Tradewinds

Are you asking how long it takes from the start of the analysis to deciding to enter a trade? I'm day trading, so I'm monitoring charts non-stop. As of this moment, I've never stopped wanting and trying to improve my trading charts and trading set up, and maybe that will never end. Are you talking about analyzing charts, deciding on a stock to pick, and then entering a long term trade?

The close of the DOW and the close of the YM move up and down almost in perfect sync. They aren't the same value, but if you chart the close of the DOW, and the close of the YM, the lines look almost identical. So, again, the YM close and the DOW close are different values. There is a spread between the two values. That spread changes as the YM gets closer to it's expiration date. During any given day, that spread between the DOW and YM stays in a range. On Apr. 18 the spread between the DOW close and the YM close was right around 61 - 62 and ranged from about 58 to 65. So the close of the DOW was about 61 higher than the close of the YM on Apr. 18, 2013, on average. Just a rough guess after charting some simple data. At 4pm the YM closed at 14480 on Apr. 18th. The DOW closed at 14537. That's a difference of 57. That night, and the next morning before the 9:30am open, the YM dropped off. At times the spread between the close of the DOW, and the YM trading overnight was as high as 124 and as low as about 32. So, it went from a narrow spread during the day, to a very wide spread overnight. Of course, the DOW isn't trading overnight, so the comparison overnight is between a YM that is changing, and the DOW close at the end of the regular day. Three minutes before the open of the next day, at 927 am, the YM had fallen to a value about 100 less than the DOW's previous day's close. On the open, the DOW immediately surged up, as the YM fell off hard. The DOW, and YM attempt to sync back up almost immediately on the open. There are various possible scenarios shortly before the open of the day. * The YM is right, the DOW moves towards the YM. * The YM is wrong, the YM moves towards the DOW. * The spread between the YM and DOW is about the same as yesterdays close * The YM is closing down on the DOW from above. * The YM is closing up on the DOW from below. * The YM close crosses yesterdays DOW close one way or the other before the open. On the 19th, there was a lot of volatility on the open of the YM. It moved in a range of almost 40 points in the first 10 minutes. A hard drop on the open. The YM dropped hard on the open, the DOW surged up in the first minute. Probably syncing back up after a 100 point spread before the open, as opposed to a 62 point spread the day before. So again, on the open the DOW and the YM were coming back together. At 933 am, the YM had a lower low, and the DOW had a higher low. The DOW wins, this divergence sends the YM higher. A divergence like this probably won't happen during the day. It's something that happens on the open. You wouldn't want to be on the wrong side of a 40 point swing in a couple of minutes on the open, so that's why most people avoid the open. At 936, the YM drops more than the DOW does. Again, this divergence is a signal. The DOW and the YM rarely diverge much at all DURING THE DAY. If they do, I'm betting on the DOW being right. At 957am on the 19th, another divergence happens. The YM hits a new low for the morning, the DOW low is actually higher than it's lows at 930 and 933. Again, I'm betting on the DOW being right. I'm not using the pre-market DOW to YM behavior here to predict what will happen after the open. I'm just observing what is. Just observe and accept what is. From that point I have a piece to the puzzle. If anyone else has specific observations, or knowledge, I'd like to hear what you know. I only know what I know. There is a lot I don't know. I'm just observing, and trying to understand, and hope not to get to far ahead of myself. I need to trade what I see, and not what I hope is going to happen. Even though I am not predicting what will happen on the open, I can be open to the possibility that large DOW to YM spreads before the open, may cause volatility and wide ranging price action. My preparation is a mental preparation to see possibilities, not to predict to far into the future.

