Frank

this EasyLanguage code for 'Rotation Factor' (a Market Profile indicator) was working fine but somehow got screwed up in a Open Ecry software upgrade. I have a headache trying to fix it because it seems to have logic that is perfectly fine. Can someone take a look at this and figure out how to re-write it in a different way that might make it work? I am thinking there is something wrong with either the word 'summation' or with the 'length' code.... What its supposed to do is track 30-min bars relative to the one preceeding it and keep a 'running score' --- such that if a low goes below the previous low, that is -1 -- if the high makes a lower high, that is -1 for a total score of -2 for that bar. these are then summed for the intraday session since the beginning of the day (you don't score the first bar of day fyi). so a rotation factor of -4 would be a day where you have made 2 lower lows and 2 lower highs after the first THREE bars. inside bars (ie, high<high[1] AND low>low[2]) and outside bars are scored as a zero. again, this WAS working but no longer does so what I need is for the same content to work --- but somehow coded differently in the hopes that OEC will be able to read the re-worked code. thanks in advance: vars: aa(0), bb(0), cc(0), dd(0), ee(0), firstbar(0), length(0); if date>date[1] then begin firstbar=currentbar; end; length=(currentbar-firstbar); if h>h[1] then aa=1 else aa=0; if h<h[1] then bb=-1 else bb=0; if h=h[1] then cc=0; if l>l[1] then dd=1 else dd=0; if l<l[1] then ee=-1 else ee=0; if l=l[1] then cc=0; value1=aa+bb+cc+dd+ee; if date = date[1] then value2=summation(value1,length); if date = date[1] then begin plot1(value2,"count1"); end; plot2(0,"0");

Brett Steenbarger yesterday did something very similar to what this blog is about: TraderFeed: VIX and Average Daily Volatility in the Stock Market "What we see is that the average daily range is closely related to VIX levels"

most brokerages have a download data option which is quite simple. I have searched for public data on this in the past and could not find it. I saw someone say that barchart.com allows this but at the time, I couldn't figure that out: http://charts.barchart.com/chart.asp?sym=ESM9&data=Z10&date=121108&den=MED&divd=n&evnt=off&grid=N&jav=ADV&size=D&sky=N&sly=N&vol=N&late=Y&ch1=011&arga=&argb=&argc=&ov1=&argd=&arge=&argf=&ch2=&argg=&argh=&argi=&ov2=&argj=&argk=&argl=&code=BSTKIC&org=stk

stats are the same for either range or % of range....

I tested those and both fall short compared to VIX. For example, using the 3-day average -- I get only 68% of days that make this number the following day -- as compared to 82%+ for VIX. This is multiplying each by 0.8x (we want high odds of 'at least XX points' in range). Using things like the 3-day median or 10-day median yield similar results -- (each method calculated using the last 18 months as my test period).

so idea is that if you know that most of the time market is going to make a key high or low in first ~3 bars - figure out if the high or low is being made first and find a trade that is consistent with that intraday move. today, the market traded up FIRST for its key high and then went directionally to take out the morning low and previous day low. so you usually have 2-3 pivots to think about -- the previous days close and the previous days high and low. the market does 'usually' like to trade towards the previous days close for a few bars at some point... so if it does this, don't be fooled that it has chosen its direction for the day -- this is normal action. so its like a puzzle every day. the last 2 days (prior to today), the correction that lasted a few bars towards the previous close still wasn't enough to take out the opening 30-min bar low and did not really threaten the previous days low --- when that happens the previous days high (or the morning high) now becomes a directional pivot for the remainder of the days direction. today, the market corrected towards the previous days close (at 854) but never really threatened to take out the previous days high... thus, the previous days low became a target --- which lined up because we were due to take out the morning low after so many consecutive days of NOT taking out the morning low (3 straight low made first's -- where a morning high is set and then subsequently taken out -- today the opening 3-bar high was set and never again violated)

I do all this stuff and can see it all perfectly with a DVI connection and cheap but powerful monitors that I upgrade every few years --- generally buying one new monitor a year and kicking the bottom one out of the rotation... just got back from Best Buy and really liked this Gateway -- http://www.amazon.com/Gateway-FHD2401-Widescreen-LCD-Monitor/dp/B001CBUTJS/ref=sr_1_3?ie=UTF8&s=electronics&qid =1239745790&sr=8-3

personally, not sure what people are talking about when talk about resolution etc... we are not doing PC gaming here, just charts --- I have never had a problem with any monitor I have ever had -- perhaps that is luck but I have also never bought a high-end monitor. what a waste of money to stay on cutting-edge..... -- 'Late Adopter' Frank

had good trade on yesterday and choked... caught a decent trade today to hopefully start a run for a while...

was thinking the same thing --- I am in slow upgrade cycle --- toss out worst/smallest monitor and replace it with bigger one that is good value. here is a 28" monitor for $349.99 http://www.amazon.com/Hanns-G-HG281DPB-Widescreen-LCD-Monitor/dp/B000TJV9KW/ref=cm_cr_pr_product_top

I wasn't even aware those contracts were available to trade... can we trade those brownsfan? I thought I tried to trade DAX once but it didn't work.

