Win Rate vs R:R

The Risk:Reward ratio is not fixed at entry. You have no guarantee of the results for any given trade.

Risk should be defined at entry.

The reward can be fixed at entry, it depends on how you trade.

For any trade, you should have some idea of your expectancy as to what hurdles are ahead of you and where the market you trade should go to.

Otherwise - how do you figure it's a trade worth the risk?
 
No, I'm not saying "no Stop is set when a trade is placed?", but rather that only determines your trade risk (assuming the market doesn't gap through the stop). I am saying a set exit doesn't determine your reward.

Shouldn't your reason for entering a trade define where your stop should lie?
 
Risk should be defined at entry.

The reward can be fixed at entry, it depends on how you trade.

For any trade, you should have some idea of your expectancy as to what hurdles are ahead of you and where the market you trade should go to.

Otherwise - how do you figure it's a trade worth the risk?

Agree with that 100%.

In my opinion each individual trade should be viewed in the context of your overall trading. With the experience of your trading method behind you, you should know how much you need to take from an individual trade (and how much you can afford to lose on it) to keep your account moving along nicely.

Assuming you always make sure you never lose more than you can afford, it follows that you must make sure to take what you need when you can - anything more than that is a "bonus" that you don't really need.

A fixed exit ensures you get what you need , but denies the opportunity of a "bonus". You can manage it to retain that opportunity, of course, if you are prepared to take the risk of the effect on your overall trading business.
 
Watched an interesting vid featuring L rasche the other day. She was talking about a project where they had taken random entries and made them all work with trade management. I can dig it up if anyone interested

Made them all work? Like 100% hit rate?

Look - the fact is that you need to have an opinion on the market and trade it. If that opinion is not expressed in the entry - then it needs to be expressed elsewhere.

You simply cannot avoid making a call in outright directional trading.
 
Watched an interesting vid featuring L rasche the other day. She was talking about a project where they had taken random entries and made them all work with trade management. I can dig it up if anyone interested

Made them all work? Like 100% hit rate?

Just to clear up the confusion, the test was run 100 times.
92-100% of those runs (variance due to different markets for test run)
were profitable in her test:
http://webcache.googleusercontent.c...TradeManagement.ppt+&cd=1&hl=en&ct=clnk&gl=uk

Look - the fact is that you need to have an opinion on the market and trade it. If that opinion is not expressed in the entry - then it needs to be expressed elsewhere.

You simply cannot avoid making a call in outright directional trading.

Agree - especially bold highlight.
That always applies even if the opinion is based on statistical data
to formulate code, its still opinion as there are no guarantees it will
hold up going forwards until its tested.

Just for the record, its not an approach I specifically use.
On another note, some sort of random entry element is best confined
to automated trading.
Personally, for discretionary trading, I think price level reactions
are a much better foundation for entry.
 
Just to clear up the confusion, the test was run 100 times.
92-100% of those runs (variance due to different markets for test run)
were profitable in her test:
http://webcache.googleusercontent.c...TradeManagement.ppt+&cd=1&hl=en&ct=clnk&gl=uk

There's some BS on that page:
"Randomly entering a position can produce a profit 65% of the time"

- sure - if your stop is 3 units and your target is 1 unit, you will win 65% of the time - AND BREAK EVEN.

"This data gives a benchmark for a systematic trader to beat in order ensure the system is successful due to more then capitalizing off noise."

No it doesn't - it's meaningless. Put your stop at 10 units and target at 1 unit and you will with 90% of the time. It's not a benchmark.

"Random entry with Small Target and Large Stop (1ATR/3ATR). The win/loss ratio jumps up to 70%+. While the chances of getting 1 ATR before a 3 ATR stop is hit are greater then 70%, it still is not a profitable system. If we use a target of .5 ATRs and leave the large stop in place, the win % jumps to 82%. The increase in % win should be encouraging for the short term discretionary trader."

Do these fnckwits not understand that win rate is IRRELEVANT? You still break even - well - you pay spreads & commissions. All they are proving is that if your target is closer than your stop, then you will hit the target first more often!

