The first image is from the CR and the second image is for the second placing. This could be easily detected by Strava. Their “threshold” for detecting if the people started/ended the segment is too high.
In regard to the cheaters, I found the following situations too frequently:
Change their devices recording GPS points to be higher than 1 second, thus making straight lines (the shortest distance between two points is a straight line);
They know the start and end positions of the segment, so they pause their device's recording before the start and after already entering the segment is when they start again the recording. They do the same for the ending, that is, pause the recording before the segment end and resume it some meters after the segment end.
My suggestions would be:
Only allow a segment to be valid if all the points from start to end of the segment happen at an interval of 1 second;
Don't allow segments with wobbling GPS data;
Only allow to count for segment if the device is collecting GPS data before and after the start and end points, that is, the recording should incorporate all the segment and not parts of the segment as now it is possible.