A855 - 5 Speed

Dave, I'd disagree with you that "torque rateing has little to do with real life". It has everything to do with real life. Ratings are used to determine the suitability of something for a particular use for virtually anything civilization uses. That's where you start. The only time this doesn't work is when the actual service conditions are different than the design service conditions. Then there usually will be trouble.

thats a very fair point
which illustrated my poor way of saying

300ftlb or 200 ftlb rateing doesn't mean you can't use it if you have a 400 ftlb motor
in the real world the rateing of the box is potentially lower than max torque of the motor i drive an illustration of this very point. but lots get hung up on making sure torque rateing is greater
if you have money to spend you will have one robust set up and you probably won't need to spend again

but you aint going to expose your trans to an instantaneous shock load of 400 ftlb or indeed run it constantly at 400ftlb ever or at least not very often. You'd need an isntantaneous engine stall caused by the wheels being locked solid at whatever RPM is peak torque.

if the wheels are moving you pass through peak torque very quicky, rpm will move up to peak BHP and then it all tails off.

rateing has real world use

but is derived from parts spec and testing theory and calculation for a purpose that is focused on longevity in standard applications, are we concerned about lengevity in the same way.....probably not

distructive testing.. well they used to with engines...relaibilty runs on dyno...run it at 4000 rpm against the dyno for 24 or 48 hours. see what broke and make it stronger. then try 5000 then try varied race track or real world rpms and load.
in theory you could simulate a le mans 24 hour or similar and see what impact it had on your new motor design....
Chrysler Australia learned how to do this from their colleagues in the US as part of the LA and D (aussie six) motor programme how Light an A engine can we make and how come we keep twisting engine blocks on the D (I6 245). there must have been some twisted and cracked engine blocks by the end of the program. vey much run it till it goes bang then re-tool a section of the block casting. which is why we are not running around with early hemi and poly weight blocks and heads, easier prototyping with furan core casting...make me a block with thicker main supports etc etc

id guess you could do it with a transmission as well

Dave