I "think" I understand it, but because I "think" it does not mean that I do. I will post later a (mathematical) sample to show what the reason is why it outperforms the rest.
Prices move up and down, but they finally result in a direction (trend) up or down. So the net (or cumulative) result of all the moves up or down is X $$$$ in 1 direction. That’s the trend. First observation: If you only trade long with the trend, your potential profits is the summation of all up moves. If you only trade short with the trend, your potential profits is the summation of all down moves. If you only trade long against the trend, your potential profits is the summation of all up moves. If you only trade short against the trend, your potential profits is the summation of all down moves. Second observation: The respective summations with the trend are always bigger than the respective summations against the trend. This is obvious because the net move of the market is equal to the summation of the longs minus the summation of the shorts. If the result is positive the trend is long, if the result is negative the trend is short. This is purely mathematical and always for 100% sure. Third observation: Statistically, average profit per trade will be bigger in trading with the trend. So risk/reward ratio will bebetter too. Fourth statement: Moves against the trend always have a higher probability to end in a loss because the risk of a fake top or bottom is bigger when trading against the trend. Moves with the trend move on average faster then against the trend, which means that anti trend traders have to react much quicker and will catch bigger losses. Fifth observation: If you take a wrong top or bottom: Trading with the trend gives you more chances to end profitable as these tops or bottoms are correctional moves against the trend that will eventually end in a move with the trend. You should only see that you are not stopped out on this bad timing. Trading against the trend gives you less chances to end profitable as these tops or bottoms are not correctional moves with the trend that will eventually end in a move with the trend. You should be stopped out on this bad timing or you will lose more money as the move will eventually go with the trend, so against your position.
BONECRUSHER has much of it right, but I don't believe you need any inflation on prices though, we know they will not just stand still cause acreage changes, weather, bugs, governments, interest rates etc, price changes and will never stop. I think dealing with Stocks, it is easier to see trend whether you doing charting by hand or computers. There are hundreds of companies that deal with commodities. And I know the title of the post deals with Commodities and trends. Now some of those who trade stocks will check charts of Commodities which certainly makes sense. Let's take Coffee for instance, November 2015 Coffee made lows and based of last 9 years, they were making quad lows, looking at monthly Starbucks it was making highs, so waiting for retracement makes sense. One way to trade stocks. What does this have to do with trends of Commodity prices? Reflection of if both going same direction there is trend. Or not being open minded, it is not mathematical at all. Trading IMHO is learning to trade risk, and most don't understand this(took me too long to have figured this out as I was sheep for long time) you learn how to control risk best you can and when you can reduce risk better than most, profits will come. I have found before trading myself and working for others that most have risk on back burners, most only want to think about momentum of taking trades in the direction of what they to believe is trend, and there are all kinds of Scalpers who take other sides as their trend might be smaller or larger than daily trend trader, plus those who have deeper pockets like Commercials, Hedge Funds, Commodity Funds and Big traders. You ever check out COT reports? The smaller trader is often wrong, they are generally trend traders. I was lucky to chat with one guy who worked for ADM for five minutes, why use trend at all in Commodities and between reading twice a book called "You Never Lose Trading Commodities" by Robert Wiest, mostly everyone loses trading this Scaled down method and I don't recommend trading that way. But what if you were able to control risks better? You would be able to take a losing system and make it a profitable system. This was one of the "ah-ha" moments in my life, I saw crappy systems in different light. I think if you are closed minded, you close off opportunities of knowledge. Trend has seldom been my friend when I enter I am generally against the masses. If people learned risk first, they would not start losing as often as they do when first starting. Many of us have sizeable egos playing in a game of high risks, but once one thinks it is his way only, can lose out at other places to play.