The way I think about it is that my statistical tendencies and market profile concepts are 'my book'. Taylors book calculations allowed him freedom to go long on sell days and short buy days etc... so to Taylor, its the calculations that matter much more than the label of the day. ---- right, if price opened high today, would be a traditional ss-day --- with a penetration of the previous high or a test of the previous high (failure to penetrate) marking the location for a good short. if the move instead happens overnight then the location is no longer favorable and have to re-think using the concepts: 1. Inside days are rare -- so market will very likely take out either Mondays High or Mondays Low at some point today. This offers a directional target to shoot for. 2. 3 low-first days in a row argue for a 'high-first' day --- so one bias would be for price to take out any morning low it forms by the end of the day --- which would then trigger a reversal in the market profile auction from up to down. you can anticipate this if you find a pattern that makes sense. 3. If the high of the first 3 30-minutes is penetrated, then the statistical odds then favor are on the side of low first. 4. If the high is not made in first 30 minutes -- then odds drop that you will close really far away from previous days close. 5. Once the morning high or low is set, you can do high-odds range calculations for a high-odds price target.

I am not an Ensign expert -- I just use their software for historical MP because its so easy (and cheap) --- just set your chart to the time period you want -- days, intraday whatever and choose 'all visible bars' here is an example setting it to 14 bar daily:

A picture is worth a thousand words (fuuuuuccckkkk) hopefully by posting this chart, I will shame myself into next time having the stones to ride out one little bullshit juke.... went LONG because felt market was tipping its hand -- can't explain the exit except that I guess I just wasn't feeling very sure about much given all the crappy churn we have had lately.

the taylor day count is just a subset to how I think about it.... I may lean that way should it fit my overall concepts or I may not. I prefer the framework of the market profile 'auction' --- as in, an auction CAN end when there is a high or low violation --- and I may look for that --- but the auction can also continue until a 'morning low or high' is violated. so today, we came in after low made first on Thursday. therefore, the 'auction' is up -- until a good morning low is taken out. But the market made its low during first 30-minutes today -- consistent with the statistics saying where it should. therefore, no matter what the day Taylor called it, I am not going to short and will try to find a spot to go long.

Let me provide some data to support what I am saying. I took all full trading days since November 2007 and divided the market into two buckets: 1) if price made its low or high for the day (low if low first, high if high first) in first 90 minutes of trading or 2) high or low made in any other time period past the opening 90 minutes. Note that there are usually 20-22 trading days per month so I am trying to show that the Taylor Cycle 'should' kick in during those first 3 bars about 75%+ of the time. In March, it was 19/22 (higher than normal). personally, I find these statistics to be quite powerful for more than just their face value. for example, if you know that the first 30-minutes often makes the low for the day, many times you can go long during that timeframe and get a nice trade on -- and price will end up reversing later on and the first 30-minutes won't be the low for the day -- but you made money on your trade anyway simply because there was enough buy-pressure to create a squeeze for those fearing that it 'could' be the low of the day.

I am not advocating this but why not wait for range on monday to approach the range made on friday and fade price at that point for counter-trend move. note that volatility is measure of 'returns' -- not range. 'returns' means the final close vs the previous close. thus, if you have a large gap -- you can have intraday range that is lower but 'volatility' will technically be higher if close is far away from the previous close. range is very much correlated to volatility -- but its not the same thing.

Monad, what I am showing is that statistically speaking -- you can/should be actually looking for the 'made first' objective in the first few 30-min bars. Not only this -- but you can quantify how often it begins its move in each period. This is highly complimentary to taylors 'rules'. For example, if you are thinking of 'buy day' the next day --- ideally you would like to get into the 'buy objective' price zone (which is a violation or test of previous days low) during the first few 30-min bars -- because this is the most likely time of day, statistically speaking, for a 'key low' to be made (made first). if price instead 'starts high' and is trading down with momentum in B, the odds may be higher on 'Buy Day/High Made First' -- because this is consistent with the statistical tendencies. again, this is not a 'system' within itself -- its just some added information when trying to come up with your 'overall read' of the situation.

