What They Are
The Persistence UP and DOWN lists are computer-generated lists of stocks that are screened for a high degree of directional pressure which is usually manifested as an attractive trend or recognizable setup on daily charts. Ordinarily, one list tends to contain stocks in very strong trends (suitable for pullback swing traders), while the other list contains stocks that are in various forms of consolidations or even breakouts in the list’s direction (usually suitable only for day traders except when breakouts are imminent).
The lists are diligently pre-screened to contain only liquid, rather tradable stocks that are usually well-behaved. Of course, the larger traders will have to scale in and out as is customary. The least volatile stocks have been removed, as are many stocks traded on foreign exchanges and many stocks headquartered in countries at war in their homeland. As of 12/24/2010, stocks from all sectors are considered (previously, a few sectors had been filtered out). Typically, 700-900 stocks are ranked.
I normally generate the lists in late morning ET of each market day and put them on this blog. I will also provide an announcement on twitter (follow @daytrend to catch them). Alert traders, who have time, can quickly locate end-of-day entry candidates or candidates to stalk over the coming days for the next trend entry.
How to Use Them and What to Expect
It pays to become familiar with the stocks on these lists, but if you’re a stock trader you probably already are. Trend traders will find numerous opportunities to enter pauses or pullbacks at all levels. Using both lists, breakout traders will find excellent early entries against a weakening main trend. Reversal traders will find superb action at V-bottoms and the end of runaway trends.
To keep the size of the lists contained, I set the limit of each screen to the top/bottom 5% of stocks only. Some market conditions will have more highly persistent stocks in one direction than that. Thus, a stock may tend to “blink” on or off its list as it moves into a healthy pause or pullback and then resumes. A trader who specializes in pullbacks is best served to stalk these stocks from day to day, and intraday, to optimize entry. To do this effectively, keep a stalk list of your own, in each direction, and feed your lists from the most attractive stocks that appear in the respective Persistence List. Then, instead of trading from the Persistence Lists, trade from your stalk lists. As a stock pulls back, you won’t forget about it because it’s on your stalk list, but in all likelihood it will drop off the Persistence List, at least momentarily. If this point is unclear, please contact me. If you try to trade from the Perisistence Lists you will find most stocks overbought / oversold virtually every day. That’s how you know it’s working! Trend traders enter the stocks when they next pull back or consolidate and breakout. From your stalk lists you will may find tradable action in one or both lists nearly every day; however, the correctness of that statement depends on your style and risk tolerance. Indeed, some days you may find too much action.
As an alternative to stalking stocks on these lists by using your own Stalk Lists, there is another way to effectively “game” the algorithm and catch many of the best stocks during market pullbacks. As the market or sectors you target are pulling back, look through several of the previous Persistence Lists, in the main trend’s direction. You are looking for stocks that were on the list in the past few days or at the peak of the swing, and have dropped off the list during the pullback. It’s a good idea to not consider these stocks inferior, unless you consider their pullback to be over done, in comparison.
In July 2010, I began rolling in improvements to the formulation of these lists. One such improvement was to run the algorithm over multiple time intervals of the recent past, instead of just one. Even in today’s correlated markets, stocks from different sectors do not turn at the same time; thus, the best measure of their relative strength should not start at the same time. However, normally, such diffusion is limited to a small number of distinct groups of days, and center point of each group is quite visible and unambiguous. I routinely watch for these unique “pivot point” days, and install a new “leg” to the algorithm for the lists (independently) as I discover them. This is normally the only ongoing maintenance required for the generation of these lists.
An ancillary benefit to paying attention to the sector makeup of these lists, and to the action of the stocks in them is that the trader gets a view of the fluctuations in the market’s aggressiveness, and potentially early indications of exhaustion or sector rotation. Important and reliable glimpses into the market’s immediate future have come to me numerous times from watching action on these lists and watching the simple statistics that I tweet throughout the day. Enough said.
The action of the stocks on the “out of favor” list is surprisingly telling. When the market is dominated by overwhelmingly strong trends, stocks on the “out-of-favor” list may, indeed, be participating in very weak or sideways versions of those trends. This can go on for weeks while numerous wedges or biased consolidations are built. About the same day as the first sign of weakening in the strong trend becomes apparent, those consolidations on stocks in the “out-of-favor” list become ominous and threatening to their holders (long or short), and more than a few will break within a tiny number of days. Over and over, I have seen these breaks occur before anyone is thinking about trading trends in the direction of the “out-of-favor” list. The out of favor list can provide a huge advantage to traders with fire-power at the time, in two ways: The breakout trends, and the warning of change in the main trend. I have seen this behavior in transitions in both directions. Thus, it pays to actively track stocks in both lists all the time. I do.
For many important reasons, stocks should not be traded simply because they appear in some computer-based screen, even a familiar one. The trader should take time to become familiar with the stock and related stocks, and have a sense of how stocks of other companies in the same industry are performing. An industry, or ideally a macro view is an important edge in making such bets. It is the trader’s responsibility (not mine) to determine the suitability of any stock(s) to his/her style, timeframe, risk tolerance, and skill level. These stocks are not for traders who have tenuous emotional control, who have unusual attention issues, who are not well oriented to trend-trading, or who have not developed a clear sense of the money management methods needed to trade trends (especially overnight) effectively.
The screens generating these lists simply rank persistence based on a recent period of usually a few weeks. Of course, as Mark Douglas says, “Anything can happen.” The behavior of any stock might change abruptly and unpredictably, and render the measured period irrelevant as a predictor of its future. Indeed, because these stocks are both relatively liquid and volatile, they are targets of algorithmic trading and options trading that may exacerbate any behavior.
The persistence algorithm does not have any sense of support/resistance, Fibonacci, Elliot Wave, Gann techniques, volume, what traders on StockTwits are doing, options expiration or earning cycles, time or price symmetry, or patterns of any kind, and cannot help in limiting risk other than putting the trader on the side of performance-based probability. The persistence algorithm is strictly a strength measurement that is not very sensitive to any one day’s movement, even an extreme one, but is quite sensitive to persistence or pressure in the recent past. If one is able to stay in a trade from the list in favor by the market, they may find themselves in one of the higher-performing stocks of that industry (or the market), analyzed in hindsight when the trend leg is over. That is a somewhat frequent outcome, but there’s never a guarantee, as sector rotations can render any backward-looking algorithm momentarily irrelevant (and the thrill is gone). My perspective is usually one or several trend trades of 1-3 days each, during the few weeks that stocks sometimes participate. Each trade normally corresponds to a market-induced pullback may resume to the favorable side of its channel, nothing more. A trader should understand this paragraph before trading any of these stocks.
Seasoned traders know:
- Experienced Eye alone > Computer-based screen alone.
- Computer-based screen + Experienced Eye > Experienced Eye alone (or it can be).
- Anyone without an Experienced Eye is unlikely to find consistent profitability in these stocks, lists, or methods.
Traders should protect themselves by seeking to limit market risk and industry concentration. That can be done by using lots of smaller positions (which happens to be my approach). Some traders will use risk-limiting option strategies, an approach I will not address.