darthtrader2.0

The problem with ninja is it can be real time event driven, but it can't be histrorically event driven...that should change in NT7.

No its actually a function of time, no matter how you cut it. Its not hard to argue that HDF5 is vastly better than SQL for storing tick data...its a non issue. There is a reason we have the concept of time series and relational databses, outside of chance. It is plain and simple. IF the time put into HDF5 that can make it better than SQL though, is the real variable then do it, if not, then don't. That doesn't change the fact that SQL is not built for high freq time series and there are DB that are. you have no real point

No, both you and blowfish are missing a very naunced point to this whole thread of ideas. That problem is in the std dev calculation being put forth here, because...the above statement acts as if the distribution has no effect on variance...comeon. Well if you break it down, if you "deviate" in a "standard" way, then you are calculating a variance away from a central mean. I agree, VWAP is probly a great way to model the central tendancy of an unknowable distribution. If your deviations though are semetric from that mean? What else could that possibly mean other than that your backtrading a normal distribution? Which anyone with half a brain knows is bogus with the markets. This strikes me as stats 101 vs probability theory...I'm self taught though , maybe I'm way off.

which you should add is one std dev of a normal distribution, which is assuming a normal distribution of price/volume, which you admit yourself is wrong.. First, I'm not argueing that you are a baddass trader. you obviously are.. I highly suspect though your "baddassness" has much more to do with years of experience with price action and order flow and your discretionary ability to negate trades than this setup you put forth has to confirm them. Maybe I'm right, maybe I'm wrong..but I dont' take anything in this game at face value just because someone says its so. Maybe we are both right and the assumption of a normal distribution is good enough but some other probability distribution is slightly better... why not try to find that function since we know a normal distribution is not optimal, but maybe that is foolish.. I do know your stuff doesn't backtest, so something else is going on..

Well again, that assumes that the VWAP is the exact average in a normal distribution of price/volume and this is what actually happens in markets. The entire idea of a "composite" seller/buyer is bogus...it negates the entire idea of a "trend day"...

Well hopefully I haven't bit the hand that...ect.. But, Jerry have you ever messed around with trying to compute the cash VWAP for the underlie of the derivative your trading? From my research, I just have fundamental problems with the VWAP on futures. I really question with the depth of something like ES along with what the ES functions as in the market that the WVAP matters that much on a futures contracts. However, on a single security that makes up an index, big institutions have some serious liquidity problems so that then puts a natural regression towards the mean on any single security, a natural regression towards the VWAP that is probly not there on the equity futures or is at least highly distorted from hedging/arbitrage. Basically, a tick precise, cash weighted agragate VWAP I suspect would be pretty baddass..

The full title is.. The Red Queen: Sex and the Evolution of Human Nature Ok this text has nothing to do with trading per se, but if you want to understand the context of why you would spend countless hours looking at numbers on a screen with your life, then I can't see how you put trading into a better "grand life philosophy" than this text. I don't have any kids yet, but if I had to give my son one text to understand life, it would be this text hands down. Its not a pretty picture, it probly is an evolutionary pyschology and biological reality though... read it at your own risk...

This thread is inspired by david22's post about neoticker in another thread and building internals for non NYSE markets... I know this is neoticker's game and this is my last ditch attempt before I just buy neoticker... I blew away one of my machines the other day and played around with the neoticker demo today but I just don't like the program. So the first question is, has anyone here ever messed around with multi instrument strategy plots in Ninja? I know currently it is possible to construct some form of the tick client side from the datafeed with onmarketdata() and strategyplot() but I do not know if Ninja/our machines/memory can handle creating the SPX in real time..I'm pretty sure not... NT7 is promised to have multiple instrument indicators so I'm on hold with neoticker until I see what that can do. How many instruments that will support in real time is the key...While I do not expect to be able to calculate the S&P cash market from the individual stocks(even though this would rule because it would gives us a real time, tick precise PREM) there should be a work around as far as pruning the list down from 500. A simple correlation between the move of security X and the S&P cash over a good sample size should do the trick there with a cutoff point, I suspect the list is much smaller than what we probly think. The other reason then I want to stick with ninja is its not that hard to override plot()..OHLC and candles are simply a terrible way to visualize the data of TICK...we already know the bounds of the data, a distribution at N time makes alot more sense.

