0
Completed

Ease of use things

Aeti 7 years ago updated by OSV Support 6 years ago 6

1. When sorting a metric in the screener for something like P/FCF I'd like to be able to start at the lowest positive value without having to scroll through all the N/A and negative values.


2. The buy and fair value lines on the stock price chart show the dcf valuation, which may or may not be related to the value score. I think the lines on the chart need to be representative of what you think the valuation is from the value score so I can see what kind of discount to the share price I'm getting.


3. I want to be able to sort by margin of safety. Preferably on some composite value score that you come up with based of fcf.

Further to point 1. It would be nice if we could range bound a metric in the screener for sorting. We might, for example, be interested in companies with a P/B between 1 and 2. We might be looking for a sweet spot, not nescesarily the highest or lowest value.

Started

@Aeti,

1. It will be hard because if the stocks are listed as part of the screen, hard to eliminate results.

What you can do is add another criteria to have P/FCF > 0.


The screeners already allow you to enter range. Just type it into the box in the edit window.


2. Valuation cant be measured accurately and is a core reason why it must not be added to the value score. Valuation must be validated and adjusted manually. Changing growth from 10 to 15% is going to drastically alter the valuation. No way of knowing which one is correct if you try to automate valuation which is a manual process.


3. Been getting this request, so I'll be adding a method to filter and list results based on valuation and DCF. BUT it will be up to the user as we don't guarantee safety of returns by using this method.

Thanks for taking the time to reply.

Hi Jae, my question is only semi related to the above.


Yesterday WMT was a triple A trifecta stock according to OSV ratings. Today WMT has a C for value. I know you have proprietary algorithms running things, but could you provide some background on what prompted such a big jump. It's a big company, so I'm not sure how such a significant change occurred so quickly with no material news.

Here's an article last year on this very question.


Why does the Action Scores fluctuate wildly? and is this volatility an error in the system?



Given the information you have at hand, I would raise the same questions.



So let me zoom out to give you the fuller picture to help you better understand the various reasons why this occurs and what affects the Action Score.



To quickly review, here are the data points that make up the Action Score. I discussed how the Action is created in an earlier email.

  • FCF/Sales
  • CROIC
  • Piotroski score
  • P/FCF
  • EV/EBIT
  • P/B
  • TTM sales percentage change
  • 5 year sales CAGR
  • Gross Profit to Asset Ratio (GPA)
Here's a funnel to make the explanation clearer.
Image


#1. Piotroski Score Causes Stocks to Rise and Fall

The Piotroski Score is the first blue layer for the Q, V and G score.
In order to keep the list and ranking as clean and high quality as possible, the Piotroski score plays a big role in what goes down the funnel or stays out.



E.g. PRKA had a Piotroski Score of 5 in 2014, but then jumped to an 8 in 2015.
Same rule applies to quarterly periods.



If a stock had a high Piotroski score of 8 but then fell to a 5 in the latest reported quarter, the Action Score will drop as a result because the Piotroski Score is importantfor Q,V and G.



The quickest way a stock will rise or drop in rank is if the Piotroski Score changes in a big way.



But why?



During my tests and studies, I wanted to leverage a proven system. I'm not interested in reinventing the wheel. The Piotroski Score has been studied and tested extensively which is why I wanted to use it.



By not having the Piotroski Score as the first line of defense, too many stocks rocketed to the top 10 because of a single crazy number.
A stock like GRSU with a CROIC 4,366.9% would squat high on the list.



By awarding stocks with Piotroski Scores in the optimal range of 7-9 and penalizing stocks with low Piotroski Scores (1-4), I'm able to pass stocks down the funnel that meet accounting and quality standards before the rest of the calculations for Q,V and G are processed.

#2. Metrics with the Highest Weightings

  • For Quality: CROIC has the highest weighting.
  • For Value: P/FCF has the highest weighting with EV/EBIT also getting a high weighting.
  • For Growth: Gross Profit to Total Asset has the highest weighting
But for all 3 groups, Piotroski score always has the highest weighting to weed out the lower Piotroski Score stocks



And you also see that Value includes 2 metrics with high weightings that rely on price. If the market is volatile, the value score is the first one to adjust, which again causes jumps as we update our scores daily. Not weekly or monthly like many other sites that rank stocks.

#3. Not Filtered

The last point to understand is that when you download the spreadsheet with the entire database of stocks and scores, it is not filtered.



If I hide stocks that don't meet certain industry or metrics, you'll have a cleaner and shorter list for sure with less volatility.
But to give you more control, I'm dumping the entire database of stocks for now.



The Action Score is also not suited for financial stocks like banks, insurance, REIT's. That's something I want to improve down the road. Therefore, these financial stocks also tend to stir the waters depending on market conditions.



Once the screener is built out and you have control over which stocks should or shouldn't be displayed, you'll realize that it's more "consistent" :)



Hope that gives you a better background and understanding of the backstory and theory behind how the Action Score is graded and ranked.