Well done. You can't compete against HFT without utilizing that approach yourself. And if u do find a way, it is likely illegal, similar to the poor fellow in London who used off the shelf software to pillage the HFT firms with his spoofing. Imagine a trader in 1998 paying $25 a round turn attempting to scalp against the floor - its not going to happen. The access and cost structure makes it impossible. Baron mentions traders vs. investors. Most every short-term trading strategy that I am aware of (in stocks) that has been somewhat durable works by dampening the excesses of "emotional" long term investors - so there is a symbiotic relationship, which is precisely why the strategy types that fit this model have been relatively durable. I'd say the best way is - put away the indicators and figure out something that actually makes sense based on how markets need to function. A principle of trading is u can make money if your counterparty is more time-sensitive and urgent than you are. From that one idea you can find stuff that works, not "competing against hft" or "buying with a stop above this level or indicator" which is doomed.
Everyone talks about HFT and algorithmic trading like they are a same thing... but that is not even remotely true. Most automated algorithmic trading systems are NOT high frequency. Many are actually very low frequency.