Session Start: Mon Mar 10 00:00:00 2008 Session Ident: TheLion 01[03:01] hello [03:01] hi 01[03:01] let me send you a chart of mine [03:02] put it on here http://imagebin.org/ [03:02] we all use that to show charts 01[03:04] http://imagebin.org/14761 01[03:04] who is we 01[03:04] people in this room? 01[03:04] who's server is this 01[03:04] uploaded 01[03:05] are you familiar with the ozfx method 01[03:05] it's on forexfactory [03:05] a friend of mine from undernet #photography put up the server 12[03:05] yeah i know ozfx its decent [03:05] great for 4 hour and daily 01[03:05] oh cool i was in that photography room before 01[03:05] :P 01[03:05] yeah 01[03:05] i use it 01[03:05] but instead of just using the 200dma as the filter 01[03:05] which to me is pretty stupid 01[03:05] :P 01[03:06] i use a variety of things 01[03:06] the ozfx is good for entries 01[03:06] but still... you need to be selective 01[03:06] after all, it's quality not quantity that counts [03:06] yeah exits are like 10x more import than entries 01[03:06] yeah 01[03:06] so i added the guppy multiple moving averages 01[03:06] multiple bollinger bands 01[03:07] and the Bands_squeeze tells me when the bollinger band is about to break out [03:07] what do you wish was in trade assistant? 01[03:07] when a pair is about to break out of the range [03:07] awesome oscillator on multiple timeframes? 01[03:07] yeah something like that 01[03:07] i still haven't thought it through 01[03:08] for example. this bands squeeze indicator is fantastic [03:08] well i gotta sleep, the baby is gonna be here soon 01[03:08] :P 01[03:08] i'll talk to yo unext time then 01[03:08] maybe we can come up with something 01[03:08] :D [03:08] think about it more though 01[03:08] sure thing [03:08] like, which would be most useful in short term and long term 01[03:08] aye 12[03:09] i have ozfx setup my own way too 01[03:09] i'll holler at you next time :P 01[03:09] oh you do? 01[03:09] lol 01[03:09] dont see it on your chart :D 01[03:09] "it's hidden" :D [03:09] its not on my chart, i dont use it most of the time [03:09] since i know what i'll find if i switch to the longer timeframes 01[03:09] k Session Close: Mon Mar 10 23:47:07 2008 ========= Session Start: Tue Mar 11 23:46:08 2008 Session Ident: TheLion 01[23:46] you have the macd set up? [23:46] yeah and squeeze 01[23:46] cool 01[23:47] now let's talk about the macd first 01[23:47] have it set to the right numbers right? [23:47] i'm tired man its almost midnight here 01[23:47] it'll be fast [23:47] yeah i got it set the way you said 01[23:47] ok 01[23:47] what we want to do is set it like your system 01[23:47] with top row showing the status and health of the bb squeeze 01[23:47] and bottom for macd 01[23:48] for macd, we will check to see if each timeframe's is ABOVE or below the 0 line [23:48] just give arrow for the macd on each timeframe? 01[23:48] in addition, we check to see if it has been increasing or decreasing [23:48] like, above or below zero on each timeframe with no distinction for decreasing or increasing against trend? 01[23:49] so maybe [23:49] ahh so you want only forming trend 01[23:49] if we're looking at the 4h macd, we check to see if it has been increasing or decreasing in the past hour 01[23:49] for 1h chart [23:49] on a long macd going downhill is bad 01[23:49] we check to see if it has been increasing or decreasing in the past 30minutes 01[23:49] so we can see the indicator 01[23:50] and see whether each timeframe is still maintaining the trend 01[23:50] above 0 01[23:50] or below 01[23:50] and whether or not they are starting to pick up 01[23:50] or slow down 01[23:50] by checking each timeframe's macd... to see if it has been icnreasing [23:50] and bands? 01[23:50] for example [23:50] the macd is simple i can program that, i know [23:50] but squeeze i never used before 01[23:50] see if the previous histogram bar is lower or higher than the current one 01[23:50] to determine that 01[23:50] and taht will maybe show the strength [23:51] i know how to program for macd, put that aside now 01[23:51] like how you use different colors or up or slant up or down arrows 01[23:51] k [23:51] focus on telling me about squeeze 01[23:51] for t he bollinger band squeeze 01[23:51] basically 01[23:51] here pull up a chart 06[23:51] * TheLion nods [23:51] i got one up 01[23:51] you should see that the lowest level bars are in dark green 01[23:51] that's really what we like [23:51] it even put up a pivot and support 01[23:51] we want to see if a timeframe is showing