Traders Ask: Risk/Reward Ratio for Setups

AdamAdam Grimes's blogs, Trading Theory11 Comments

Here is another question, this time from reader Mark Ramsey:

Hi guys.

Probably a question for Adam as it’s in reference to his recent post “A little thing that has made a big difference”.

Adam makes comment that he doesn’t like to enter before a level nor during a range, he wants in while a trend is in play. To enter this trend I’m guessing your looking for a pullback (possibly even switching to a smaller time-frame say 1min view while watching the 5min) to find an entry.

This certainly sounds like a recipe for success and a great away to avoid reversals however market trends are usually brief before re-entering another aimless range. Could you please talk a little about your risk to reward ratio preference when you take your trades as I just can’t see you always getting a 5:1 consistently with this method except in the strongest of trends.

For a beginner/intermediate trainer without yet a solid book of profitable plays is entering trades with only a 2:1 or 3:1 reward to risk seem acceptable if the setup looks good?

Hi Mark.  ou hit on some good questions. I confess that I have an important series of blogs that I need to write, but I have been avoiding them because of the mental energy required 🙂 and this topic of risk/reward is definitely on the list. Let me hit the important points here, and I’ll follow up with something more in depth in the near future.

First of all, I think there is a difference between what I mean when I say reward/risk ratio and what some people mean.  If we don’t clarify this at the beginning, we will have confusion.  Some people mean that they think they maybe, possibly, have good potential to make X times what they are risking on the play. For instance, there’s no major resistance in the way, the stock really has the volatility and room to run X times their stop, etc. They don’t actually mean that they will make X times their risk, nor do they intend to hold the stock to that level necessarily, if something happens before that gives them good reason to get out. Another variation of reward/risk ratio is the “ex post”, or “after the fact” version. People will say “Wow, I was risking .05 on that trade and made a full point.  That was a 20 to 1 risk reward ratio!”

When I talk about reward/risk ratio, I mean that if I do the trade 1,000 times my win will be x times my loss, and I also have a good set of realized trades that support that assertion. This is a much more strict and quantitative way to look at risk/reward, but it is the only one that I have found really useful over the years.

Another piece of the puzzle is that there is an undeniable link between your probability of win and your reward/risk ratio. If you have no edge (for instance, are trading in a random market (hmmm… you can see where that set of blog posts was going maybe?) or are entering random trades in actual market, your win ratio will scale with your reward risk ratio so that, over a large set of trades, you are breakeven. If you are entering with 1:1 reward:risk, your winning percentage will be very near 50%. If you increase the ratio to 2:1, your winning percentage will drop to 33%, again leaving you net breakeven over a large sample size. Winning percentage is meaningless by itself. Reward/risk ratio is also meaningless by itself. They are only significant in combination. To sum this up, higher risk/reward ratios in actual trading are always combined with lower winning percentages. When you hear of someone trading with a very high win percentage, their reward/risk ratio is, necessarily, very low.

That’s kind of just scratching the surface. As you can see, there’s a lot more to be said, and I’ll write that future blog series very soon.

To answer your question, I am very comfortable with a 2:1 reward/risk ratio. Truly exceptional trades will over 3:1, but I don’t have any trade setups that offer 5:1 consistently as you suggest. (And if I did, they would also have a low winning percentage of trades.) I’m even comfortable with 1:1 or lower, and traded successfully for years with risk 3 times my wins and very high winning percentages. I think this is a question of nailing down your trading setups and also deciding what kind of trader you are. Do you need a high win ratio psychologically? Nothing wrong with that, but just know that you are going to have to find lower reward:risk ratio trades to find that very high winning ratio.

I’ll follow up on this topic soon. I’m concerned that I confused more than clarified with this answer! Let me know if you have more questions about this.

11 Comments on “Traders Ask: Risk/Reward Ratio for Setups”

  1. Great post! Clean some of my confusions. High win percentage, low reward/risk ratio; Low win percentage, high reward/risk ratio. Maybe it is the essence of the highly competed and often-random market. I confused how a trader define the reward/risk before entering the market, it is so subjective. The more you expect the market go your direction, a high reward/risk ratio, the more factors will interrupt, thus the lower probability the market will reach your target, thus a low win percentage. On the contrary, the win ratio is more objective, often a statistics number.

