Annotations for observations that don't exist?

[wolfram06]

Sometimes I see this happening, where there’s more annotations for observations than actual observations at any given time. Is this a glitch?
Edit: this screenshot is showing how many observations are research grade, rather than being an annotation. However I see this happening with annotations too, but I’m assuming it’s because sometimes observations have more annotations than don’t at any given time.

[matthewvosper]

Not really. The only real datapoints are the dots. I think the line is just a best fit line for illustration purposes. In this case the total number of observations is low too, which will make the best fit line a bit more arbitrary and inaccurate. As more observations are made the anomaly will correct itself.