What creates spikes?

Discussion in 'Trading' started by k p, Mar 3, 2015.

  1. k p

    k p

    Yes, this is correct. I meant to say that I would sell after a support level broke, hoping I'm getting in early on the down move, only to see that I sold on the lowest tick! :(

    I see, thank you for the clarification. I guess I was thinking along the lines that if you are coming down to support, you almost want to buy at the lowest tick, and if price should happen to break through support and stop you out, then so be it. If you wait for support to be confirmed, price will sometimes reverse very quickly, so waiting for a bit of confirmation might cost you a couple of points by having to buy 2 points above the support level you were expecting to hold. So then of course you have to figure out if by waiting for confirmation increases your win rate sufficiently that it offsets the higher price paid over the times you do loose when you buy, and price just keeps dropping.

    In your example where you said your 4 tick stop wasn't hit, since the low of that down wave was 2085.25, you would have had to buy no higher than 2086.25, and seeing what is happening there, at least on the 5 second chart, makes it certainly seem like you were catching a falling knife! Now you clearly sound like you know what you're doing, so I don't mean to imply at all that you were just throwing the dice here, and price certainly did stop dropping, so I'm only saying that based on what I see with price, there is no indication that its going to stop, other than if a clear level of support was right there, or you have some mathematical/statistical framework that tells you when and how these down spikes stop. (ie. If you know most spikes are 3 points wide, then you can start buying once price has spiked down 2.5 points, and if price keeps dropping after dropping 3.5 points, you're out with a loss of only 1 point. If you gain 3 points for every time you lose 1 point, this has to work only 50% or even slightly less to be profitable)

    In some respects, given that I had the ability to always sell on the lowest tick, or buy on the highest tick, I could have used my emotional trading in reverse to put on really good trades! :)
     
    #31     Mar 4, 2015
  2. k p

    k p

    Sorry i960... I'm not sure what post you're referring to here...
     
    #32     Mar 4, 2015
  3. i960

    i960

    Too early. The market likes to do things more than once. You need to watch for the reaction a bit, then watch for the reaction to that reaction. IMO you should also have a DOM up around the same time.

    As far as thinking about it as a falling knife on that guy's ES play, classic case where lines could actually help you.

    es_2086.png
     
    #33     Mar 4, 2015
  4. k p

    k p

    Ah, that does look nice, but I have a problem with diagonal lines because often, you can draw them in a few different ways to make them do what you want them to do. In this case, the spikes down are pronounced, so they do stand out and hence make them less ambiguous, but horizontal levels I find more objective, not so subjective as I find diagonal trend lines tend to be.

    This was one of the issues I had with trying to learn SLA as per DbPhoenix. In real time, drawing the supply lines and demand lines was a bit more tough than how I would do it at the end of the day. And by the time its obvious which of those swing points ends up being the major one you need to use to connect your line, that trade is long gone. Swing points are of course more telling because they are an actual level beyond which nobody wanted to buy or sell for that time being.
     
    #34     Mar 4, 2015
  5. Right. So think - what else can I use? What geometry exists that I can coordinate? What is static, and always present, and how can I coordinate these areas to profit from?
     
    #35     Mar 4, 2015
  6. k p

    k p

    Well, having learned from DbPhoenix, I'm not so much into trading patterns but rather following demand and supply. Saying that though, if I absolutely see a geometry that allows for a proven statistical edge, I'm more than happy to be the "monkey in a cage pulling the lever each time to get a treat". (ie. If I risk 2 points to make 5, and I'm right 50% of the time, and I see this 3 or 4 times a day... sign me up!)

    But to answer your question, I need a bit more information as its too much of a riddle at this point and I was never good with riddles. What is always static and present?
     
    #36     Mar 4, 2015
  7. monoid

    monoid

    That is one way of looking at the trade. But this way of looking at the trade implies that your decision to enter was not made before the level is hit. You are responding to what is happening after the level is hit. On this particular trade, I made my decision to enter when my context was set in motion -- this was based on the information I was able to glean from what was available at that time. What happens after the level is hit, I care only to the extend that the price action is conforming to my expectations. If it does not conform, there is no trade (nor do I reverse my opinion). If it confirms, which it did in this particular scenario, I enter the trade as per plan. I really don't care if price falls like a knife or a big axe -- the context is there, my conformations are there, I pull the trigger. Actually, I want it to fall like a big axe 'cos that is my expectation!

