Interesting. How did you come to that conclusion? By that I mean, on what information are you basing your assertion that past performance is a good gauge, and that it gets better with time?
Think of it this way. Let's say you're an employer looking to hire an employee. You have an application and references. You call the references and they say this person has not been very good at their job, poor performance, etc. You interview them, they say, "Look...for the past 10 years, yes I've been a bad employee. I've not done a good job. But when I work for you, trust me, I'll work harder, be better!
You interview another employee who has a history of success and doing well, working hard. His references say the same.
In both cases you are not absolutely guaranteed success from either candidate, but which one will you choose?
Great example! You're very close to being spot-on here, you just need to shift your perspective a bit.
Instead of thinking of the funds as the employees, think of them as Indexing vs Non-Indexing. Ok, let's try it, and I'll fill in some numbers:
-------------------
Let's say you're an employer looking to hire an employee for a very important job. This job is so important, that if you have a bad year, it's almost impossible to catch up to your competition. In fact, if the employee performs
half as well one year, they'd need to do
twice as well the next year just to break even with your competitors. So it's very important to not pick anyone who might have a bad year.
You have an application and references. You call the references and they say this person has not been very good at their job, poor performance, etc. You interview them, they say:
"Look...
for the past 100+ years (every year on record), yes I've been a bad employee.
There's less than a 1% chance I'll be successful for you over the long run. In any particular year, I have a 37% chance of being successful, but since it's so hard to catchup, the longer I work for you, the less of a chance I have at being successful overall. Even worse, it's almost impossible to predict
when I'm going to do a good job. Most people think I'm more likely to succeed after I've had a successful year, but I'm actually
less likely to succeed at those times. I've not done a good job. But when I work for you, trust me, I'll work harder, be better!
You interview another employee who has a history of success and doing well, working hard. His references say the same.
He is also mathematically guaranteed to not have a bad year. Every year, he is guaranteed to beat or match at least half of his competition.
In both cases you are not absolutely guaranteed success from either candidate, but which one will you choose?
-------------------