I hate those graphs. You can manipulate the axes on any graph to make it look like the two variables closely match each other and increase at the same rate

US spending on science, space, and technology vs Suicides by hanging, strangulation and suffocation overlay

https://www.tylervigen.com/spurious-correlations

Total revenue generated by arcades vs. computer science doctorates... wait a second I think this one might actually be onto something

So both have a similar shaped growth curve, but large deviations from that curve on the suggested independent variable are not reflected at all in the suggested dependent variable? Sounds like two unrelated phenomenon with similar long-term shape but no direct link.

The correlation is perhaps not the smoking gun OP thinks but your analysis seems to be based on the assumption there is no time lag in the population of children 6-21 with autism and glyphosate use on grains.

The most obvious one would be a six year lag. But it might have variance depending on crop cycles and whether glyphosate is handed off through the placenta and/or mother’s milk, and depending on the proposed mechanism of action and weaning age.

> The correlation is perhaps not the smoking gun OP thinks but your analysis seems to be based on the assumption there is no time lag in the population of children 6-21 with autism and glyphosate use on grains.

No, if there was just a time lag, than the variations would still be reflected, but with a time lag.

There is variation in the data that could be reflective of that time lag.

If the population had no awareness of the effect and no ability to observe the impact of grain consumption on their children, then maybe it would proceed like some chemical process. But the population can learn and change behavior.

My point is that your dismissal, in the way you framed it, is possibly just as much a random walk as the correlation in the way it is presented. Neither is conclusive.

Correlation isn't causation (Wet streets cause rain).

Unless you have some kind of mechanism to explain it.

Well glyphosate accumulates on wheat products and ends up in your food (e.g. breakfast cereal) [1]. So it's not impossible glyphosate is affecting us negatively, but there just isn't any real evidence yet.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8622992/

Wouldn't we expect to see sharply reduced incidences of ASD in gluten intolerant population if this if grain is the primary vector for this?

Unless glyphosate is involved with the development and presentation of gluten intolerance as well.

There is enough evidence that Bayer recently had to set aside $10B to settle current and future lawsuits:

https://www.cnn.com/2020/06/24/health/bayer-monsanto-roundup...

More than one-half of human microbiome are intrinsically sensitive to glyphosate

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9145961/

To be fair

https://www.tylervigen.com/spurious-correlations

Autism prevalence growth may be driven by increased medicalization of society causing more diagnoses and increased medical training to hand out the diagnosis. We have little reliable data on the past with diagnoses only becoming a thing in the 1990s. One should be very careful extrapolating from the existing data since it's not a reading of actual numbers of people affected, but the rates of diagnosis.