Well, I have to admit I'm in Mea's camp. While I love Steenbarger's work taken as a whole, I find even his more specific pyschology stuff to be a bunch of wishy washy nonsense. I do know I'm a bit biased though as I don't respect the field of pyschology much and believe Freud to be the biggest intellectual fraud in human history, Skinner to be outdated and wrong, Jung absurd...ect, ect... To me the best "trading pyschology" text is "Judgment under Uncertainty: Heuristics and Biases" by Kahneman, Slovic and Tversky. To me though the ideas in this book have little to do with live trading and more in how you come up with setups. You should at least be aware of the hardwired heuristics and bias you were born with, then maybe you can plan ahead in your research to combat those. To me a huge cultural bias that should be combatted is that I believe everyone finds this game with the idea of making tons of money quickly. When traders "luck" card/capital runs out, they leave the table with ideas of not being "good" enough, trading is bogus, ect...as opposed to coming to the table with the idea of spending decades sharpening a skill to razor sharp perpection. The Dalai Lama would not make a good trader if he is trading a system that simply doesn't work. I don't care if he can pretend to visualize it over several lives... The biggest thing I don't get with "trading psychology" is why very obvious things are not mentioned instead of this "new age" garbage... If you don't feel creative, then learn a right brain craft...painting abstract art, learn to improvise jazz, learn to sequence original music on your computer.. If you don't think probabilistically enough then get some text on probability theory, get good at other probabilistic games.. If you don't think left brained enough then take a math class at community college, learn to code.. If you take too much risk trading then take up a hobby that is really boring to an "adrenaline junky"... If you don't take enough risk, take up skydiving.

that is ingenius..only a sucker would trade off time and sales after watching that.

Getting back to the title of this thread. I opperate from the standpoint that most traders are not fundamentally undercapitalized, but undercapitalized in comparison with the spread and tick size of the instrument they are trading. To me to truely quantify a market edge, you would need to have an infinite player game theory model, otherwise you will simply not know how your setup is effecting the other players. Obviously, that model is null, so you have to accept a random element of chance in your setup along with every other player in the game having a random element to what they are doing. Scaling out is a probabilistic good bet if you can stay in the game cheaply while holding on "luck" wise..If someone makes me a bet that I know will only happen 1 out a million but it cost me nothing to get in/out, i'll put on the bet as much as possible. From what I see most traders do the opposite of that. They risk far more on the spread/tick size, ignoring the randomn element of trading as far as being able to use many many "tight" wrong bets, just to capture the big move and "be right", "probability wise" with a "bad" bet. A bad bet is still a bad bet, if you get lucky or not.

BlowFish, have you read Evidence Based Techincal Analysis yet? If you have not you should get it ASAP. All it really does is put forth the case for starting from a mindset that everything we talk about does not work and its our job to prove to ourselves that it does. If you think about it, this is the exact opposite of what anyone who buys a trading book does. Someone buys a trading book on fib trading because they assume it works, but that obviously leads to delusions and confirmation bias. "price hit this golden ratio today so obviously this fib stuff has utility".. I've come to the point that I believe a good trader could actually be profitable trading off a random level generator...that setups are simply ways to filter out trades so that a good trader can focus on price action.. As far as market profile, I do believe there is something to low volume zones..if no one wanted to transact at a level 5 minutes ago what has changed that they want to transact now? I do not believe though our market profile tools do an optimal job of quantifying the difference between a low volume breakout and a low volume fade...the VAL/VAH is just nonsense to me...I would bet they are not better than random. They are no different than pivots, pivots have nothing to do with reality..the fact some people can trade them is simply because those traders are damn good at reading order flow and price action and use pivots to filter out what not to look at.. Dalton isn't Jim Simmons, and Jim Simmons is never going to write a book. To me the more interesting question is how do we build a better mouse trap with market profile? One thing I can think of is that the bid/ask and book have to act differently when retracing near a low volume zone than when price established that point as a zone to start with. What that looks like is a very interesting question.. How to do that without moving to neoticker? I don't think it can be realistically done..