"At this point we know we can randomly enter a trade and get better then an 80% win rate on most markets. By adding a trend filter, random entries can be made into a profitable system."

"Only long entries are taken in an up trend and only short entries are taken in a downtrend. (Again, all entries are at randomly generated points). The Profit Target is 4 ATRs and the Stop Level is set to 3 ATRs (just outside of the “noise”). This ensures that if we can get better then a 50% win rate, the system will be profitable."

OK - so no longer random :rolleyes:

"The system was profitable between 92 – 100% of the time on all markets with the exception of live cows and hogs, (which tend not to trend for extended periods of time), as well as silver and natural gas – (probably due to its extreme spiky nature in recent years). All index futures were profitable 100% of the time over 100 randomly generated runs over 10 years of daily data!"

So they developed a trend following system that was profitable 92-100% of the time?

It's on the internet - must be true.

Do you really buy this LV? Even a trend following system with the stops & targets they describe would never generate 100% winners in live trading.

What utter nonsense.
 
Just for the record, its not an approach I specifically use.
Do you really buy this LV? Even a trend following system with the stops & targets they describe would never generate 100% winners in live trading.

Guess you missed that bit then...
No I think ATR and a whole load of stuff in there is dubious.
MM mentioned it, there was confusion over strike rate,
I cleared that up, nothing more.

BTW, random can mean many things.
It can be time based as mentioned in that link (which I don't buy into),
or directional based, or even random occurrence of mean reversion after
session open which is what I do.
Random is a very broad and simplistic term.
 
...................... Take some free advice and go learn the basics of trading... the lack of understanding regarding basic concepts on this forum is really worrying..............

:LOL::LOL: DT is way, way past the basics so I doubt he'll heed your advice. Perhaps you should think harder about what he has been saying if you want to understand it.
 
I understand what you are trying to say, but actually risk should be defined before entry, and reward is not fixed at entry (an no it doesn;t depend on how you trade). Take some free advice and go learn the basics of trading... the lack of understanding regarding basic concepts on this forum is really worrying...

So let's get this right - no trader can have a trade where he sets a target and just leaves the market to reach it or not?

If they can, then this is fixing the reward at entry and leaving the market to hit it (or not).
 
Of course a trader can set a fixed exit. But again, that is not "fixing the reward at entry" - that is just setting a profit target.

Since people are missing the point I'm trying to make, let me be more clear: the majority of retail participants misunderstand R:R ratios. The idea that one can set an exit point of 30 pips & a stop-loss of 10 pips does not mean you have a 3:1 reward:risk trade set-up. Your reward is not "fixed" at entry (although your desired target point may be). What happens if you run that trade set-up 20 times and every trade fails? How is your reward in any way fixed when you can't guarantee what the market will do next?

I know I'm being a little pedantic in my wording... it is aimed at making a point.

Actually, you weren't being pedantic, you were being downright rude.

If you have a 30 pip target a and a 10 pip stop, then your potential reward on that trade is 30 pips and your potential loss on that trade is 10 pips.

Both are fixed for that trade. You cannot make more than 30 or lose more than 10 on any given trade (ignoring slippage for a minute)

Neither have anything to do with the probability of being stopped out. In fact, on a random entry, the probability of hitting the stop is 3 x more likely than hitting the target.

It is only with the introduction of an edge that you can hope to hit the fixed target enough times to turn a profit.
 
Last edited:
With a random entry system, you need your winners to be bigger than your losers. You should expect to lose 50% of the time.

So a win rate of 50% is expected and yet at entry time there is no reason for your winners to be larger than your losers because your entry is random.

Anyway - say your starting point is a 50% win rate system. Which direction would you prefer to take that in and why?

1 - Improve the win rate and possibly accept a lower R:R
2 - Improve the R:R and possibly accept a lower win rate
3 - Look to both improve the R:R AND win rate

Obviously number 3 sounds like the best option but the "why" should really explain why you feel it's feasible.

As an aside, it might also be worth considering how your choice would impact the number of opportunities that would occur.

