correct, the red is the highest stress and blue is lowest (see scale on the right) and the only way to distinguish compressive and tensile is by intuition. the loads are in-plane (these are some of the limitations i mentioned, there are others as well). the alignment of the suspenson is actually pretty good so there isn't a lot of off-axis component, but some is there. i don't think it's a very large factor overall, though it would contribute to earlier failure in extreme cases.
yes, making it more of a triangle or even just increasing the radius in the highest-stress area would make it considerably stronger. i initially thought that the crank is shaped this way to clear the shock at full extension except the shock doesnt extend far enough for it to be an issue. maybe previous versions did? i can only guess.
anyway, as i mentioned this wasn't meant to be a definitive analysis, just a rough pass to get a feel for whether design is reasonable or marginal. that's all i use simulations for - to get a reality check and do some simple optimization (like relieving the stress risers). and of course to make pretty pictures
when i designed the aluminum bellcranks for the dp1 i found that the rodend pivot would yield at 4g in original design because the material was too thin in that spot so after simulation i made the part thicker and sized it for 8g. the simulation was quite useful for that.
