Bearing Clearances: Tighter on the mains or rods and why?

I'd like to have a discussion with some others on their view with this topic. I do understand the standard is to go with 0.001" of oil clearance per inch of journal, but i've seen many successful engine builders mix things up and loosen the clearances or go tighter on the rods than mains vice versa. For me I have put together quite a few 4g63/4g64 engines and always used 0.0025"-0.0028" on the mains and 0.0020"-0.0023" on the big ends and used a thicker oil, based off what i've learned from others on here or evolution forums.

Recently I had a discussion with a friend who owns a very reputable 4g63/evo shop in the states about clearances, and on all his 600-1200hp engines he always runs 0.0020" on mains and 0.0025" on the big ends. Now i can understand why he would go with looser rods on a high horsepower engine, the rods take MUCH more load. They take every downward beating as opposed to the crank that just spins and absorbs some of the motion. With tighter mains the rods will see less pressure loss and allow more clearance. I'm debating switching up the clearances I run in my 4g63 engines, and going slightly looser on the rods than the mains. I wanted to know what others think based off their their own knowledge or experiences!

Are you having bearing issues with your current clearances? Given that you're at approximately the same clearances I've run on all our drag engines (1000-1300 hp and 10,500 rpm), I'd venture a guess that you're not. As the old saying goes, if it isn't broken, don't fix it. The potential problem with running tighter clearances on the mains is that you need to have a very stiff crank design to resist the inevitable flex you'll see with high power and rpm. If the crank flexes and you don't have sufficient clearance (and a strong enough oil film), you'll end up with metal to metal contact. This really comes down to the crank design and material it's made from. Talking to Jim from Nitto Performance Engineering in Australia over the weekend, they recently stepped up their Nissan RB crankshaft design as the 4340 material they were using started showing flex at the 2500 hp mark. Moving to EN40B stopped this and the cranks are now proven at the 3000 hp mark.

My point is that there's a little more to consider than just what clearances someone else is running. You need to understand what they're doing in relation to the other components selected. Ultimately you also need to have a good reason to change what you're doing if it's proven to work.