Personally, I'd run a mile from anyone with an 85-90% win-rate offering mentoring services, because I'd expect mentoring services (rather than trading) to be his primary source of income.
Win-rate doesn't matter. Expectancy matters: you need reliably to win more, collectively, from your winning trades than you lose, collectively, from your losing trades.
Methods that can (for a while) produce win-rates of 90% are typically unsound and will eventually crash and burn.
Introductory trading textbooks (like Van Tharp's Trade Your Way to Financial Freedom) explain in more detail why it's generally easier and better and safer to make decent profits from methods with win-rates below 50% than above 50%. I don't suggest that you can't do very well with a method that happens to have a win-rate around 60% or so, but if you're looking to learn from someone with a win-rate of 90%, you're really focusing on the wrong parameter, there.
The following 3 users say Thank You to Tymbeline for this post:
I can highly recommend Al Brooks' video trading course. $250 for a whole 48 hours and hundreds of slides, with a massive update coming early 2016. Get it and get studying! Learn to read the price action bar by bar and you will begin to understand what is happening on the chart in front of you.
Re. win %, as Big Mike pointed out it's about more than that, it's also about reward/risk.
Scalpers go for a R/R roughly equal to 1, and try and keep their win % high i.e. >60%. (very difficult)
Swing traders go for a R/R >= 2 and try and achieve a win % of 50% or more.
The second one seems to be the best, if you think about it if you risk 10 ticks to make 20 every time, you only need to win 40% of the time to become profitable....
I asked Hoag from TST this before and he said, most of their successful traders are using a R/R of 2-3 or more. He also has stated on numerous occasions that, if you can have a R/R of 2 and win just half of the time, you stand to make a lot of money!
Not successful with futures yet by any means, but:
Many trend following systems have 35% wins and make money. ie, big wins on huge trend break-outs pay for the much more common failed break-out attempts.
Helpful for me was to get a tradestation account and run through some back-testing of strategies just to get an idea.
You can add lots of filters to increase your win %. Eventually you get very few trades but they do win. Can you scale this enough to actually trade for a living. Probably not, simulations assume you get filled with whatever size you like which does not happen in real life.