So, don't move your target and stop as a way to keep your emotions from affecting you? Not a bad idea in some circumstances. Denial is my enemy. When everything looks bad, I can have a hard time accepting the loss. In that situation, leaving the stop loss in place is better. I prefer to move the profit target to a loss, take a smaller loss than the stop loss, and reverse the trade in the other direction if the indicators are telling me the trade is failing, and there is reversal potential. If I enter a trade, and shortly after the entry, see that it was a bad entry, I'd rather take a smaller profit, and get out. If you get the perfect entry according to your rules, and everything looks fine after the entry, there's no need to do much. I rarely get the absolute perfect entry I'm looking for. If I did, I probably wouldn't make any changes. Basically, I have a stop loss as catastrophic situation insurance. Like if the power goes out, the computer crashes, etc. Your target and stop will be affected by volatility, and price ranges. I enter a trade with the target and stop exactly the same every time, but adjust it immediately depending on things like the last swing hi and low levels, and volatility. I could see leaving those initial changes in place as a matter of discipline and consistency, unless I quickly see that I made a bad decision on the entry.

The OPTIMAL decision window is often very small, . . . not a lot of time to make the correct decision. (You can make a bad decision anytime you want.) It is impossible to make a perfect entry on every trade. So, there must be an acceptance of imperfection. I'm imperfect, and will always make some percentage of bad trades. It WILL happen. There is no way to avoid it. But I can get better at recognizing when I get into a bad trade, and manage it well. How well I manage the bad trades is at least 50% of the determining factor of whether I'll make money that day or not. The rules of the trade must be prioritized. If your setup needs 5 criteria to be met, which one is the priority? Which one happens at the very last second? Are there any that happen ahead of time? If 3 of the 5 criteria happen first, then 2 happen at then end, practice the order in which you evaluate your criteria. Cycle through the criteria in order by priority. It's very boring and tedious to be constantly monitoring the same things over and over and over again, the same way, in the same order. Your mind can wander. You can be undisciplined, careless, and impatient. So what is the motivation to stay focused and pay attention? Without the motivation, there is no energy to draw on. The mental discipline requires mental energy. The mental energy has to come from somewhere. There are positive and negative motivators. You can be so fearful of making the wrong decision, that you are hyper attentive. That can provide the energy to be disciplined, but the fear is also very harsh on your nerves and mind. It's better to want the satisfaction of knowing that you executed your trades with precision and competency. If you executed your trades with precision and competency, and you loose money, then your strategy is no good. But at least that way, you know it's your strategy, and not you. If you execute your trades with imprecision and doubt, and you consistently make money, you probably have a really good strategy. (As a side note, if you consistently loose money all the time, consider what would happen if you had always done the opposite of what you are doing. The issue is consistency) Why do we make the same mistakes over and over and over again? Partly, because we are undisciplined and unmotivated to change. If I am motivated to change, there is a good chance that I'll change. This is why I think that searching for motivation is critical. I can't just wake up one morning and use my willpower to become disciplined. I can wake up one morning, and make the decision to start the process to become disciplined. It's a process. People who make New Years resolutions, and break them 2 days later, started the process, but lacked the motivation to sustain the effort. You can make the decision, and engage your will power to initiate the change process, but without the motivation you can't sustain the effort. That's why no real progress gets made. The long term energy to sustain the effort has to come from somewhere. Where are you going to get the energy and motivation to sustain the long term effort? That's the critical issue. I do believe that we are genetically programmed to behave a certain way, and we often behave more like wild animals just trying to individually survive today, than civilized people trying to insure indefinite existence into the future. The point is, we are flawed. Trading brings those flaws to the surface. Being in denial about personal weaknesses and flaws that affect trading makes it almost certain that you will fail.

Since you are using thinkScript, this may be of interest to you. Least Squares Method. Yahoo ThinkScript Group Least Square Method. Wikipedia #This program created by tradescripter #Dec. 24, 2010 Merry Christmas #It uses the Least-Squares Method to forecast a new price #For an explanation of Least-Squares method see: #http://en.wikiversity.org/wiki/Least-Squares_Method input length = 9;input price = close; def AvgPrice=average(price,length); def SumTime=fold i = 1 to length + 1 with x = 0 do x + i;def AvgTime=SumTime/length; #sx is the sum of all the deviations from time for the last x bars def sx=fold j = 1 to length + 1 with y = 0 do y + ((j-AvgTime) * (getValue(price,length-j,length+1) - AvgPrice)); #sy is the sum of all the deviations from price for the last x bars def sy=fold k = 1 to length + 1 with z = 0 do z + (power(k-AvgTime,2)); #m is the slope of the line, b is the slope intercept of the line in the equation y = mx + b def m=sx/sy;def b=avgPrice-m*AvgTime; plot FuturePrediction=(m*(length+1))+b; This code was also posted in the Thinkscripter site, and improved upon by others.

