Pat,
I appreciate your sharing of your hunt—successes and failures. Let me add a bit to your observations of Ashby’s studies. To his credit, he tried to normalize the study of arrow penetration in real skin, tissue and bone. But working with biological systems is tough. So many variables, and most outside the experimenters control.
I’ve read all the reports and looked at the regression analysis he did while examining all the variables he tested. The regression analysis attempts to see what variable(s) can actually be ‘modeled’ using a formula to predict results that match experimental measurements. It’s been a while since I went through all the reports and statistics, so I may not cite everything just right (a dangerous thing to do on this site). Statisticians use the ‘r squared’ value to say whether or not a variable can be accurately modeled. R-squared is a number displayed in percents or as a decimal fraction. For example, an R squared value of 0.5 or 50% can roughly be interpreted as the model will match reality about half the time. Most statisticians don’t get excited about a model until the value of R squared gets above 0.8 or higher, meaning the model and reality will agree 80% of the time or more. Of all the tests done by Ashby, only one variable was statistically shown to have a predictable effect on penetration and that was the amount of FOC. I believe this variable was the only one to show an R squared value of more than 0.5.
Now hold on with the poisoned arrows. All of the data collected is valuable in the sense it gives us some empirical insight, but FOC is the only one that was consistently able to explain the real world results.
So, Pat, I’m not surprised to hear your single bevel results did not show great penetration. It’s one of the variables that matched predicted results less than half the time.
I understand many folks think Ashby’s results are infallible and will flame my comments. But folks, the statistics I’m quoting are not mine, they are the ones Dr. Ashby published. I’m just trying to say which ones are most likely to be replicated by others. And I believe the only one that fits that criteria is the amount of FOC.