% | $
Quotes you view appear here for quick access.

American Capital Agency Corp. Message Board

  • reits_r_us reits_r_us Mar 21, 2013 11:47 PM Flag

    The Basis of Probability Trading

    I have been doing a lot of soul searching these past few years into the the cyclic nature of the market, I first found AGNC when Ben and a few others were regulars here talking about the runs to EX div and how this was such a great stock to trade.

    I had found AGNC because I determined that folks wanted yield and that they would be drawn most to the highest yielding securities around the time that security paid up. Well I looked and noted a distinct payday for those buying one month out to EX and selling on or about the EX date minus one day(DBEX). In fact that method of trading usually made as much as the 1.40 dividend at the time and more often much more.

    In all of the roughly 20 quarters of AGNC's life I believe we fizzled on only one quarter doing exactly that. One out of 20...those are odds!! So I have enjoyed the rewards of AGNC and a bunch of us starting with Ben, I believe, branched off into the MLPs between the AGNC runs. It was a convenient distraction to fight boredom. The MLPs were not as consistent as AGNC(who is?), and offered hit and miss rewards. They often had lower volume on options and much wider spreads which made for less return on investment.

    The cyclic pattern was there also, though, and those who played consistently, quarter after quarter, were rewarded for their efforts. The trouble was consistency, money management(bet consistency), and the will to continue in the face of losses which were often devastating when risk was not well managed. I remember swearing I would nhever trade the hated HTS or NLY or KMP or LINE again. On many of those I haven't. Not that there is not money to be made but I got mad at the companies for not giving me what I felt I deserved. Wrong!!

    The market is neutral and owes me nothing. It is an ATM teaming with lots of card holders and cash for the taking. And here we are swiping our cards trying to get the ATM to pay. Why do some card holders receive nice green 50's and 100's dropping into the ATM tray, ..more

    SortNewest  |  Oldest  |  Most Replied Expand all replies
    • The only way you can be sure of a successful strategy is to correct for the stock price changes that have occurred in AGNC during the 20 quarters over which you did your analysis. Since AGNC has had a return to shareholders of over 30% per year for the last four years, many trading strategies are likely to look successful.

    • nice one Doc. but before any readers take a flyer in AGNC probability trading, they need to remember a few things.

      1) the dice CAN come up snake eyes more times than you have the cash to stand the losses. so don't bet the entire farm every roll.
      2) if losing even one time is more than you can stand, don't do it.
      3) transaction costs are going to be higher
      4) since you can't hover over the keyboard constantly, get used to using limit orders.
      5) don't second guess yourself. stick with the plan unless you have a real reason to abort it
      6) a long run of wins is just as likely as a long run of losses. don't get cocky and throw the whole farm in there.

      • 4 Replies to mrwizard9090
      • Erratum,

        ""85% of 24 cycles is 19 wins and 5 losses""

        Should read "80% of..."".


      • 2 (cont...)

        should follow the pattern of the past. Oh, the company can go BK, or the board found fraudulent, then yes, the terms change(look at AT and the misfeasance claimed this past few weeks). That is why I selected in the SS, stocks which have a history of consistency, good management, high volume(200k+), etc.

        My earlier gum ball lottery analogy with red and blue balls might be better explained to illustrate this point I am belaboring re historical cyclic probability with GLAD/3%. The securities(all of them) have a 80+% historic win rate using GLAD/3%. Most over 24 months, since the inception of some's monthly dividend. 85% of 24 cycles is 19 wins and 5 losses. Float those in the Lotto mixer and a blue ball(loss) shows up two times in a row. What do you have left? One might say an 80% chance of a blue ball turning up again. NO!! This is not a coin flip. Based on historical probability the two blue balls are removed from the mix and NOW(this is key) you have a probability based upon GLAD/3% and historical probability of 3 losses and 19 wins or 86.4% chance of a red(win) turning up next time.

        The current event is totally random and anything can happen but those are the odds. If two more blue balls turn up in a row, you are getting happy while your fellow trader is quitting. Now your odds for the next cycle are 19 out of 20 or 95% odds of a red(win) ball turning up .

        So that was a lengthy answer to this statement of yours, "" a long run of wins is just as likely as a long run of losses""". You voiced exactly what most people think. It is a fallacy. As I have shown, GLAD/3% and the SS is antithetical to "Long run of losses" and "likely" being bed-mates.



      • Excellent advice Mrwizard,

        If I may comment on several points. On your points as numbered:

        1) Money management is key. 10% of risk capital should be the limit in this model for any trade
        2) Exactly, which is why most will not make the attempt
        3)True, but moot, as a round trip/month on each security is at most $16.00(for one round trip)
        4) There is a specific time to enter a trade with GLAD/3% and a specific time out. No need for hovering, hence less stress and uncertainty
        5) absolutely and again a second major reason why most will not be successful. They will want to tweak the rules, get in or out early or late and abandon the method's rules, not trade the same stock "next" time, just to see it zoom forward with themselves on the sidelines, etc.
        6) Disagree. This idea abrogates or disavows probability theory and its reality in being a predictor of future events. The words "likely" and "long" should not be used in the same context when speaking about future probability when we have gamed the system for the future outcome

        This is a common mistake and why, again, traders will not see the 'forest for the trees'. We look back at the certainty of the past.. There are those reading this at this moment, who take pause at that statement because they assume the past is uncertain. It is not. We can look at the historical record and "see" how a specific stock performed. There is no uncertainty how it performed. It is recorded for all to is certain.

