The description of the APD Ratio is incorrect and should be changed…
The following can’t all be true…
18% of systems have an APD of .20 or higher (high risk)
13.4% have an APD of .40 or higher (medium risk)
7.9% have an APD of 1.0 or higher (low risk)
So, is an APD Ratio of >.20 considered high risk?
I think those figures need to be rethought…
It would seem that perhaps an APD <.20 or even <.40 might be considered high risk
and are there really only (13.4 - 7.9) 5.5% of systems that have an APD Ratio of between .40 and 1.0?
I think you are misinterpreting. The percentages seem correct. The higher the APD, the lower the percentage. Remember, the APD can be negative, for a system with a negative net profit.
I would consider 0.40 as a decent system. You are comparing a net profit total against a max DD total, so you would expect a fraction to be a decent value.
"The following can’t all be true…
18% of systems have an APD of .20 or higher (high risk)
13.4% have an APD of .40 or higher (medium risk)
7.9% have an APD of 1.0 or higher (low risk)"
They can because the figures are inclusive. I have argued several times before regardless of the merits of the APD as a statistic the guide on how to interpret it is very misleading. The problem is that the statistics are all inclusive, ie the 7.9% that have an APD of 1.0 or higher are also part of the 18% at 0.20 or higher. Personally I think it would be a lot more intuitive to bracket them ie. using the stats from the above example:
5.4% of systems have an APD of .20 to 0.39 (high risk)
5.5% have an APD of .40 to 0.99 (medium risk)
7.9% have an APD of 1.0 or higher (low risk)
The other thing is the summary is also phrased negatively, ie whilst it’s true to say 18.8% of systems may have an APD better than your 19.9 rating, you could just as easily say your 19.9 rating is better than 81.2% of all systems. Obviously it’s the same bottom line but the latter interpretation is psychologically more favourable or forgiving.
The figures could possibly be correct if they are inclusive… but you can’t describe an APD >.20 as high risk… while a an APD of greater than .40 or 1.00 is medium or low risk…
1 is greater than .20… that makes it a part of those systems that are greater than .20 and are then “high risk”
.40 is greater than .20 also…
The description is inaccurate. Systems greater than .20 can’t be described as high risk… if .4 and 1.0 are not…
Maybe, I’m not stating my case clearly.
You can’t describe systems with an APD greater than .20 as high risk while at the same time systems greater than .40 and 1.0 are labled as medium to low risk. Both the .40 and the 1.0 are greater than .20…
See the problem?
Isn’t it listed as <.20 is high risk, not higher than?
Nope…
It says 18% of systems have an APD of .20 "or higher" (high risk)
I agree. I get what you are saying, the problem is as I described, that the percentages are inclusive, and as you say the labels (low/med/high risk) are not. Bracketing them would solve this. I have mentioned this previously to MK and also on this forum before but no-one else seems to think it’s a problem.
It might easily be solved by changing the wording to
"18% of systems have an APD of .20 or lower (high risk) "
That is, assuming that’s still a true statement…
But that would leave a slightly less messy problem of wondering what to term those systems that have an APD that fall between .20 and .40… 
The more confusing thing is how this relates to the "Grid" results.
For example, the Grid shows there are currently 576 active systems. Sorting by APD, reveals 26 systems have an APD >1, which is 4.5% instead of 7.9%.
We also need to take into account that about half of these systems (12) have not yet made more than 10 trades, so can you really say they are "low risk"?
Which brings up an entirely different issue for me… but first…
As I implied before… I don’t really think the percentages are correct in terms of the APD description and unless they are dynamic (change as the systems change)…they can’t be correct.
But regardless of the correct or incorrect percentages it is totally incorrect to say systems with APD’s over .20 are high risk while system over .40 are medium risk… etc…
But back to the entirely different issue… why if C2 advertises that they are monitoring several thousand systems does the grid only pull up less than 600…?