here is a quick glimpse from recent days. the arrows mark (with hindsight) the important high or low that was 'made first'. obviously, in real-time trading you will have to be making a judgment of whether its high or low first and then proceed accordingly -- your judgment might be right or wrong but there are other concepts to help guide you on that -- this is just to demonstrate that an important high or low is 'often' made in the first 2 hours of the day --- with particular emphasis on the first 30 minutes. the first 30-minute bar is very important (as I will show later) in that if a high or low is NOT made in the first 30 minutes, then the odds of a large move away from the previous days close drop in a big way -- and you need to proceed cautiously once price has moved AWAY from the previous close. As you can see on Thursday, the low WAS made in the first 30-minutes and price closed FAR away from the previous close, which is consistent. Note how on March 27th, March 31st, April 3rd, April 7th and April 8th -- price did not make a high or low in the opening 30-minutes and its as if there is gravity of the previous close not allowing price to close TOO far away from the previous close. One other thing -- a key question which is kind of market profile based but fits in here --- is the market showing 'urgency' vs the previous days range early in day (in first 30-minutes)? This is the 'don't get run over' indicator -- ie March 30th and April 1st -- and April 2nd and April 9th to a lesser extent. On these days, market is showing that it has no interest in staying near the previous days high or low. This price action is a strong indicator that something unusual is going on. The point of this comment is just to show importance of the opening 30-minute bar -- not only in its necessity to create a large move vs previous close -- but also interpreting strong action early in day vs the previous high or low as a 'Taylor signal' that an important high or low was just made. (Huge gaps do make this tougher to actually trade in reality).

I wanted to discuss one concept that I think helps 'quantify' Taylor. Taylors concept of 'high made first' and/or 'low made first' is quantifiable. If you think of the daily taylor rhythm in quantifiable terms, it become easier to understand, at least for me. From studying the time of days that the 'made first' occured (low for day if low first, high for day if high first), it turns out that the market makes its high or low for the day 45-50% of the time in the opening 30 minutes. This is a very powerful tendency. If you add in the next 30-min zone (the classic 'opening range'/'initial balance'), then its about 65%. Toss in the next 2 30-min timezones (first 2 hours of day) -- then you capture ~75-80% of the days. Thus, if you 'define' the upcycles and downcycles as movement relative to the opening 2-hour range, you have somewhat re-created taylors system for the intraday market. So look at this past week -- Tuesday made a 'high first' and a 'low last' -- the cycle was therefore down. This cycle does not mechanically 'switch' until a morning high is taken out. Now, price tested the Tuesday low on Wednesday for 'low first' -- and it did so in the opening 2 hours -- then the high was taken out for 'low made first' --- this signals that the market is in an up-cycle until a 'morning low' is violated to the downside. On Thursday the market gapped up and put in its low in the opening 30-minutes, which it will do nearly 50% of the time. I find this to be a very useful framework to stay in the rhythm of the Taylor cycle. The market remains in an up-cycle until a 'morning low' is taken out, at which point the intraday statistics point toward a down-cycle. This framework can be used within a buy day/sell-day/ss-day daily count -- or it can be used on its own. But either way, you know that you are on the side of the statistical tendencies if you use this as a 'mental guide'. Just my 2 cents on making Taylor concepts easier to understand.

closed my system early yesterday and was out last night so will post yesterday and today now. yesterday was very brutal --- I won a nice pot early in session and then pissed it away (and then some) trying to set up for the main range expansion move of the day --- I finally caught the directional short, but market just died in the narrowest range session in many months. so I never got to earn the lost blinds and antes back. kind of like waiting to flop a hand and then finally hitting your straight for a big pot, only to learn your oppenent was holding the same exact straight hand so no payday (ended up stopping breakeven despite a decent entry and an initial downward thrust). today (Wednesday) was another narrow day but I made a nice play early and led to a pretty decent day to wipe out Tuesdays loss and then some. still, this is a tough market, range-wise. looking forward to more 'normal' days ahead.

one more post: page 48 of TTT:

so true elovemer. here are all the 'close to close' changes for ES this year (ranked from high to low) -- you can see how its not so common to get huge point moves vs the previous close -- so the location for shorting today was tricky with the gap. in shorting, you would be implicitly betting on getting a large move vs previous close -- if you expected continuation down that is... a high violation/high test would have allowed much better location for the market to 'start high' and would allow more profit room to the same level/zone --- 815-820.... shorting was fine -- but had to keep expectations in check and respect the lack of continuation below the previous days low... we ended up right in the lower middle of the std dev bands at 9.75 pts (all numbers are in absolute terms fyi)

what do you mean Fan? what is going on with the moderation?