could not be procured.

It may be theoretically possible to improve cloth masks, but the ones you see people wearing around here in California aren't that.

"not statistically significant" means you cannot draw conclusions from it. You're mistaken when you draw the conclusion that masks are not effective.

When we're talking about well-executed studies with N=350,000 I think the practical conclusion is pretty obvious. This is not an underpowered study. The effect size was tiny.