        We can devise a mathematical model which I have done which takes advantage of that past certainty, which if it had been employed over the two to three year look back would have produced 80+% wins. The sample size needs to be large. (the larger the more probable of repeat-ability in the future). Now you are thinking yes, but that was the past, and this is the present and the present is uncertain. Agreed. That is why anything can happen on the current trade. Given a years worth of trades(12 cycles) your results (cont)...

      • Thumbs up! Very well said.

    • cont.

      While others keep depositing money into the machine, with no return? It is , IMO, the lack of four things:

      1) The cognitive recognition that there is money to be made in the market. The realization that the market is neutral and owes me nothing, nor is it evil and want to ruin me. It is a blind machine hoisted or lowered on the backs of all the card holders depositing and removing their monies. The same card holders never trade the same at any separate moment. There is a continual stream of new card holders depositing and removing their money each moment. Hence the uncertainty of the moment as new and old players enter their card swipes on a tic by tic basis throughout the day, week, month and years.

      2) The lack of a consistent plan. The mechanics. I smile as I see the different posters here suggest this stock or that stock that they got a tip on from a friend of a friend of a friend. Often they are right, like Ray85's NTI(Thanks Ray..I made a little money on that). But this "tip" business while fine, is not a mechanical consistent, reliable source of income on a monthly basis. It relies on "others" to supply you with another bone, to last maybe a while or not. We need a trading paradigm. You can buy these for as much as you want to spend or get them free from lugs like me...;-) The point is you need "RULES". Mechanical entry and exit points.

      3) You need securities to plug into your mechanical machine in order to benefit from what the machine was created to produce. It won't do you much good to plug any old stock into most mechanical trading systems, because most machines were created to take advantage of an edge. That edge might be found in energy, or utilities or precious metals and interest rates, ad infinitum. I sought out a machine that would take advantage of predicting buy points for cyclic securities. Cyclic securities I have found often surround the feeding of the fish. The fish get fed at EX div time. The higher the...more...

      • 2 Replies to reits_r_us
      • cont(4)

        So can you do it? Can you tell yourself that you are tired of staring at the ATM, teaming with money while everyone else is getting theirs and you are more often than not buying high and selling low. I just responded to a SA article, where some genius was stating that these Mopays(monthly paying dividends), were not good to trade, because of the measly(his word) change to be made. I told him I will be GLAD(sorry) to take my 20 measly cents/month times the SS, which equals 1.20 x 10,000 shares/month. What a maroon...;-)

        So do you have the will to trade each month even in the face of draw downs in order to reap eventual reward? Do you think that a year of trading these stocks which I have shown to produce winners 80-90% of the time are all of a sudden going to produce winners only 10% of the time?? Only 60% of the time?? NO!!! Can it happen? Yes!! Look at it this way..this example is a little lame(I'll tell you later why) buy it hopefully will cement my point in those teetering doubtful minds. We have one of those Lotto ball gum machines with 9 Red balls and one blue ball, all randomly blowing around inside.

        The machine stops when you exit CLM, red you win(93% this past 15 months), and blue you lose(7% this past 15 months). Now there is a chance, since each drop is random, that the blue ball will turn up every single time. Why our model is different than this random drop is that we skew the ball by tilting it with our GLAD/3% Rules, so that we are forcing 9 drops out of ten to turn up red even though we know that blue might show up, and it does 7% of the time.

        So this was the final point...get your mind right and trade without fear. Read Mark Douglas's book Trading in the Zone. I am reading back through it a second time. He is a kindred spirit and states much more eloquently these simple themes I have been discussing. I am away on the road so I wish you all well while I am in Sunny FL this next week...;-)

        Much happiness and success!!


      • cont.

        yield, the more the food. The more the food the more fish gather. the more the fish gather the higher the price goes. Very simple. So lets select high yielding stocks that have consistent low to high(EX) runs on a consistent greater than 80% historical probability basis. Lets get paid every month(I like that). If a trade is going against us we only have a few days before we exit as we stay in the market the fewest number of days as possible.

        That is why I don't like buying early for KMP, LINE or any security. Randomness and uncertainty. You have your balls hanging out in the air waiting to get home(EX div date) and there is more opportunity to get them chopped off. Very simple. I will take small consistent meals each month than wait for a possible meal every two to three months(think MTGE...wait for the meal a few weeks B4 EX or at the spo), or on these monthly SS.

        OK so you need the right stocks.

        4) Finally, the most important and the one that 98% of traders will fail to employ is mental discipline. You have the knowledge that there is money to be made. You have a mechanical method to enter and exit the market(GLAD/3%). You have the food(securities) to feed the model(SS). Now all you have to do is simple, but, trust me, it is the most difficult. It is the resolute gut determination to trade without fear and to obey the rules. There is no fear when you have probability on your side. You see the stock go down. You rejoice and buy more with 3% Rule. You know you will be exiting the trade in a few days to weeks. What is the worse case scenario. You lose 20 cents, so are that much closer to a win as probability tells you that.

        If you have a 90% probability of winning and you lost once, its not like flipping a coin where each flip is a new 50/50 probability, You have dramatically moved the odds of a win in your favor on the next month. Sure each month is random and you can have another loss but the odds (80-90%) argue way against that outcome. more..

18.88-0.13(-0.68%)May 27 4:00 PMEDT