red or blue [23:52] yeah, the levels under the red line 01[23:52] you'll see price starts to break down when the red bars show up 01[23:52] and vice versa 01[23:52] right? [23:52] so red or blue, disregard anything anything else? 01[23:52] do this for each timeframe 01[23:52] no 01[23:52] we'll do something similar for the bands 01[23:52] like we do for macd 01[23:52] see here is what i'm not sure of 01[23:52] maybe we can see what works best 01[23:52] here are two ideas i have [23:53] forget macd for 10 mins, talk only about bands [23:53] i know macd like the back of my hand, i can write it from scratch in 5 mins, no joke 01[23:53] fist we check the blue and red........ this will be like we cehck macd's above or below 01[23:53] but to check the strength... 01[23:53] perhaps we can check if the bars are icnreasing 01[23:53] or decreasing [23:53] yeah i can do that 01[23:53] 'cuz the blue and red 01[23:53] only shows the start 01[23:54] and it could just turn green 01[23:54] but if it's increasing that's better 01[23:54] another t hing 01[23:54] or perhaps. [23:54] so its a bollinger bands version of the histogram of woodie's cci? 01[23:54] we can try how many red or blue bars it has shown in the past 10 historgram bars [23:54] at first its grey, then established its yellow, then its either red or green depending on trend 01[23:54] 'cuz sometimes a price move dont just move right away... sometimes it shows 2 blue, follow by a green, then 1 blue, etc 01[23:54] yo ukinow what i mean 01[23:54] so that's 2 ideas i ahve [23:55] does bbands redraw? [23:55] thats super important 01[23:55] the color of those histograms? 01[23:55] hmm [23:55] like if you close a chart with it on it, then open a fresh chart and put it back on, will it say the same thing? 01[23:55] not sure 01[23:55] oh no [23:55] check, because what you propose takes a lot of coding 01[23:55] it'll show the thing for each individual chart 01[23:56] and individual timeframe [23:56] there are 5 buffers to keep track of on say six timeframes, thats a 30 item array i have to program for 01[23:56] dont really worry about 5min down 01[23:56] 'cuz i've checked the 5min 01[23:56] and it's not really that great 01[23:57] so 5min up is great 01[23:57] up to 4h [23:57] ok screenshot it, then close that chart, open it back up and put a settings file on it 01[23:57] 4h,1h,30m,15m,5m [23:57] see if it look identical 01[23:57] or just4h to 15min 01[23:57] k [23:57] because if it redraws history i can't program for it 01[23:57] oh ANOTHER THING 01[23:58] this is a good determine of trend 01[23:58] and i think could be done really easily 01[23:58] so this will be the third row for the indicator 01[23:58] have the guppy multiple moving averages 01[23:58] check on each timeframe, when the trend changes 01[23:59] when the Fastest longterm MMA and the slowest Longerm MMA crosses [23:59] there is one for guppy existing already 01[23:59] i know 01[23:59] but we want the indicator to show when the longterm MMA crosses [23:59] you realize how much programming you are proposing? 01[23:59] :P 01[23:59] the guppy one should be ok. 'cuz we're just checking one moving average against another Session Close: Wed Mar 12 00:00:00 2008 ==== Session Start: Wed Mar 12 00:00:00 2008 Session Ident: TheLion 01[00:00] :P 01[00:00] but yeah 01[00:00] i know.. hehe 01[00:00] sorry [00:00] what does lime green signify on bands? [00:00] forget about macd, guppy or anything else 01[00:01] regular price action [00:01] if i can't get this in my head, the concept of bands i wont do it 01[00:01] basically the red 01[00:01] and blue 01[00:01] they will appear as the price breaks away from a tight bollinger area [00:01] the red and blue are main things we're interested in? 01[00:01] when the prices are consolidating 01[00:01] and they break away from it 01[00:01] and you will see bollinger bands expanding outward 01[00:02] from its contracting form [00:02] so i can actually program using bollinger that is included, i dont even need to use bbands really 01[00:02] no [00:02] is there extra math that bbands does? 01[00:02] just use the bb squeeze indicator you downloaded 01[00:02] you wont need the actual bollinger bands indicator 01[00:02] just he squeeze [00:02] he uses ibands and isqueeze 01[00:02] who [00:03] the programmer 01[00:03] what are ibands and isqueeze? [00:03] actual indicators 01[00:03] default indicators? [00:03] yeah 01[00:03] oh [00:03] but he's using a massive amount of arrays actually [00:03] i'll just grab from his buffers 01[00:03] cool [00:04] do sell and buy arrows appear on the price action when you see a new bollinger breakout? 