RSI vs Stochastic

I`ve been demoing for quite a time now and [B]Stochastic[/B] indicator is basically my guardian angel.I guess it works 7 times out of 10 and false signal can be also evaded with trend analysis sometimes.I find it very useful in my strategy.

So ive heard of the [B]RSI [/B]indicator before but could really not get it so i wanted to test it, and basically ive found that its almost the same except there is only 1 line and the levels are 30 and 70 instead of 20 and 80.I don’t know if that smoothens out fakeouts,but i guess it also slows down reversal signals.

I want to hear opinions about the 2 indicators,i`m really confused since they are almost the same.Which one is better or more accurate and which one is [B]lagging behind slowing down signals but prevent fake ones [/B] vs [B]leading the signal and causing more fakeout/false signal[/B]?

1 Like

Have previously traded with / observed RSI with MACD, but couldn’t see all that much signal accuracy in predicting movements; many of the market interpretations of these older tech. analysis indicators are normally only confirmed once FX price data is available.

1 Like

Don’t panic!

Stochastic & RSI are both oscillators, so they’re doing the same job albeit slightly differently. Choosing the right one simply comes down to which one confirms / identifies your trade entries and exits most reliably.

Try both. Experiment with the settings so that the peaks and troughs most accurately coincide with peaks and troughs in the price action. Ideally you want these peaks to hit the overbought / oversold zones. The optimum settings can change depending on the currency and timeframe.

I use both but prefer Stochastic and find that 8,3,3 is a good starting point for settings (at least to start optimising from). If you want the best of both worlds you can get a Stochastic & RSI combined indicator calld DTOsc (DT Oscillator), which I really like too.

+1. I have used both. I like and use stochastic. The 8,3,3, setting is great for scalping/day trading.

1 Like

I find the stochastic more visual especially when spotting divergences. RSI…not quite

Looks like this thread is going in the favor of Stochastic.So what about the 15,3,3 settings? Its a bit more smooth but not that much fakeouts,besides a trend lasts longer usually than 8 candles,so what about those settings?

Depends on the instrument and timeframe. On your chart, draw horizontal lines where you have highs and lows, then adjust your Stoch settings to get the swings as close as possible, with the peaks being in the OB/OS zones. Stochastic will always lag by a bar or two, but you should be able to get it so that the when the fast line crosses the slow in the OB/OS zone it becomes a reliable reversal confirmation.

Just had a look at 15,3,3 on my EURUSD 15m chart (which was open) and the result was horrible!

Yea maybe on that low timeframe you need the 8,3,3 setting.But anyway lower timeframe = more uncertainty,bigger spikes and huge price movements.I would probably set up the stoch on a 1 hour chart and from there look upon the 15 min if that would be the target.On the other hand a 15 min chart may indicate trend reversal signs for the 1h chart,so Its more important to look on more than 1 timeframe.

I agree - my trading requires correlation across three timeframes, either D, 4H and 1H or 4H, 1H and M15. The criteria that has to be met is very specific and the result is 15m is usually reliable enough to make my first target (where I take half my position off).

To me, Stochastic reacts faster to price changes that the RSI (which could sometimes result in fakeouts), but the RSI is more accurate when it comes to the overbought and oversold conditions. The RSI is also used to show a bullish bias and a bearish bias, i.e. when the price is above or below the level 50. Therefore, it looks like the RSI is more favored than the Stochastic.

Neither stochastics nor RSI reacts to [I][U]price[/U][/I]-changes at all. They’re indicators of [B]momentum[/B], [U]not[/U] of [I]price[/I].

That may be,but if you look at pure price action and you identify a trend reversal to be happening, and that was not confirmed by an oscillator,it has a greater chance of not happening,since other traders will look at an oscillator and decide not to trade it.So in other words a pure price action might be the way of the market but not the way of the traders who are in that market,and if the majority of the traders decide something thats contrary to price action expectations,no matter how well a price action indicates an event,it will not happen if the big traders don’t want it to.
In other words: stochastic and RSI is not just a momentum indicator,but because of their popularity its also a market sentiment indicator (especially if the oscillator is in the overbought/oversold zone for a long time,a reversal is inevitable)

Yes - I won’t argue with that. :slight_smile:

I don’t quite agree with your inference, in this part: I don’t think the “inevitability” is in itself useful for trading purposes. Momentum and price are like the acceleration and speed of a car. If its acceleration is diminishing, however slowly, then it’s “inevitable” that at some point it will stop altogether, but it’s entirely unpredictable when. I’m biased and outspoken on this subject, I know: I think “overbought” and “oversold” are useless and misleading concepts, in this context. Prices can continue rising (albeit with less acceleration) for ages after momentum oscillators are telling you that they’re “overbought”. It’s not helpful. To me, anyway. :slight_smile:

Of course it doesn’t tell when will the reversal happen,that would make it 100% accurate,which is not quite true.If it would be then I would be now drinking martini on my 16m dollar yacht in the Mediterranean Sea :slight_smile:
What it does it signals a warning,that something is stinky here,and you should pay attention to further events,and use other tools to confirm a signal not only stochastic,stochastic is morelike a warning sign not a confirmation sign; used properly with other indicators or price action analysis,it works like magic.
The trend may continue for ages,or it would reverse,and also take in consideration the timeframes…A normal retracement in a trend is a reversal on a smaller timeframe,so using an oscillator in multiple timeframes is even better.In my experience
i would say that it has a 65-70% accuracy,and it works mostly for me on 15-30 min charts,when there are no fundamental
factors like upcoming news/announements disrupting the price action.

That would be more than enough to make a living just from that on its own: very few of us can trade with 65-70% accuracy - it’s an enormous strike-rate! :slight_smile:

Oscillators are objective,besides market conditions are changing,what works today may not work tomorrow.Ok maybe 70% is exaggerated,it’s more like 60%,but still it works.The problem is mostly with the trader who has emotions not with the indicator,for example 60% doesn’t mean that 6 times out of 10 you win guaranteed,it needs a much larger set of data.It’s pure mathematical statistics and the phenomena is called “standard deviation” (sorry for nerdness :D). Like maybe you win 60.000 trades out of 100.000, but you can start even with 100 losing trades in a row and just after that you start to win.And traders usually after 5-6 losing trades get panicked and abandon their system(otherwise good system)
just because the odds worked against them.I don’t say that 5-6 losing trades doesn’t need an evaluation of your strategy but mostly the problem is with the individual trader ,aka those nasty fear & greed symptom appear,oversized lot,random trades,addiction,etc.And at the end of the day they would have a nice 15-20% growth/month theoretically,and practically they have now a blown up account because they could not discipline themselves…

Yes … you make many good points, and you make them well. I appreciate the discussion with you, anyway, even if we don’t agree about the underlying issue (or “even if we have differing experiences of it”, I should perhaps say). :slight_smile: