No experience trying to chase them down on the deck in an A6M2 I see.
No, I have done that as well occasionally. And then there is the great deal of my experience being in TBM's that are chased down by A6M2's. I don't think that you have flown TBM's in Coral Sea scenarios, though. Nevertheless, my argument isn't that TBM's are fast as judged by A6M's. See below for a very precise explanation of what I am saying.
A later war sub for early and mid; From that data, A stacked deck, a loaded die, and Pure bloody mindedness not the see sense, logic and reason.
I have a good grasp of sense, logic, and reason and a long education in and then a professional background in science and analysis. There are lots of occasions (in physics, biology, chemistry, engineering, economics, social policy, finance, etc.) where people have an opinion: "X is very important. If you have X, clearly it changes things." This is based on the person's thoughts, but to see if it really is true, you use a couple of tools on actual data to verify whether those thoughts are correct or mistaken. The tools are statistical analysis and especially regression analysis. It is a fancy name for a straightforward, simple-to-understand technique. Regression analysis is used in situations where an effect can be due to lots of different variables, including the one that you are interested in. In other words, you are interested in finding out if X is really important or not, but there are also variables Y and Z that matter. So what you do is that you look at past data and compare all of the cases you can find where Y and Z are the same (or as similar as you can get given the data) and the only (or then main) difference is that X is changing. Then, if there is a statistically significant difference in result between when X is one value vs. another value, you know that X is significant. If there is no statistically significant difference, the changes to X are not significant.
What I have shown you above is a quick version of regression analysis on past data, compared results when X = TBM and when X = B5N, and found that the result does not vary significantly.
Thus, the feeling that TBM's are a significant advantage is not born out by the data.
The one caveat to the analysis is that, if the A6M2 and D3A are significantly more capable than the F4F and SBD, it could mask the B5N being worse in effect than the TBM. This is because I have no data of events where both sides have the same fighters and divebombers and only the torpedo bombers are being varied.
Note in the above that I am definitely *not* saying that TBM's are harder to catch under some circumstances (namely going full speed) than TBD's or B5N's. What I am saying is that the data says that it doesn't matter. This might be because most torpedo bombers are lost on their run to target, on the deck, toward a known target, when they are going ~200 mph no matter what their top speed is, for a substantial distance, while the pilot has to focus on his flying and can't man his guns. But the explanation doesn't matter -- the data shows what it shows.
When data shows that a result goes contrary to your thoughts on the matter, sense, logic, and reason demand that you update your thoughts.