Yes. You can use thinkScript. There are a couple of groups where you can find support: TOS_thinkscript : Think or Swim Thinkscript programmer ThinkScripter | thinkScript Indicators for thinkorswim There is also a programing forum within TradersLaboratory. coding-forum/ This indicator automatically draws price level lines on the chart, and extends previous lines. It uses subroutines in the program flow. I programmed this, but offer it freely to anyone wanting to use it. I do not ask for any reference to myself as the author. I'm offering the code so that I can go on my little ego trip, but trying to go on less of an ego trip than most. //10-3-12 input: ShowOnlyToday(True), ShowAftrHrs(True); // True will plot lines for after hours also //hh=higher high, ll=lower low var: DayHrs(False), AftrHrs(False), NewDay(False), hl(False), CntHiDwn(0), CntLwUp(0), CntLwDwn(0), CntHiUp(0), CntMjrHiDwn(0), CntMjrLwUp(0), Peak(False),Bttm(False), pk2(False), Bttm2(False), CrrntHi(0), CrrntLw(0), MajorHigh(0), MajorLw(0), MajorHigh2(0), MajorLw2(0), HighExt(0), LowExt(0), MjrHiExt(0), MjrLwExt(0); Method void MainMethod() Begin Signals(); // Call Method subroutine to calc Peak and Bttm signals If NewDay then Begin // Reset the Hi and Low at the Beg of a new day CrrntHi = H; CrrntLw = L; End; // If it's after hours, and the input is set to exclude After Hours, then quit here If ShowAftrHrs = False and Aftrhrs[1] = False and Aftrhrs then Begin CrrntHi = 0; MajorHigh = 0; CrrntLw = 0; MajorLw = 0; Return; End; If Peak[1] or pk2[1] then CrrntHi = H[2]; If Bttm[1] or Bttm2[1] then CrrntLw = L[2]; If (Peak or pk2) and CrrntHi < CrrntHi[1] then CntHiDwn = CntHiDwn + 1 Else If (Peak or pk2) and CrrntHi > CrrntHi[1] then CntHiDwn = 0; If (Bttm or Bttm2) and CrrntLw > CrrntLw[1] then CntLwUp = CntLwUp + 1 Else if (Bttm or Bttm2) and CrrntLw < CrrntLw[1] then CntLwUp = 0; If CrrntLw < CrrntLw[1] then CntLwDwn = CntLwDwn + 1 Else if CrrntLw > CrrntLw[1] then CntLwDwn = 0; If CrrntHi > CrrntHi[1] then CntHiUp = CntHiUp + 1 Else if CrrntHi < CrrntHi[1] then CntHiUp = 0; If CrrntHi < CrrntHi[1] and CntHiDwn = 1 then MajorHigh = CrrntHi[1] Else if CntHiUp = 1 and L < CrrntLw then MajorHigh = CrrntHi; If CrrntLw > CrrntLw[1] and CntLwUp = 1 then MajorLw = CrrntLw[1] Else if CntLwDwn = 1 and H > CrrntHi then MajorLw = CrrntLw; If MajorHigh < MajorHigh[1] then CntMjrHiDwn = CntMjrHiDwn + 1 Else If MajorHigh > MajorHigh[1] then CntMjrHiDwn = 0; If MajorLw > MajorLw[1] then CntMjrLwUp = CntMjrLwUp + 1 Else if MajorLw < MajorLw[1] then CntMjrLwUp = 0; If MajorHigh < MajorHigh[1] and CntMjrHiDwn = 1 then MajorHigh2 = MajorHigh[1]; If MajorLw > MajorLw[1] and CntMjrLwUp = 1 then MajorLw2 = MajorLw[1]; If CrrntHi < CrrntHi[1] then HighExt = CrrntHi[1] Else if H[2] > HighExt[2] and HighExt[2] > 0 then HighExt = 0; If CrrntLw > CrrntLw[1] then LowExt = CrrntLw[1] Else if L[2] < LowExt[2] and LowExt[2] > 0 then LowExt = 0; If MajorHigh <> MajorHigh[1] Then MjrHiExt = MajorHigh[1] Else if H[2] > MjrHiExt[2] and MjrHiExt[2] > 0 then MjrHiExt = 0; If MajorLw <> MajorLw[1] Then MjrLwExt = MajorLw[1] Else if L[2] < MjrLwExt[2] and MjrLwExt[2] > 0 then MjrLwExt = 0; RunPlots(); End; Method void Signals() Begin DayHrs=T>930 and T<1558; AftrHrs = (T>1558 and T<2300) or (T>200 and T<930); NewDay = D[2] <> D[1]; Peak = h<=h[1] and H[1] >= H[2]; Pk2 = h<=h[1] and C[1] > O[1] and C[1] < O[2]; Bttm = l>l[1] and L[1] <= L[2]; Bttm2 = L>L[1] and C[1] < O[1] and C[1] > O[2]; End; Method void RunPlots() Begin if CrrntHi > 0 then plot1(CrrntHi, "Current High") else NoPlot(1); // Set to points so that the price of the line shows up on the axis if CrrntLw > 0 then plot2(CrrntLw, "Current Low") else NoPlot(2); if MajorHigh > 0 and MajorHigh-C < 20 and MajorHigh-C > -20 then plot3(MajorHigh, "Major High") else NoPlot(3); if MajorLw > 0 and MajorLw-C < 20 and MajorLw-C > -20 then plot4(MajorLw, "Major Bottom") else NoPlot(4); if MajorHigh2 > 0 and MajorHigh2-C < 20 and MajorHigh2-C > -20 then plot5(MajorHigh2, "Major High2") else NoPlot(5); if MajorLw2 > 0 and MajorLw2-C < 20 and MajorLw2-C > -20 then plot6(MajorLw2, "Major Bttm2") else NoPlot(6); if HighExt > 0 and HighExt-C < 20 and HighExt-C > -20 then plot7(HighExt, "High Ext") else NoPlot(7); if LowExt > 0 and LowExt-C < 20 and LowExt-C > -20 then plot8(LowExt, "Low Ext") else NoPlot(8); if MjrHiExt > 0 and MjrHiExt-C < 20 and MjrHiExt-C > -20 then plot9(MjrHiExt, "Mjr Hi Ext") else NoPlot(9); if MjrLwExt > 0 and MjrLwExt-C < 20 and MjrLwExt-C > -20 then plot10(MjrLwExt, "Mjr Lw Ext") else NoPlot(10); End; Method void OnError(int ErrorNum) Begin print("Error! ", ErrorNum, "Time of Error=", T); End; { This triggers the Main Method subroutine to run, and restricts the lines to the number of days set to show in the INPUT} //If DateToJulian(D) > (LastCalcJDate - DaysToShow) then Begin if ShowOnlyToday and D = CurrentDate then MainMethod() // Process the lines being drawn Else if ShowOnlyToday = False then MainMethod();

