Yet another flaw in the indicator…it has no measure as it relates to the number of contracts traded and has no relation to total risk as it relates to account size.
A trade that draws down $1000 on a single contract and then exits for $200 profit is much different then a trade that draws down $1000 on 10 contracts and then exits for $200 profit.
A trade that draws down $1000 on a single contract and then exits for $200 profit in a $200,000 account is also much different in terms of risk than a trade that draws down $1000 and the exits for $200 profit on a $50,000 account.
The Hold and Hope indicator is useless and completely flawed in its ability to measure the performance of a system.
SixSigma Trading
Let me be simple. I’m simple uneducated guy.
Who cares? I gave an example ~2K of indicators and there aren’t Holy Grail. I know the indicators lags and problems. Well, not all of the 2K of course, but somehow I can use rest that I know lol I don’t see anything insulting in the “Ross indicator”. Again, who cares? It might be useful or not. It’s only new indicator. Why to not appreciate his creative approach? 
From my point of view it’s one of the ways to estimate a system. It’s not perfect, but it’s not very bad to be insulated. The method has some realistic base (again from my point of view).
Why to not use it?
Eu
I agree that the indicator leaves much to be desired for. Introducing an indicator like this, C2 stops reporting generally accepted statistics objectively, but cross the line into inserting it’s own bias with regards to a system and start giving them labels based on the preferences from one person. Systems get labeled and rated now based on arbitrary thresholds set by one person based on his own preferences.
We are not talking about something like Sharp ratio which is well accepted through the industry. We are talking about an indicator made up by own person, introducing his own bias and preference on how a system should be rated.
I would rather see that C2 continues to present just the stats. I think a much fairer alternative would be to report just the average drawdown per trade. This can be reported in the same box where Avg Win and Avg Loss are reported. Just add another line and call it Avg DD/trade. This is also know as Maximum Adverse Excursion (MAE). This would be just objectively reporting the stats without introducing any kind of bias. That way potential subscribers can see the average drawdown they can expect per trade and decide for themselves what they would be comfortable with, without Ross’ arbitrary thresholds deciding when a system is any good or not.
I find it very disappointing that C2 allows one individual to dictates how systems should be labeled, based on his own preferences.
Regards
- Fanus
Even the average drawdown would be biased though. One would have the note the average drawdown per contract on a trade when there is more than 1 contract involved and then relate that as a measure of percentage drawdown on the account as a whole to measure the degree of risk.
Nevertheless, it still completely ignores the basis of hedging and trading two opposing contracts simultanously but it would be a much more accurate measure of risk on a per trade basis if the above is taken into consideration as well.
Hmmm…if this uh…“indicator” INDICATES a system as being poor, risky, useless, etc. while the system is consistently pulling month-month profits while mantaining 10% or less max drawdown, it’s useless as far as I’m concerned…we shall see!
There’s a lot of truth in the criticisms of the drawbacks of the “Hold and Hope” indicator even while some of the critics have been quite a bit more vociferous than I have.
I think the description should be changed to indicate what it really is… as Ross describes above… a relationship between net profit and drawdown. Not an indicator that tells you a system is holding longer than “it should” or that a low score is a result of nefarious trading manipulation. As I stated, and as SixSigma pointed out even more strongly… the indicator can be misrepresentative of what’s going on in a system. It may be spot on for many… but it can be way off base for others.
Still… I think the statistic itself has some use as it does give one pause to consider more deeply how a system is trading and what kind of drawdowns one might expect. Nothing wrong with that, I think.
An indicator may mean different things to different people. I know some who swear by “Volume” indicators… and their usefulness… but a Forex trader would find it completely useless. … Hold and Hope might need some changes to description of its stated purpose and possibly to its name… But I still think the raw statistic can have some value to C2 users…
Fanus S & SixSigma Trading
I don’t understand your complains. If I want to estimate a system I’ll use some industry proven standards + something from me. I’ll never use something hmm… “creative” from others if I don’t trust it.
From other side, who cares about new indicator? Gosh. In zillions of indicators “Ross indicator” became the final true. lol
Guys, what is the problem? The indicator isn’t perfect, it’s averaging, it’s doesn’t have time frame in base. It’s just next attempt of prediction of a system.