Well first I should say that to me Mind Over Markets is the best trading book ever written, if you throw out or don't get into some kind of "religious" mindset about ideas. No one has ever done a better book on how markets breathe but that doesn't mean its exactly right, its moving into the realm of mythology at this point. The fundamental problem with market profile is assuming a normal distribution of either time or volume between the high and low of period X is worse than wrong because if we can say anthing about the market at all its that between price X and price Y, a normal distribtion of anything has nothing to do with reality. To me the only thing with can objectively say about the market is between X and Y, the probability distribution given enough samples will exhibit fat tails, and that distribution is random enough to not be quantifiable in any real way, but its not exactly a stochastic process either. If you try to get too solid about it then the stochastic element will get you, if you try to get too stochastic then the solid element will get you. To me this points to an objective view of trading in that trading is damn hard, you probly want to do something other than what anyone says to do on message boards/literature/anything you can get second hand, liquidity becomes issue number 1 and how to "surf" waves of liquidity.

any chance you could do a video of this thing and show it moving in real time? I remember running across the picture on CQG website awhile back, but I kind of assume static pictures don't quite do it justice. Its basically the tape with a book level component right? Also,you ever mess around with these delta kind of charts? Also note the above chart is what I ment at the start of that last post as far as postive/negative delta on the opposite side of the move. Maybe a move is a bad word but that chart would almost be more usefull of the deltas were flipped. Why that is...I haven't a clue besides for the fact that trading is a damn hard.

its actually interesting too if you look at a market delta chart, because on almost all moves all the time the top of the move is met with very strong positive delta and very negative delta near the bottom...this somewhat negates the usefullness of the market delta chart but thats a tangent. One thing to keep in mind is that I'm not sure how much importance should be given to above or below the bid/offer. I kind of wonder if that is simply our software/datafeeds not keeping up with the millisecond execution speed of the algo arbitrage guys. I say that because usually when I see that on the tape its at the break of a critical level when you see the "flood" on the tape. For me, I didn't get much from the tape until I started to ignore the individual tick volume size and view the tape more in spirts of trades going off at the bid, trades going off at the offer and how that is moving price. Thats why I don't understand putting a filter on the tape. All your seeing then is the large discretionary guys like what James was doing at the bank but your losing all the algo guys with insanely fast execution who can print 50 1 lots with a fragmention algorithm in a few seconds instead of printing a 50 lot trade. I'm still pretty new as far as the tape not looking like noise but a few patterns I look for are: a. the poor sap who got sold the bottom/bought the top print of a move..depending on how fast the tape is moving this print is usually put up sequentially if its more than one print, bang bang bang. You can even see this sometimes on SPY(SPY to me is great to practice on because its so insanely fast, which I assume is because its such an arbitraged instrument) If you see that, then what I'm looking for is I don't want to see the opposite side of the bounce bid/offer wise being hit with a stream of prints but not moving price very far away from the "sap" prints. To me that says that although there are participants that want to do business not far away from the "sap", they may very well be the next level down of "fools" if they can't move price much and are just causing a distribution instead of an accumulation that actually moves price away from the "sap". b. rejection of the breakout, pullback, the same kind of distribution as in A on the opposite side of the bid/offer with price stalling. Currently, if I see that I'm going to take it and front run the breakout, if that is a good idea or not I can't really say yet without more trades. So then if price moves back to the breakout level, what I'm looking for on the tape is for a breach of the prior "sap" print, how it is transacted doesn't matter in my current thinking. Usually, from my experience when that happens if your clicking a mouse you won't have time to do much before you get the "blast" of prints on the tape and probly a price move..Thats usually when I see prints above/below the bid/offer..Since if I'm in at this point I've already front ran the breakout, I'm gone at this point. I'm sure in time I'll hone some patterns on the tape as far as what to look for to probly add to my position but I haven't a clue there. To me what is interesting about that pattern is someone who is using buy/sell stops to play a breakout is buying/selling the otherside of my trade

Personally, I think the whole trading plan thing has got a bit out of control and leads to ideas like this. Somewhat akin to renting an office, hiring a staff, then having a conference on what exactly the business is going to do. A business has to start from an idea that can be exploited to make money...then you add the structure to help that idea grow. If you don't have the idea though, then you have nothing.

I wouldn't worry too much, Mr. Duffy isn't blowing smoke up your ass there. The CME didn't cause any of the current problems, OTC derivatives are the problem, CME is the solution to the problem. I imagine alot of hedging strategies need the leverage that a 5k margin provides..all it would do is hurt the CME and open the door up for an exchange to offer lower margin futures outside the US. Worst case I imagine is a few cents a car tax...raising margin requirements by 500% is just not logical.