Every price pause is a Support/Resistance level. I have two basic formulas that define a peak or a bottom: Peak = h<=h[1] and H[1] >= H[2]; Pk2 = h<=h[1] and C[1] > O[1] and C[1] < O[2]; Bttm = l>l[1] and L[1] <= L[2]; Bttm2 = L>L[1] and C[1] < O[1] and C[1] > O[2]; I have an indicator that draws support and resistance levels automatically. Previous high and low levels have their lines extended. The above code is just a small piece. If the program needs to keep track of previous highs and lows for comparison to other data, an array, or multiple variables need to be defined. The price level indicator I programed is in Tradestation. I'll post the entire code if anyone wants it. It could easily be translated to some other language. If the program needed to use price levels from a different time frame than the chart being used, that might be more complicated. Some platforms can easily get previous daily or weekly highs and lows even in an intraday chart, and if that's what you need, it might not be that difficult. (Relatively)

If looking for price patterns is what you wanted a computer program to do, the program would need to capture different conditions, store them, then constantly be looking for condition combinations that matched your definition. For example, if the price pattern required two lower highs, the computer program would need to identify a lower high, set the count to 1, and start looking for another lower high. If the program was looking for correlations between different ticker symbols, multiple symbols or multiple data feeds would need to be accessed by the program. You would need a platform that can handle whatever the requirements are. Scans can filter for lots of different criteria, and might not need a lot of programing knowledge.

