Add this argument to ./configure --disable-tests

I'm not sure what you're taking issue with, here.  Given a long enough time mining, you'll have actually a greater chance of mining coins.  With pools you would need to contribute a share greater than the pool difficulty.  That's very unlikely if you're hashing 200 vs a pool of 2000s.  Luck is fair.  Diluting hashes to give fractions of coins while a pool operator takes commission, is not. 

The lesser of the two evils here results in A.) You receiving more coins.  B.) No single operator being able to launch a 51% attack and you potentially suffer a loss in an outright theft of coins.

Not sure what happened there.  It's not entirely improbable that you would mine 3 or 4 blocks in a row.  What I believe would be a somewhat likely explanation: you found a block that was not the best solution (this happens frequently with alternate chains on any blockchain), but was still valid. It follows that there would be a succession of blocks thereafter that were correct solutions but not optimal ones.  When that occurs, the other miners can be mining on a separate chain from you, until the nodes reach a consensus as to the best proof submitted for work completed (blocks solved). The longer chain prevails, and multiple chains can exist simultaneously before one chain shows itself to be longer, and the network agrees.

This is the exact reason that I increased the # of blocks before a mined reward can be spent.  If it were lower than 25... there would be issues.  I've been contemplating raising it even further as a buffer. Hope that all helps.

Anyone already tried this? I checked the repo and there is no commits for the balloon algo. So Ill wait first before trying this out.