Because there are many many systems that are not being actively supported. People pay the 6 month subscription, and then let it go dead (IE. flatline)
Craig,
I understand that… but I think it’s somewhat (being purposely generous) misleading to say your monitoring thousands of systems when only 600 are still active.
It’s like (for hypothetical example) AOL saying it has 50 million subscribers when 40 million have stopped paying and gone elsewhere…
I believe that being straightforward is always the best path to credibility.
That is something I have always thought. I don’t think it impresses anyone to hear there are 4000 odd systems. It is obvious quickly after coming here, that it doesn’t hold water…
> 18% of systems have an APD of .20 or higher (high risk)
> So, is an APD Ratio of >.20 considered high risk?
Of course not. IMHO it is fairly obvious that the phrase “high risk” refers to “.20” and not to “.20 or higher”. An alternative formulation would be “.20 (high risk) or higher”, but then people can as well misread it and think that “higher” refers to “risk” instead of “.20” - which is also wrong. Your suggestion "18% of systems have an APD of .20 or lower (high risk) " is wrong too; that should be 82%.
For a clear reference the percentage should be cumuative (or what you call “inclusive”), exactly as they are now. This tells you what percentage of systems performs better than the reference point (according to APD).
In other words, I think the formulation is basically good. Nevertheless, since people apparently misunderstand it, I suggest to use more words in the formulation, and make it something like "Systems with an APD of .20 or less are considered ‘high risk’. 18% of the systems have a value higher than .20."
With respect to the problem that ST pointed out, the fact that it seems that the percentages are incorrect, I suspect the text is not automatically updated and that these are values from several months ago.
So, is an APD Ratio of >.20 considered high risk?
Of course not. IMHO it is fairly obvious that the phrase “high risk” refers to “.20” and not to “.20 or higher”. An alternative formulation would be ".20 (high risk) or higher
Jules,
To me, its not about what’s obvious to those paying attention enough to notice the error… it’s just an absolutely incorrect statement and your alternate formulation doesn’t improve upon it at atll.
The statement that 18% of systems have an APD of .20 or higher (high risk) is simply wrong. In your formulation, an APD of .19 wouldn’t be considered “high risk” …?
If you want to use these non-standard statistics like APD they should at least be correct in what they are attempting to describe.
The simpler change is just to say Systems with APD’s of .20 or less may be considered high risk.
With respect to the problem that ST pointed out, the fact that it seems that the percentages are incorrect, I suspect the text is not automatically updated and that these are values from several months ago.
I am not so sure about that, because it now reads "8.0% have an APD of 1.0 or higher (low risk)", i.e. the number just changed from 7.9 to 8.0 which would be more consistent with daily updates.
It’s nice to know that the numbers seem to be dynamic at least.
So, maybe the discrepancy that was pointed out as to the percentages on the grid vs those in the apd pop up box may be due to the apd ratio being figured on the phantom 4000 systems and not just the 576 active ones on the grid.
I still have a problem with the
.20 or higher wording…
But I disagree that it is “an absolutely incorrect statement”. I think the statement has two interpretations: a correct and an incorrect one. In my experience that is something that usually happens when someone tries to describe a math formula in natural language while trying to be brief at the same time. I don’t consider that an error. The reader should seek the logical interpretation. That my first alternative formulation doesn’t improve on it is true; I said that myself. This was an example of why it is difficult to formulate it unambiguously, except if you use more words, which is what I actually suggested. I agreed that a formulation like “Systems with APD’s of .20 or less may be considered high risk” is better because it probably causes less confusion.
Well Jules… at least we agree on something.
The statment should say .20 or less (high risk)
But I honestly don’t think there’s any math/english conflict in the sentence.
While the error is pretty obvious… it can not be true that systems with APD’s of .20 or higher are “high risk” (Which is exactly what the sentence says)
if
Systems of .40 or higher are medium risk…
They conflict with each other and can’t both be true since .40 is inherently greater than .20 which has already been called "high risk"
But I do think this horse has been about beaten to death. There is a really simple solution… just change the wording to
.20 or lower (high risk)