My problem with all this stuff is that its viewing the micro structure in a form that simply has nothing to do with reality at this point. With fragmentation and millisecond algo execution, the size of of a print contains literally no information about the size of the capital behind the trade. The ironic thing to me with all this is a 30 lot YM print is probly "dumber" money than alot of the machine gun single prints, because alot of the single prints are guys like Rentech doing their thing. I really can't see how it makes sense to operate from the idea of "operators" or "Syndicates"...we don't trade in a world of Goldman and Rentech colluding to take traders on traderslaboratory.com's money..We trade in a world of Goldman, Rentech and a bunch of other 5 ton elephants trying to take eachothers money, we are all just in the way of that battle and either get stepped on without notice or pick up a few scraps off the table without notice. Even if everyone on this board pooled their money, our "syndicate" would get utterly destroyed if it tried to play games with the market.

I'm been thinking about adding something like this to what I watch but just for the opening range to try to watch out for trend days. I think if you do it for the whole day, a broader view of things might have more utility because I'm not exactly sure what information you are gaining from knowing that the current volume from 10:45-11:00am is higher than some windowed average of prior days at that exact time. Some good data points to me would be 8:30 - 9:30 8:30 - 10:00 8:30 - 11:00 After that to me your getting into a higher/lower overall volume day. If a genie looked into the future and told you monday would have higher/lower volume overall on the day than the average of the last 20 days...well I really don't see how that information would be of much use.

I still don't think thats what I'm talking about. I can't believe that "good" would be part of that since its so illogical to fit in there, one has to take bias though that this was during the dot com era . I was refering to an idea that you just throw as many ideas at AI as possible and hopefully because of the number of samples something "good" emerges. Total tangent, nothing to do with trading software if you want to nit pick because your self interest sits in software engineering and robust applications....:þ

Mandelbrot found his patterns in the clouds in futures markets yet no one has been able to make pragmatic sense of his ideas since. If you really want to get into modern reseach your best bet is to include either the words "stochastic" and/or "bayesian" in your search, better yet both. "game theory" is also a good key phrase but we are not even close to computing power needed to game markets, but its still an interesting idea.

Go back to the day you started this thread and compare it to today.... How much more could you possibly be wrong about shorting vega? You did not do anything wrong as far what you know...the problem simply sits in what you do not know. Frequentist stats and probablity are nonsense. Markets are Bayesian, I've bet my life on that idea. If you want to read about stochastic process being ripped to shreads then get E T Jaynes book. It is literally the bible if your interested in the "quantification of market data".

As are fibonacci nonsense or a random number generator "predicting" support and resistance. Who cares what you call it if the end result is no better than random. Noise is a pretty good term. If you want to take statistical assumptions off the table there is really a more fundamental problem with assumptions of the VWAP of futures here: Who is hedging other bets risk wise vs who is taking outright positions to add to risk is unknowable?? You can throw all the statistics you want at the unknown, yet even if I can find a mathematical relationship that defines that I can see a dog in the clouds, no matter how abstract and probabilistically informed it is, does not mean an actual dog is in the clouds. Betting on what a fake dog in the clouds is going to do is a clear logical absurdity. You can not seperate the two variables there so we are basically both talking about horseshit, that has little to do with what we are actually trying to argue about. The difference is, I at least understand this perspective, while you do not. Maybe this perspective is wrong, maybe it is not.

nice! how accurate was the assumptions in frequentist black scholes nonsense on vega in what you can price now on the instrument compared with when you posted this? oops...if vega doesn't make sense all the other greeks are just random. I didn't come to this via a stochastic process. A random number generator didn't just pick me to most this.

Well its 2:53am and I'm drunk...maybe I can be clearer this time. I take issue that you are not assuming a gaussian distribution...what distribution in your code is the code deviating from in this thread if not gaussian? When it gets down to it there is a far more fundamental problem here that has to do with a fundamental and not computable moving nash equilibrium price intraday and the probability distributions above/below that equilibrium in order to quantify a bet. You can quantify all day on a random variable, however if you are not quantifying distributions on the "correct" variable, I don't see how that differs from straight out single agent/brain/heuristic "guessing".