When I program a signal, then have my indicator display that signal on the chart, invariably what happens, is that many of the signals are bad or false signals. The calculations don't lie. What I see on the chart, and think is a good signal, often isn't as good as I thought. Or the raw calculations cause me to see something I didn't see before. So it's good to have a programed indicator generate unbiased output as a test. So, to quantify trading signals will take some programing skills. A person could invest years, and thousands of hours, and not be guaranteed success. I like to convert data into oscillating indicators. So I'm using the change in the data from bar to bar as opposed to the net value. The change in values usually operate within a range, or within certain levels. One thing to look for is price or data to break a previous level. So there needs to be a way to define when a level was established. The price went higher on the last bar, but didn't on this bar. Maybe that was a new price level. But it's not always that simple. If you want to quantify trading data on your own, it's like being a research scientist. There's a lot of trial and error. Regardless of how much time and effort anyone puts in, or how skilled you are, the odds are against you.

The HL2 is from the same bar. The high of this bar, and the low of this bar. Add the two values together, and divide by two, and that is the center of the bar. So if the HL2 was on your chart as an indicator line, it would go right through the center of each bar. The change in the HL2 is calculated from this bar compared to the last bar. Some people speak in terms of 'Slope', how steep the line is. I just use regular candlestick bars. I don't read trading books. I don't study other peoples strategies. I do read about what people are doing, looking for information or ideas, but have always worked on my own strategy. The bottom line is trying to validate what works and what doesn't work. I try to break the data down in different ways, looking for some kind of correlation. I do a lot of programing, and am constantly tweaking my indicators, looking for some kind of an 'edge'. I like indicators that are oscillators. I do pay attention to Floor Trader Pivots, and a basic Exponential Moving Average. At 15:36 and 15:37 today, the price was trying to move up a little on the YM. The $TIKI HL2 failed to go positive, The YM volume failed to go up, then the price had a big drop.

I like to use the HL2 as a basic input in comparing price to other things. (High + Low) / 2 It's simple, it follows the price moves fairly accurately compared to other calculations. Then you can take the change in the HL2 from one bar to the next. HL2 - HL2[1] Then I like to compare what the price did to what other things did. For example: Did the HL2 price change go up more than the change in volume? I'm constantly trying to determine whether the amount change in price seems justified compared to the change in other things. What trading platform are you using?

It's very difficult to determine when price will turn in the opposite direction. If if were easy, everyone would get into trading, and everyone would make money. It's like a casino; if everyone made money at the casino, they would be open for one day, then go out of business, or quickly find a way to put the odds overwhelmingly in their favor. Price moves very quickly to new levels, making it difficult to react in time. This leaves the trader with two basic choices: Enter an order at a target price in advance, and hope for the best. Try to time the order point, and hope you can react fast enough. Both choices have serious flaws, putting the trader at a severe disadvantage no matter what you do. If price always moved very slowly, that could be seen as a lack of opportunity to the trader. In order to make money, there needs to be price movement. So either way, there are disadvantages. If price moves fast, it poses challenges. If price moves very slow, there is less opportunity within that time frame. So what's the better choice? Try to enter orders very quickly, or guess at target levels and enter the order in advance? It was pointed out that the price pauses after each price level, and at that point a decision needs to be made. During that price pause, there is plenty of time to exit, lock in profit, and try to decide if the trend will continue or not. If you exit, and re-enter in the same direction, then the trade goes against you, at least you've locked in some profit. You may loose money on the next trade, but hopefully, overall you won't loose. If you can somehow break even on the bad trades, or not loose to much, that's half the battle. If you react too fast, and get a bad entry, then a fast reaction time is not an advantage. A fast reaction can be either bad or good. If you try to act very fast, but fail to analyze the situation because you didn't have enough time, it's basically just trading randomly. You might get lucky, you might not. No matter what perspective you take in your strategy, there are advantages and disadvantages. Good decisions need to be made that put the odds in your favor. I'm not saying that I do that. I'm speaking from experiencing and knowing how stupid I can be. One mistake I make, is that I take profit, then immediately get back in a position at a better price without having time to analyze whether it was a good decision or not. Locking in the profit, and getting back in at a better price isn't the mistake, that's fine. The problem is that I'm just 'Rolling the Dice', acting on a hunch, and taking a chance. Later, I can look at my charts and see what I should have done, but I have the luxury of time 'after the fact'. I need to take advantage of those sideways price pauses as a way to have enough time to make a good decision.

