To me, if you wanted to hack the system, you would look at creative approaches to make the market behave in ways that it is not supposed to.
An example that I'm working on is the idea of an alternate currency where the deposits are invested in socially responsible / clean energy companies etc. People would trade their currency via NFC wallets or square equivalents etc. for goods and services. Meanwhile the investment stays with the same company.
This to me hacks the trading system because it uses private contracts & a virtual currency to shift the choice about how money is invested to the consumer & opens up opportunities to make longer sighted investments (that benefit the economy as a whole & the people using this currency) rather than just trying to maximize the profits of an investment fund.
From a "computer hacking" perspective, this is just a mash up of different platforms that have recently developed & using it to create the previously only dreamed of "indexed currencies" etc.
Every HFT outfit has someone on the lookout for these kind of ideas. When this system is found, every idea in a commit in an obscure branch is going to be quickly reviewed, tested inside that outfit's backtest system (which likely has years of microsecond stamped tick data to test on, and a 1000-core strong compute cluster to do that quickly), and if it has merit -- applied before you have a chance to test it in trading yourself.
Your own success in trading this system is not required (and in fact, unlikely, even if you implement things that would work if done in a non-public way).
I'm not trying to get you not to do it - on the contrary, I love to follow interesting ideas in the field even though I'm not in it anymore. I'm just trying to give my sober, experienced, point of view - if you want it to succeed, this particular thing has to be closed source - you should only commit to public git after your edge has waned....
