Is it possible to get 70-80% accuracy?

[Tonievanu]

I am starting to think that people claiming to get a high accuracy in trading is lying. Testing different strategies always worked well at some point and then the market changes and it doesn't work anymore. Is there a profitable strategy that gets at least 80% accuracy? Is that even possible?

 

[Nick]

There are strategies which can produce very high probabilities of success and accuracy is important. However, the size of the average win vs. the size of the average loss (2 to 1, 3 to 1, etc.) is the key to making money in the market. If the gain/loss ratio is high, you can do very well with low win rates (> 40%). Good trading!

 

[zeek]

Like Nick said above, it’s more about the win/loss ratio rather than the winrate. But to answer your question, yes it is possible to have 80% accuracy. Requires extremely good discipline.

 

[Alex]

if your strategy is based on high winrate you would almost always be sacrificing your risk to reward and vice versa. Personally I prefer a higher winrate (55% - 65%) as I find it to be easier on the mental side and a smoother equity curve. In the past I tried trading a low winrate (25% - 35%) high reward strategy and the equity curve swings would have a big impact on my emotions.

 

[DeepThinker]

Per my experience, the level of accuracy can be improved by employing a helper bot.
Also, I agree with earlier comment about experience and discipline.
There is no magic wand. Find your targets, trading style and comfort zone.
Paper-trade, Paper-trade, Paper-trade
Best of luck!

 

[GiantBull]

When testing a strategy, I want to aim for a win % greater than 60%. So, 6 out of 10 trades will be profitable. If I could for example, make $1,000 6 times and lose $300-$400 4 times I would say that is a solid trading system. Ideally a win % should be greater 70% but honestly that seems to be a little hard when day trading/scalping. What win % makes you comfortable to implement in real time the strategy you are testing?

So if you managed your stop losses well and had a 70% loss rate with 7 losses of $100 and 3 gains of $1000, you would call that a bad strategy to make $2300/day?

Considering commissions I would say that isn’t a strong strategy. Just reading trading books and doing research online the ideal % seems to be 70%. Imagine taking those 7 losses back to back and being down $700 on a small account. Now cut that loss number by half then it is reasonable to implement that strategy since taking 3 or 4 losses continuously won’t affect your account by a large degree. I’m curious about what win % you currently have for your strategy? I see that you are very active in this community and assume you also trade.

 

[MerryDay]

I don't pay commissions. In bullish times, I was a swing trader. I bought dips in a small corral of high growth stocks, and the losses were minimal.
I almost never hit my stop loss.

In the current market, with my strategies:
https://usethinkscript.com/threads/mean-reversion-strategy-study.13448/#post-112841
I get stopped out fairly often, even on big bear rally days. Many stocks make double bottoms, sometimes triple which stop me out several times before I get into the big rally.

In this bear market, it is more about choosing the best stocks to trade on the best bear rally days, and using price action to time your entries and exits. As long as I pay the rent at the end of the day. I don't pay any attention to what number of losing trades I had.

That is why I was curious why you would focus on that percentage to determine a good or bad strategy, if you are making a good profit.

ToS Backtesting is helpful to fine-tune your entry-exit logic of your strategy.
But if you are using it to create your win/loss percentage, do you realize that requires 10-15 years of data. With a broad testing field of instruments?
https://blog.quantinsti.com/backtesting/#:~:text=What should be the time period for backtesting a trading strategy?

The generally accepted "financial" standard for determining profitability is:

Average Profitability Per Trade (APPT)⁵, which is actually the cornerstone to determine if a strategy is profitable or not.

https://towardsdatascience.com/backtesting-trading-strategies-less-sorcery-and-more-statistics-on-your-side-241ac41d18b0#:~:text=Average Profitability Per Trade (APPT)⁵, which is actually the cornerstone to determine if a strategy is profitable or not.

At the end of the day, whatever you consider to be a successful strategy and if it works for you, then you can set whatever parameters on it that you want.