01[00:05] oh 01[00:05] no 01[00:05] just do it like your TA or TRO indicator 01[00:05] and remember how we talked about strength of these breakout 01[00:06] check if there have been red or blue historgrams on the bbandsqueeze indicator in the previous.... let's say 5 bars or so mething [00:06] purpose being? 01[00:07] well say we want to know if the lower timeframe is still strong [00:07] 5 bars ago is useless to trade on 01[00:07] i know 01[00:07] well this is great for monitoring your existing ttrades 01[00:07] so you konw how strong the trend is 01[00:07] or whether or not you should get in later or sooner [00:07] adx does that [00:07] i'l try it though 01[00:08] but i find your adx, rsi, cci changes real fast 01[00:08] like suddenyl i'll see 4h good [00:08] its a weird indicator 01[00:08] 30 bad 01[00:08] 15m good 01[00:08] :P [00:08] he makes no distinction between sell and buy trends 01[00:08] the blue are the buys 01[00:08] reds are the sell 01[00:08] but we're not using it as a buy sell signal [00:08] right but visually its a poor presentation, he got lazy 01[00:09] hhe yeah 01[00:09] actaully let me rephrase it better 01[00:09] so say we're checking the bar 01[00:09] we first check if there are at least say 2 red or blue bars 01[00:09] to make sure we are still in a trend 01[00:10] however due to the fact that our current bar could be a green bar [00:10] dude with this indicator if you wait too long the trend is aleady over 01[00:10] but still in a trend 01[00:10] we check if the current bar is lower than the previous bar... if it is... we signal that it is a buy due to the amount of blue bars in the surrounding area, but becuase it's lower than previous bar... it's a weak trend up 01[00:10] and vice versa 01[00:10] and if we have a view of this on all timeframe 01[00:11] it'll really help a lot 01[00:11] same with teh macd 01[00:11] the macd is great 01[00:11] i've tested it a few times 01[00:11] eventhough we're not coding it as a trade signal [00:11] ok what timeframe is most important with this system? 01[00:11] but it's a great way to help you determine when would be the better time to enter or exit a trade 01[00:11] i would do 15,30,1h,4h 01[00:11] that should suffice [00:12] without 5 min you will fuckup your exits, i guarantee 01[00:12] hmm... ok 01[00:12] add taht then 01[00:12] it wouldn't hurt :D 01[00:12] so let me just recap [00:12] 1 min seems to be less useful 01[00:12] yeah 01[00:12] taht's what i meant 01[00:12] :P 01[00:12] so you know about this bbandsqueeze 01[00:13] first we check if there are surrounding blue or red bars 01[00:13] or if itself is a red or blue bar 01[00:13] amount can set to 2 01[00:13] i find 2 is enough [00:13] 2 what [00:13] your thinking is erratic you automatically think i know what is in your mind 01[00:13] 2 red, or 2 blue 01[00:13] sorry :P [00:14] if 2 red || 2 blue, what 01[00:14] in the surround bars 01[00:14] then we determine that as a trend up and down 01[00:14] based on red or blue [00:14] so wait till the 2nd red, assuming there was a first red? [00:14] ditto for blue 01[00:15] in the past say 5-10 bars 01[00:15] yeah 01[00:15] to see if we are still in a trend [00:15] then check the green histograms as long as they are uphill call that a trend 01[00:15] or if our current bar is a color bar [00:15] once downhill, cancel any signal? [00:15] ok stop your thinking is erratic, go in order 01[00:16] thelion... well maybe our current bar is a blue still..... but the second thing we check is to see if they are still going up or down... the histograms 01[00:16] yeah [00:16] look for 1 red, count 2nd red, assume trend is down, check green bars and display until they start going downhill 01[00:16] to see if current trend is still healthy [00:16] you're talking to a programmer, you have to present the information in order or its not programmable 01[00:16] ok 01[00:16] let me start all over then 01[00:16] it'll be fast 01[00:16] i'll talk in steps 01[00:16] ===================================== [00:17] look for 1 red, count 2nd red, assume trend is down, check green bars and display until they start going downhill [00:17] that is sell trend? [00:17] look for 1 blue, count 2nd blue, assume trend is up, check green bars and display until they start going downhill [00:17] since the indicator is unilateral 01[00:17] yeah 01[00:18] and we can check back 5-10 bars [00:18] you type about 500x more information