>And then there's a fear of "and what if I lose"...

Which is not necessarily remote. In Ticketmaster v. RMG, the smaller company, which extracted non-copyrightable data from Ticketmaster's site, lost because the Court ruled that the momentary existence of a web page in RAM constituted an unauthorized copy. The real-world analog is claiming that an illegal copy is made every time a copyrighted work reflects against your retina. The only explanation for that kind of ruling is simple technological ignorance, but it's what we have to deal with.

While the American system has safeguards in place to prevent it, you do have to ask yourself how badly a particular judge is going to want to offend a massive company that could potentially make him tons of money if they believe he's a reliable ally; if not through explicit corruption and bribery, through the possibility of a post-retirement relationship as a "legal consultant" or similar. What benefit does the judge get by favoring a tiny company that's probably not going to be around for more than a few years anyway v. favoring a megacorp that has practically infinite money to burn? Whatever the deterrents are, we do have to recognize that the natural incentive is to favor the larger company, and remain cognizant of that.

We've seen a minor version of this happen with the Eastern District of Texas, which has become the go-to jurisdiction for patent trolls because they believe that judges that sit there are more sympathetic to their cases than other judges may be. The incentive exists for the circuit to encourage a "specialist" reputation, because they collect more filing fees that way, and the judges get more attention, which is a very valuable currency indeed.

It doesn't solve it entirely, but it helps.

- An attorney doesn't need to be afraid of not getting paid if they're supporting the side that is 'objectively right'. Obviously, most cases aren't crystal clear from the outset, but some are.

- The cost of bullying goes up in the cases where the bully loses.

But you are right, there are also instances where it wouldn't help much. This isn't a silver bullet, but there are good reasons why almost all other modern countries has adapted this principle.

>But does it really solve the problem of bullying?

I think it's a hard problem to reorganize the system to prevent big corporate bullying. I think any potential solution would be pretty sweeping.

One idea would be to require MegaCorp (defined maybe as $100m+ annual revenue) to pay the legal fees of the small defendant (defined maybe as <= $20m annual revenue). MegaCorp would have to bear this cost even if they won, meaning they could only sue little guys if it was really worth it. MegaCorp would pay the same amount they pay their attorneys into a trust managed by the court, which would allow the small defendant to select his own lawyers. This would ensure that both sides have equally-adequate representation.

A second solution is to lower the cost of litigation in general, which would be accomplished by de-formalization of the system and greater local sovereignty. Why does it have to take 10 years and $5 million to have a dispute settled? Companies are already trying to make something like this the de facto legal system by forcing arbitration against their helpless victims; arbitration might not be bad in theory, but under the current practice, it's an unnerving prospect, since you're going into a black hole.

I was thinking it would be effective to have a variation of your first solution. But rather than having special rules for big vs small cases, just have a rule that the instigator has to set aside money for the defendant, equal to it's own court and legal costs. Then, if they win, the defendant has to repay it (although if you are suing a tiny company, they may cease to exist after, leaving no means of collection). So, if MegaCorp sues Little Guy, Inc, and puts 5 mil into it's case, it has to stick another 5 mil into a trust account for Little Guy to use for defense. So, it better be willing to lose that much or be very certain about it's case, and Little Guy has nothing to worry about or any reason to cave unless there's a significant chance of losing, even with equal resources. Seems to me the defender should have an advantage in most cases.

One possibility is to take a mixed strategy approach. Companies facing this situation can buy into a pool, and then flip a weighted coin. If the coin comes up "fight" they get a team of heavy-hitting lawyers who will slug it out against the big corporations team of 100. This will ensure that at least some of the time, the bully will lose several million dollars. Thus, if the bully is given the odds, and motivated by their own bottom line, they may back down. You just need to turn the game theory against them.

Sounds like https://en.wikipedia.org/wiki/Legal_expenses_insurance to me.