Being open minded has nothing to do with mathematics. Being open minded means not take everything for granted what others tell you. It is an attitude (state of mind), whereas math has nothing to do with that. You should always be skeptical and see everything as questionable. Many new inventions find their origin in getting out of the already available knowledge and exploring new unknown territory. Or from redefining existing rules. Trading is not different from that. Learning to control risk starts with measuring the risk, which is pure math. Controlling the risk is measured by calculating data, stats and calculating probabilities. So again purely math. Math always gives you the same result if you recalculate data, as it should be to be valid. If you would do this without math, the result will not always(or even never) be the same. It will be influenced by external factors, and that's an additional risk when building a objective system. I trade a mathematical system in a discretionary way. I have to because I am unable to replace the discretionary part by math. When backtesting and repeating the same trades, the mathematical part is always identical but the discretionary part is not. So the risk is in the discretionary part. For the mathematical part I have fairly reliable stats about the future results of my trading, but the discretionary part messes up things a bit and makes that the stats of the trades can vary depending of what the discretionary part does. So the calculated risk is different from the real risk because of the discretionary part. The big advantage of trendfollowing is that it does a big part of the job for you. Theoretically you should take the start of the trend, so the point where prices reverse and never come back at the entry level (so you should be against the masses) . Because that's very difficult you should use a stop that gives you some room for error. Optimal stops can be calculated mathematically. These stops should tell you where the risk is and how big it is, with high probability. Math does not know, or have, "ego" too, which is a big advantage. But professional traders don't have any ego while trading.
I would think that the runaway inflation of the 70's had something to do with it. The market went only up back then and for much longer. Keep in mind that while you pointed out Richard Dennis' inability to adjust to changing markets, the other guys who came up during that era went on to become billionaires. PTJ, Bruce Kovner, Louis Bacon, Helmet Weymar, etc. Markets still trend (always have, always will) but the idea of thinking that you can only buy and prices will continue to rise just isn't gonna happen. Bruce Kovner himself even said that it took him a while to learn to trade outside of the 70s inflation market. He said that he and the others had to get used to trading bear markets.
Working what? How does this chart tell you that TF is working? It tells it is not working unless you're a fool a a born loser.
It tells you that TF worked between around 1980 and 2015, contrary to what the OP said. I don't see how there can be any argument about that. Whether you think it is working "now" depends on your definition of "now"; TF is down in August but up since the start of the year. These are periods of time that are too short to say for sure (in a statistical sense) whether something has worked. It doesn't tell you if TF will work in the future. GAT (Fool, and born loser apparently )
One of original Turtle traders, Jerry Parker, sad in a recent interview that trend trading still performs very good - in fact he said the holding time he uses now is much longer (1-2 years) than any time in the past. It only lost popularity with investors that became so short term focused that they could not tolerate the larger draw downs from trend trading and the fact only 10% of the trades accounted for the profits. The investors opted to have minimal draw downs at the expense of far less profit. There is a lesson in her from one of the greatest trader of all times.
I have been using no trend for Long term trading Commodities since 1991 and have done very well, I changed how I think and trade six years ago going from profit to risk(comparing ratios), have learned how to do trading that many don't seem to do and last six years have added 8 times more signals based on not using trend plus extended min profits from 30% to 75% plus of the nine year range, I have learned hedging much better, but I don't use much of Math at all(I have limited abilities dealing with Logic side of brain as I had surgeries to eradicate tumors)Math is by far too hard for me to understand any more, I test going back 25 plus years and rely on those figures. I am most fortunate I had doctors telling me what most likely will happen and hired and got automated. But there all kinds of traders and even day trading systems I go against the immediate trend on most of the signals, someone has to take other side of the trend traders entries. I really like not caring about trends Long Term Commodities. Back in Oct 22, 2014 in "Live Cattle-Top", I found the highs in Live Cattle and Feeders, took 5-6 trades to find it and in all lost like $20, I recently reversed in Live Cattle but believe it go lower, plenty of rollovers and I think I added two dozen add-ons selling and kept 11 of them as they didn't get stopped out since Oct 2014. So I had a dozen entries all against different trends and took little less than 2 years to complete percentage of nine year range and made 75%, I do this on over 40 commodities and thankfully all automated. OH, once price is in my favor by so much, I keep protective stops alone and never trail, backtesting shows trailing cheats you. I am very delighted by the thousands of trend traders, they help push the markets to other side of nine year ranges. I am glad you have something you believe in and I am always open to risk adverse ways for me to improve but not on entries. I have found a niche that works for me where risk is very low at extremes. ROFLMAO, yea Professionals have some large egos. I have never found Mathematicians to show much emotion at all. Very serious dedicated folks, without them we would never have Great defense and Nasa as we do.