  2. I’m definitely confused. I thought that trades with the highest risk/reward ratio were the “safest” trades. If these trades have the lowest winning percentage, are these trades recommended for newer traders? Doesn’t Bella talk about trades with a r/r of 1:5 and 60-70% win rate in the training program?

  3. The relationship between R/R ration and W% is expectancy. If expectancy > 0, then system might be profitable over time. If expectancy is <0, then system isn't profitable. From my humble experience, I think (it might not be true for others) that the shorter system's time frame (scalping), the lower Reward/Risk Ratio, and vice versa, the longer system's time frame, the higher RR ratio. Of course, it depends on market structure too. Psychologically, it is very important for a trader to "discover" his/hers RR and %W, and that combination should produce positive expectancy.

  4. great post, thanks Adam. i like kras’ point too.

    one thing i have been doing recently is looking at my median stats rather than mean stats as often my stats are skewed by one or two very profitable trades. obviously there is value in these trades and catching those moves however they do not represent the majority of trades. would be interested on your thoughts about the relative merits of these.

  5. Thanks Adam. I think what I was looking for was confirmation that lowering my expectation for higher R:R when evaluating a trade in favor of a higher probability of a win was acceptable. You summed this up perfectly with some great examples.

    I’ve a had a truly shocking year as the Aussie market has just slogged nowhere and am desperately looking for some wins simply to break the drought and restore confidence. I’ve avoided some good setups because of low R:R but given runs on the board is more important to me (psychologically) than return at the moment I think I’ll aim to drop my R:R and concentrate on my w% best probability setups.

    Thanks so much 🙂

  6. Adam, you hit the nail on the head.

    “When I talk about reward/risk ratio, I mean that if I do the trade 1,000 times my win will be X times my loss, and I also have a good set of realized trades that support that assertion”

    I agree with the following link where Dr. Menaker explains that you do not choose your reward because the market dictates that. Instead you choose your risk and place your trading strategy to take as much out of the trade as possible

    In my opinion, saying I’m going to get 5:1 on this trade is an exercise in futility.

  7. Eric,

    There’s another way to think about this. Imagine you enter a trade and “bracket” the market with a profit target and a stop loss OCO. Sure, you CAN set that up with 1:5 or 1:3 or 3:1 or whatever risk:reward, but you have to realize that your win% will scale with that. I’ll write another post soon digging into this.

  8. I look at both mean and median stats yes. (am a huge believer in nonparametrics in general.)

  9. This is all wrong. Good traders do not think in these terms while Adam is technically correct and provided a most succinct explanation. What he left out is that a good trader doesn’t need to know risk/reward.. this is an illusion. he only needs to know where the market is going. Risk/reward can sometimes be an advanced factor worth considering but for a beginner you should never consider it.

  10. Basically, you need to ask yourself, Do I feel I can be the best in the world at trading? If you don’t you should walk away. Have you ever see a broker who claims you are risking his money but any money you lose is your own? Can you make sense of that? I can’t either. Now, you may ask where does risk/reward factor in? This is an advanced topic and only factors in for gaming other players which is very advanced. Typically it is only valuable during abnormal market conditions when traders may require a higher reward/risk ratio.

  11. Adam, your response makes perfectly good sense. I would like however if you can clarify, in calculating your risk and returns, do you include commission costs. The reason I ask is because I find that these costs can skew your ratio down quite significantly. The best commision rates I can find (I trade in Australia) are of order of 25 bps round turn (on stock). Stocks that provide a greater than 1% “tradeable” move are limited. So if my stop is 50bps below entry, my risk reward becomes 1:1 as opposed to 2:1 excluding commissions. But it is very difficult to find trades with a 50% win rate given these ratios. This leads to the question: at what level of comm does intraday trading become impractical?

Leave a Reply