than is necessary [00:18] ok, fact, you dont need to count backwards 01[00:18] and if our current bar is a blue or red..... 01[00:18] and it's still increasing in height 01[00:18] that will be a stronger signal [00:18] if you aren't near the computer you shouldn't try to enter the trend late anyway 01[00:18] since the height of the bar tells us how strong hte signal is 01[00:19] ====================== 01[00:19] and the macd is simple 01[00:19] just check if it's above 0 [00:19] i know macd 01[00:19] or below 01[00:19] but remember the second condition though 01[00:19] if it's increasing or decreasing [00:19] i told you can write macd from scratch in about 5 minutes [00:19] i'm not joking either 01[00:19] by checking the previous histogram 01[00:19] i know i just want to make sure yo uremember the second condition :D 01[00:20] and the third thing we check is this: [00:20] if you read my source code you know i do that already with everything i write 01[00:20] check if EMA 30 and EMA 60 is crossing 01[00:20] i use guppy mma 01[00:20] but basically these ar ethe two thing we need to check [00:20] its why people like my indicators so much, i consider history in all my multi timeframe indicators 01[00:21] to see if long term trend on each timeframe is shifting trends 01[00:21] so if ema 30 crosses the ema 60 01[00:21] on 15min chart [00:21] there are indicators for that already 01[00:21] that's another indicator for us that the trend might be shifting from downtrend to uptrend on 15min 01[00:21] and we will ahve this for all timeframe 01[00:22] so essentially we will ahve 3 pretty decent indicators to tell us if we are stillin a trend on all these timeframes 01[00:22] and we can use it to help us get a better entry 01[00:22] say i want to enter eurjpy today... and i trade on the daily chart.... i can use this system to determien when would be a good place to enter my short [00:22] you know that guy that wrote gmacd, he gives no consideration for history 01[00:22] gmacd? [00:23] yup 01[00:23] never used it :P 01[00:23] i've only used the regular macd :P [00:23] he only compares the current timeframes and compares them [00:23] for consolidation 01[00:24] that's stupid 01[00:24] :P 01[00:24] indicators are only useful if we can see its progression :P [00:24] you and i know that [00:24] not everyone knows that [00:24] i'm looking at his source now, he doesnt seem to know arrays and loops 01[00:24] hehe 01[00:24] oh hey lion 01[00:24] one more thing 01[00:25] the moving average check i talked about earlier 01[00:25] check if EMA 30 and EMA 60 is crossing 01[00:25] to determien if the trend is chanign 01[00:25] let's add a second condition to that 01[00:25] like oru other indicators 01[00:25] if EMA 3 and ema 15 crosses 01[00:25] that will tell us if the signal is strong or weak 01[00:26] so if the faster oen is on up [00:26] 3 and 15 or 30 and 60? 01[00:26] yeah [00:26] which? [00:26] or both 01[00:26] both 01[00:26] first we check 30,60 01[00:26] ema 01[00:26] see which one is higher 01[00:26] to know which way they crosses 01[00:26] if 30 crosses 60 01[00:27] in otherwords if 30 is higher than 60 at hte time 01[00:27] trend is still up 01[00:27] right? 01[00:27] now we check 3, 15 01[00:27] ema [00:27] you type about 500x more information than is necessary 01[00:27] however if 3 is below 15 01[00:27] it means the trend up is not as healthy as we would like 01[00:27] :P [00:27] if 3<15 && 30>60, trend is up but weak 01[00:28] and we'll just use the same way we display the strength for each one 01[00:28] yeah [00:28] if 3>15 && 30>60, trend is strong 01[00:28] yes 01[00:28] speaking like a true scientist [00:28] ok, from now on, try to think like that instead of 20 sentences which are detached 01[00:28] lol 01[00:28] roger that [00:28] i can't follow discordant thoughts [00:28] not while i'm trying to boil down how to program something 01[00:28] i understand [00:29] if 3<15 && 30<60 trend is strong down [00:29] if 3>15 && 30<60 trend is weak down 01[00:29] yeah [00:29] end of story, i can do that [00:29] ok so three rows presented [00:30] bands, macd and what i just said with the EMAs 01[00:30] one for macd, one for bbandsqueeze, and one for MMA 01[00:30] yeah 01[00:30] each with 2 conditions to check each 01[00:30] for the strength obviously [00:30] yes history is very important 01[00:30] what do you think [00:30] i can program that [00:31] EMAs for the moving averages right? 01[00:31] think it'll be effective? 01[00:31] yeah 01[00:31] playing around with it 01[00:31]