    If NoDoji indeed said you were looking for a perfect trade, I can see why now! :) Look at it this way: If you take two successful traders their methods of analyzing the market are going to be most likely different. However, in many cases, they will arrive at the same conclusion. Markets move because many traders' conclusions converge. No one trader knows if they are going to be right or not when they enter the trade, but where there is a quorum, the trade works. So, what is important is not trying to find that perfect setup, but having an objective framework that you can use to understand the market. From that understanding comes your trade setups.

    All the best.

    Regards,
    Monoid.
     
    #37     Mar 4, 2015
  8. k p

    k p

    So let me see if I have this straight. First you have a plan for what you need to see as price is coming down, and since you are looking for a long, you will need to see something that tells you that demand will be stepping in. When price hits this level, you either go long because you saw what you need to see, or you don't see it, and hence no trade, correct?

    I find that what I'm looking for at the moment isn't as advanced. Once I see a range, I want to place a trade anticipating the rejection. Yes that range will eventually break, but if the stop is only 2 points, and the other range extreme is 10 points, a R:R ratio of 1:5 means I can be right even just 50% of the time and still come out ahead.

    The trouble is that I'm just not coming up with any filters for what I need to see as price approaches this level, and of course, once price bounces off, then trying to enter is chasing. Sometimes price doesn't even get to this level and turns a point or so before even hitting the level. But given that this idea about needing to see what I need to see in order to initiate the trade sounds a bit too much like thinking, and any hesitation on my part will always get a worse price, then entering in anticipation and getting the best possible price seems to be where I'm at. Most levels, if important enough, produce at least an initial bounce. If price comes back to you, getting out at BE is often possible, but the best rejections happen very quickly. And if a quick penetration should happen.. so be it. A 3 point stop on the NQ seems to cover most cases where price rarely penetrates more than 3 points if it will still end up reversing. 2 points I see was a bit tricky as sometimes you'd be stopped out with a 2 point penetration just before price comes back into the range.

    So as you can see, I'm just trying to focus on the odds/stats right now, but it sounds like you have way more filters that tell you actually in advance whether you will put the trade on or not as the level is approached... is this correct?
     
    #38     Mar 4, 2015
  9. Well, I'm not in the business of giving away years of my or others' work, lol. I can't answer your questions; I'm simply posing them for you to ask yourself. Think about what you have at this moment right now to ID current market conditions. Do you even have that? I don't think you do. Can you ID a trend v congestion? Can you ID termination points that will give you a highly probable & low risk entry? If not, are these things that you want? If so, then go find them. I would query every word in this message with regards to trading and see what you can come up with.

    You really have to be a good detective to figure this stuff out. When you write things like "Yes that range will eventually break" - well... how do you know this? You don't know if it will break or not, so you can't base a decision on a trade with that. You're looking at a branch and trying to guess how many trees are in the forest.
     
    #39     Mar 4, 2015
  10. k p

    k p

    If you have someone else's work... they must have given it to you... so pay it forward! LOL

    I do understand what you say. At some point though, I risk falling into the trap of looking for the perfect trade once again, or having such stringent filters, that I could sit for days without a trade presenting itself.

    When I say the range will eventually break, what I mean to say is that sometimes the trade won't work, and it is therefore a guarantee that the range will eventually break. I might not know which which end price will pick, but if I was to keep fading the extremes, it won't be long until one trade gets stopped out. Plus, the idea isn't to know if the range will break or not, its taking a trade regardless, and having a plan for if you're right or if you're wrong and hence need to get out.

    Anyway, yes, high probability with low risk is the ticket, and the phrase termination point I haven't heard before, but I certainly imagine this to be a major swing point, which does once again equate to trading the extremes. So at least I'm on the right track.
     
    #40     Mar 4, 2015