Ken, Thanks for thoughtful reply. Basically you are adding such a heavy discretionary component that backtesting is difficult. I don't have a problem with that, and as you say, results are best confirmation. Good luck in 2002.
Three and Candle..........I tend to agree that Stochastics are useful in a range using Overbought conditions to confirm a test up to high end of range for shorts & Oversold to confirm a test down ....... I disagree that they are useless in a trend......if yoru shorting a overbought stocastic in a uptrend , you are right....its harmful..... BUT using that Stoch when goes oversold in a uptrend on a pullback maybe with a bullflag as confirm or a 38-50% retracement to confirm its perfectly fine to use & a valid indicator....... George Lane said that stochastics should be used to confirm only & not a stand alone indicator.....see chart as example for what I mean......... I used a example from my site becuase I dont have another in front of me to show you. But this is the BEST way to use a stochastic in a trend & is a valid idea. Chris
Ken.....i cannot belive you said that..... You test your techniques in your chatroom ? I hope your members dont know they are used at guinie pigs for your testing of your ideas..........maybe test them on paper or do it with your own money BUT not in your chat so you dont put your members at risk ......I would never test anything in my chatroom as it would put my people at risk & its unfair to them...I test with my own money & not anybody elses. Chris
Agree vadym ... I had first been exposed to stochs in a different room, and it 'sounded good' when the op did it, but upon testing it myself, I found far too many false positive signals being generated. re testing, what I mean is, if anyone wants to test out my ideas, they can see the results and performance for themselves in the live room, I stand behind everything of course .. chatrooms are for learning in a live trading environment, not making actual trades from, imo.. I've tested the techniques out over the last couple of years extensively and have found that they produce the most consistent results for me, so I share them with others in the live room.. Stochastic crossovers are in the secondary group of signals I may use as a red flag to exit a position once in, much like a 1-min red candle doji after a volume run, or declining volume, or a 50/50 xmas tree time & sales tape etc.. something that may help me to decide to exit a position, but not an entry signal.. overall I've found that simply doing the 2day high breakouts/ 2day low shorts is the most consistent technique... the intrarange and fade trades are much trickier to get right, if I were to "do it all over again", that would be my advice to myself, stop bottomfishing and be a breakout trader instead. Today, 12/26 provided a bunch of good opportunities to test out this technique, here's AMAT for example: http://www.dtu1.com/amatdec26.gif It's a relatively simple, but effective pattern ... that's the most important pattern I want to be able to share with everyone, it seems to produce the best results. It's not 100% of course, a good example of how I'd likely be stopped out even/small is in today's VRTS, eg: http://www.dtu1.com/vrtsdec26.gif the point being, even when the technique produces stops they're likely to be small, eg even or .2 max .. As I have both time and price value stops, an entry at 45.5+ here I would've stopped myself out on at 45.4 for a close to even trade, I'm personally not patient to have waited for the large later-in-the day move in it (10 mins is long to me), I like the amat style trade, eg a decent +1/2 or better fast clip and out, ready for the next trade. check out today's charts, all the basket stocks I listed, you'll see that the technique should've produced +3/8 or better wins for the majority of them .. it would be good to refine the method, look for confirming indicators that others may come up with (ideas, traders?), personally I like the trinq, time & sales balance and sector relative strength ... today's a good example, COMPQ produced a 2-day high signal when it cleared the 1954, for a nice sharp run, stalled at 1970, a good first exit area etc. ken agree chartwiz, stochs not a standalone, a secondary reversal exit type indicator, just so-so in my book, i like time & sales as the best realtime indicator, also i like exit once I'm in a position and see a 1-min red candle doji, if the tape is 50/50: http://www.dtu1.com/nvdadec26.gif
Ken, Good examples. Very surprising to see a losing trade displayed by a vendor. Might be a first. Particularly one where you could have fudged it a bit, not that any other vendor would have done that. Hats off to you. How do you select the basket stocks? Do you look at their patterns on the daily charts or are you only looking for good volatility and liquidity?
Ken I was wondering what happened to your stochastics on your charts. lol. I agree with Chartwiz's qoute from Gearge Lane "Stocastics are not a stand alone indicator." There is no stand alone indicator{except for price} When you find one let me know. lol. I also find stochastics quite usefull and have done quite well incorporating them into my charts. I find myself using them more for exiting trades then I do entering them. I will sell when in overbought territory and cover when oversold. When entering I only use them as one of many confirmations. Great thread guys, thanks for sharing the info, charts, ideas etc.
Chartwiz, I think it was pretty clear Ken meant he considered the results of his room to constitute backtesting, not that he tried out new ideas on his students.
Actually, this might be good indicator of honesty and real purpose since any hohest room leader with intention to educate rather than to brag and solicit would show examples of stop taken. Two kinds of such examples, both useful: first, when stock continued lower, to reinforse necessity of stop, and second, when stock reversed after hitting stop level, to reinforse the notion of market being the game of probabilities. Vadym
excellent points all.. re losing trades too, sure i make em all the time... just like winning trades... i wish i was 100%, but i'll never be, like tony says, you have to be able to take stops, to trade.. that's life.. a personal goal is to learn how to start to make larger trading profits by expanding share size to the 5K+ type blocks... I need to build my skill in this area.. anyone have any tips, eg how to breakthrough to feeling comfortable swing large size? that's how 'market wizard' wealth is created, i'm not there yet, hope to get there in the next couple of years. open to ideas... I still classify myself as an intermediate-level trader, hope to learn more about genuine size trading, the next step.. since I'm extremely risk averse, limiting share size is one way I do this, though I limit profits as well. it's a 'traders' block' for me, for me to breakthrough i need to be ok with 5k/10k share size trades.. ideas? ken i agree too with you all, that we should be hard on/skeptical of vendors, myself included, the best thing is to test out the ideas, see how they work, if they help you become a better trader, then great, if not, look for someone else to work with. it's a good point too, that different people respond to, and learn from different trading techniques individually, so Experiment, go try out everyone, that's what I did, it's good to read all the books, see all the sites, then look for things that, at least, they all seem to agree on (eg something as basic as 'keep small stops') , and then look for what matches your own unique trading needs and style. and check em out, test it, see if you can duplicate it yourself, and try to move on, become independent. thanks all (hey i gotta get off the pc, it's holiday time where's the eggnog? lol.
Ken, good commentary, but I do have a question regarding your first setup, AMAT..... You say only "entry long .5 to .6 over whole number above previous day's high But in this AMAT trade you entered long @ .25 within the whole number. Yesterday's high was 40.35, but you entered the trade @ 40.60?? Why? The actual entry would have been @ 41.50 or so, would it not? (.5 over whole number above previous day's high) Just curious as to why you seemingly broke your rules on this one? ProTrader1