Who cares about all the attempts of new scientist ? lol
Eu
I fail to see why this measure has all these flaws. Instead of being insulted and feeling unfair, and giving whines and complaints, address the underlying issues.
The higher total drawdown a system has, the higher the risk and more problems it has. Systems that average down or hang on until trades turn positive, have significantly higher drawdown, than those that don’t. These kinds of systems are rampant here.
Unless you clearly show why this is wrong, then I see little validity to the complaints. This was not directed at anyone, and a simple comparison of net profit against max DDs is hardly one person’s directives.
Every measurement can have its detractors. But facts are a lot more useful than rants or opinions.
As a “vendor” (not really, but technically) , this measure has no effect on me. I couldn’t care less.
As a subscriber, I love the concept but the implementation is lacking. I reviewed the new metric as it relates to several of the systems I follow regularly. As it stands, it is not working as intended. Perhaps it should be flagged as “This metric is in beta” or some such until it is generally agreed it is working as intended.
Jules, much like the “P/L Unit” issue, it’s all messed up by trades that are scaled in and out of (possibly several times), and for the same reasons. There’s no real comprehensive solution to that, but I’d love to see the new metric stripped down to a per-unit value, as suggested above. Perhaps that will satisfy the accuracy I’m looking for, I’d have to see the effect of those recalculations on the systems I follow.
I’m also concerned by “Cheetah’s” comments recently about drawdown figures being incorrect. I’ve never looked at them, and don’t know if they are working properly or not.
I certainly don’t agree with the “hedging” argument presented above; it is a mathematical logical fallacy. “Flat” is a position. If you want to “hedge”, take a flat position. It costs less in terms of commissions and execution, but doesn’t look as pretty in statistical terms. “Hedging” is only useful when one can’t effectively exit the original position due to tax issues or position size, or perhaps in some options combinations.
You can’t call taking opposing positions in related commodities “hedging”. That is what is known as a “spread”, and even then, the term only applies of you take both positions simultaneously.
In any case, the metric in question has no need to consider such a thing.
The problem is that you display an unproven “indicator” together with well established measurements. That creates the impression that this measurement is just as important than the others, when in reality it is just based on one person’s preferences.
Regards
- Fanus
I have clearly stated where the flaws of the indicator are and why it is misrepresentative as a whole. What did you not understand?
Unless you take into consideration total contracts per trade as well as give the drawdown per contract a representation of percentage to the account, the statistic by itself is completely meaningless to measure risk. There is nothing to relate the number to and that is a very valid argument to prove it is useless.
Hedging is a very common practice used to reduce risk.
http://en.wikipedia.org/wiki/Hedging
If I take a trade, but then later see an opportunity to make a profit in the opposing direction while the maintaining the original exposure, then I would hedge. I could also hedge if the bigger picture maintains my bias but the short term looks unclear. It’s all about reducing risk for the short term. Going long one index and going short another that is similiar is exactly what hedging is. They do not need to be opened simultaneously, but when the second is taken, the first is then ‘hedged’.
If you take a hedge and close it succesfully, you incur charges for the trade. If you close the original only to get back in later, you still incur the same trade charge. There is no difference in cost. However, if I take a succesful hedge, I have avoided the additional drawdown which is a profit and account growth benefit. It is not for everyone, but I trade multiple time frames. When I when I want to keep exposure on one side for a longer period, but I also see a short term trade in the opposite direction for a profit opportunity, a hedge is a nice trade.
In addition, taking a long on an index and then taking a short reduces risk because the margin is effectively reduced to the difference between the two and is not additive. If it is, you need another broker.
If each long and short trade were taken individually at different times, then there would not be a problem. However, the Hold and Hope does not distinguish situations where a drawdown is effectively negated by a hedge trade.
Ross,
I didn’t rant or give you an opinion. I just proved with a very simple and straightforward example that many situations are possible in which a system with a low H&H has both lower overall risk and higher profit than a system with high H&H. I’m not saying the indicator is useless, but it should be understood and interpreted in the context of other system parameters.
I strongly believe that the indicator by itself is not sufficient to classify a system as “high risk”, “averaging down”, “holding on to losing trades”, or “boosting winning-trade statistics”.
Unless you clearly show why this is wrong, then I see little validity to the complaints.