Just realised I never actually answered this.

None of them TBH.
If you want efficiency, a random or semi random entry is simply the wrong place to start.

So I would say 4 - go back to square 1.
One possible exception I can think of would be random entry at
session start with something like reverse at stop, so subsequent entries
are controlled by price.
Tight BE, stepped trails and widish target / time window / max loss to cancel,
or whatever else you prefer.
Either way, it will be very hard if not impossible to improve strike and av. RR together.
One or the other is the best you can hope for taking this approach.
Thats assuming you wanted to stay in the systematic domain.

If not, level reactions and so on, are more efficient than random entry.
Thats the whole point, inefficiency goes with the turf with a systematic
approach, its largely unavoidable due to the fact its largely impossible
for code to interpret level reactions, tape and volume, at the retail end at least.
The bonus is the ability to more rigorously test it, which of course is absolutely
no use if you can't accept the inefficiency that goes hand in hand with such
an approach.

That simple fact is the reason why most people attempting a systematic
approach fail.
They attempt to alleviate that inefficiency, more often than not by short
term curve fitting, end result isn't robust.

Discretionary trading is like fly fishing, you need to master it, not everyone
can or even wants to.
Done well, it is a more efficient and targeted approach, although more labour intensive.
A systematic approach on the other hand is like a trawler dragnet.
It is impossible to be efficient.

Thats the real choice.
Strike rate and average RR are a by product of that choice.
You can't have both.

That choice is largely out of your hands - its based on availability,
time and the willingness / ability to actively screen watch or not.
 
Just realised I never actually answered this.

None of them TBH.
If you want efficiency, a random or semi random entry is simply the wrong place to start.

The question is simply -

"Anyway - say your starting point is a 50% win rate system. Which direction would you prefer to take that in and why?"

This does not mean the starting point is a random entry system. Just a system that is winning half the time.

Talk about that stick again...
 
The question is simply -

"Anyway - say your starting point is a 50% win rate system. Which direction would you prefer to take that in and why?"

This does not mean the starting point is a random entry system. Just a system that is winning half the time.

Talk about that stick again...

OK my turn to get the wrong end of the stick :LOL:
Obviously 3 in that instance.

If that wasn't doable, logic says 2, psychology says 1.
Not that there is necessarily anything wrong with going for
increased strike at the expense of RR, pretty much
all the work I've done on that tends to suggest that favouring strike
over RR will reduce potential profit.
Then again drawdown is likely to become more of an issue if you choose to improve RR.
But if it makes it easier to trade on a practical and personal level,
then that is a perfectly valid reason to choose improved strike.

So 3,2,1 in order of preference.
 
increased strike at the expense of RR, pretty much
all the work I've done on that tends to suggest that favouring strike
over RR will reduce potential profit.

Have to agree with that.

The thing is - an ideal trade for me is where:
- I get in and I can see that a price 1-4 ticks away should hold.
- There is s good chance of a reaction and if the reaction does occur, it should make a decent more (more than 4 ticks)

That's a perfect trade for me - I think there's going to be a reaction, if there is I make more money than I lose if there isn't.

I don't think a really high hit rate is at all important.
 
Have to agree with that.

The thing is - an ideal trade for me is where:
- I get in and I can see that a price 1-4 ticks away should hold.
- There is s good chance of a reaction and if the reaction does occur, it should make a decent more (more than 4 ticks)

That's a perfect trade for me - I think there's going to be a reaction, if there is I make more money than I lose if there isn't.

I don't think a really high hit rate is at all important.

Agree.
One instance where strike and RR stats can get skewed is with
small profit scratch trades.
Obviously strike climbs and RR falls off.

I spose that is one situation which will result in a higher strike
and lower RR, despite the less frequent and larger MFE / MAE trades
where BE + tick or 2 doesn't kick in.
 
IMO with 50% success rate, reward:risk is all I need to improve.

If I have a 5:1 reward:risk and 50% success rate I am happy. Even a 2:1 Reward to risk and 50% probability is attractive.
 
Top