It's true that the $TICK is not perfectly correlated with price moves. I have not found anything that is perfectly correlated with price moves. Here are some Tradestation ticker symbols: $TICK - NYSE all stocks $TCKSP - SP500 $TICKC - Composite $TICK across different exchanges $TIKI - DOW stocks $ADVI - DOW Advancing Issues $DECLI - DOW Declining Issues $ADV - NYSE Advancing Issues $DECL - NYSE Declining Issues $UVOL - NYSE Up Volume $DVOL - NYSE Down Volume $UVOLSP - SP Up Volume $DVOLSP - SP Down Volume $UVOLI - Dow Up Volume $DVOLI - DOW Down Volume $AUVOLI - Dow Alternative Up Volume $ADVOLI - DOW Alternative Down Volume There is also data for AMEX, NASDAQ, Russell, SP400, ARCX, OPRA. As far as using the $TICK, or any of the internals data as indicators, it's not just a matter of simple divergences. There could be a higher $TICK high at a peak, instead of a lower $TICK high, but a lower CLOSE on the $TICK at the peak. There could be a higher $TICK high, but a divergence on the Advance/Decline. There are many possible combinations. There are patterns that are reliable. A decision also needs to be made about what might take priority. If there is a huge divergence on volume, I might use that as the most valid signal. Here is an example of a VOL divergence today. The DOW volume has just gone positive, and had a higher low, the price of the YM went lower. Right at that point, at 11:15am the price of the YM goes into an uptrend. So there is a divergence. Price went lower, DOW volume did not go lower.

Thank you for your response. Is the market cap being constantly updated second my second? Is there any way to get that info in real time? In other words, do exchanges and/or brokers provide that kind of data? If that is the way the indexes are calculated, then someone must have that data and be using it second by second? Would I need to be feeding data from all 500 stocks into a spreadsheet and calculating the formula myself? I'm trying to understand how the calculation for the index could be used in real time for actual trading.

Thanks, I appreciate the explanation. That makes sense. People on all sides of the transaction looking for opportunity. Short term, (Minute by Minute) it's probably whatever side of the transaction has more activity. Someone could see that 1000 contracts of the ES were just bought, and decide that the 'best value' instrument available at the time, would be to buy a stock rather than the futures contract. That would result in more buying volume of the stock at the same time the ES volume is going up. Probably if I understood options and hedging I'd be able to have a more complete understanding. Basically, if someone sees that 1000 contracts of the ES were just bought, and there is relative less buying activity on the SP500, someone is going to attempt to take advantage of that discrepancy. For the buyer of the 1000 contracts, at some point, they need to get out of that transaction. Someone has to be on the other side when they want to get out. So someone is going to try to be there on the other side when they want to get out, hoping they will make money. And I'm assuming there are multiple ways to do that. I'm just guessing here. I'm assuming that's the way it happens. The extent of my understanding is just trying to absorb information here and there. Is there a good book, or article, web post that explains the basics of how different instruments interact to each other? Probably plenty of books on strategy based on some indicator, but I'd like to know a simple overview of all the instruments and basic tactics.