I cannot disagree more. If something new is introduced, the onus is on the person introducing it to show why this adds value and why specific thresholds are right. One cannot just come up with some arbitrary thresholds and say "show me wrong".
If this is your argument then I can say that values below 0.05 is High Risk, between 0.1 and 0.3 is tradable, above 0.3 is good and above 0.5 outstanding. My thresholds are right and yours are wrong. If you don't agree, then clearly show why my thresholds are wrong.
Here is an example of one system:
Profit DD
-100 -100
200 -90
-100 -100
-100 -100
-100 -100
200 -90
-100 -100
200 -90
-100 -100
200 -90
Total Profit = 200
Total DD = 960
HH = 0.21
Classifiction: high risk
You have a system with average wins twice that of average loss and 40% accuracy which is well in the realms of a longer term trend following systems, which typically have many small losses and fewer big wins. Many people are comfortable trading systems like that. But since someone decided that HH below 0.30 is bad, the system is flagged as such by C2 as a system which are holding on to its trades for too long.
This indicator is very narrow minded and assume all systems manage risk on a trade by trade basis and ignore the fact that other systems manage risk on a portfolio based level.
Sam's suggestion of Return divided by Max Drawdown will do a much better job of reporting risk/return for the whole system instead of boxing down systems and assume all are the same.
This “indicator” does not measure total system drawdown. It measures the intra-trade drawdown, which has no value in making a judgement about overall system performance or tradability.
To give you a specific example - since you are looking for one - a system that trades volatile equities may have 5-10 positions open at once, serving to avoid risk and smooth the equity curve. However, by your measure, this system could be judged to be “risky” if each of the positions, while open, at different and various times, each experienced weakness before closing at a profit. However, it is often the case that while one position is experiencing weakness, others are experiencing strength. YOUR indicator makes absolutely no sense in these cases, and serves no purpose but to penalize systems that use such an approach.
I believe your posts serve only to display the fact that you don’t understand the trading methodology behind such systems, which can be low-risk and profitable.
Further, I don’t think that anyone here needs your personal admonitions about the value of holding positions for whatever reason the trader desires. If those are your opinions, you are fee to keep them to yourself. But I repeat again, to impose those on others in the form of a public “rating” is simply insulting, and a real negative for this otherwise excellent site.
If you want to publish a judgment about systems that are here to take traders for a ride beyond their expectations, then you need to find another way to do it.
“averaging down” and “scaling in” are two entirely different concepts, but I fail to see how this “indicator” can differentiate between the two.
When a stock or currency pair is tanking, there are support levels…usually at least 3.
If one were to scale-in at all 3 levels, we’re looking at 2 bad entries??? A habitual “average downer”??? That’s bogus!
I agree to the extent that any indicator is biased to an underlying timeframe and/or strategy. Show me criterion and Ill show you optimization, and vice versa. Markets play many tunes.
Very nicely put Super Investor!
I like the indicator itself- its a nice statistic to have…
but it does need to lose its derogatory name.
I love collective2 in that all its users seem to provide input into making it better, whether negative or positive… its great that all feedback is listened too and debated.
I fail to see how an indicator can be insulting. It doesn’t have a mouth. The problem is with the name and the accompanying text, not the indicator itself.
It has flaws. So what? Every indicator has flaws. A single index will never be enough to characterize systems, because there are too many dimensions that are important.
Having limitations does not mean that it is “useless and completely flawed”. That logic is itself flawed. It is seeing things black & white.
Many similar objections can be made against the Win%. Everyone knows that this can be high for a bad system. That’s not a reason to omit it either.
These two statistics (H&H and Win%) can help each other: If the Win% is extremely high then I would look at the H&H to see whether that might cause the high Win%. Conversely, if the H&H is low, then I would look at the Win% to see whether this also indicates that the vendor is unable to cut losses.
This "indicator" does not measure total system drawdown. It measures the intra-trade drawdown, which has no value in making a judgement about overall system performance or tradability.
There are already plenty of systems that measure system drawdown. This has nothing to do with worrying about this. This indicator will still generally push systems with much higher average per-trade drawdowns towards one end and those with much higher average per-trade drawdowns towards the other.
The first group has much higher percentage of systems that average down or that Hold& Hope (hang on until the trade becomes profitable). Earlier readings of these forums (especially spring) will reveal many many discussions about these problems
