BlueHorseshoe

Sure: Bidsize( ) Summary: Quote field returns the number of units offered at the best bid. Cheers, BlueHorseshoe

Why not try calculating R:R ratios like this: (Stop*Probability of stop being hit)/(Target*Probability of target being hit) BlueHorseshoe

Didn't you quote Ed Seykota in another thread - "everybody gets what they want out of the market"? I think one of the difficulties here is that discretionary input tends to fall into one of three categories: Pure 'instinct' or gut-feeling. Decisions based upon a set of rules that the trader is unable to formulate consciously, but exists within the neural structures of the brain in an objective way. Decisions based upon a set of rules that the trader is consciously aware of, but is unable to formulate with sufficient (in)accuracy to describe them to a computer. The first category is a bit like gambling, and where anybody is consistently successful with it in a way that defies statistical probability, one has to assume that their decision making falls into one of the other two categories. The second category is, I imagine, very frustrating - it's a bit like the trader being their own black box and unable to peer inside their own head. Not knowing, in spite of profitability, is not something that many people find terribly comfortable. The third category is the most interesting. It stems from the fact that the human brain is able to recognise what Wittgenstein called 'family resemblances'. Getting a computer to do this is not easy, but is probably a neccessity if a trader wants to remove human discretionary input altogether. Hope that's a useful way to think about this. BlueHorseshoe

No - if you look more carefully there's a 'u' in the middle - it's 'f**k u angle'. Very mystical . . . BlueHorseshoe

Open Interest might give you the answer to this question, but it's not something I've ever used, so I'm unsure of the granularity of it. BlueHorseshoe

Hello, There are various possible answers to your question . . . Firstly, what 'price' are you talking about? You seem to be charting the ES. The ES/S&Ps track 500 stocks, whereas the $TICK symbol tracks, if I remember rightly, over 3000 stocks - pretty much everything that trades on the NYSE in fact. There is no real reason to expect on to perfectly follow the other. Looking at your chart, I'm not sure if your question is more literal? The $TICK is an aggregate of said stocks; therefore it is banded (rather like an oscillator such as the stochastic or RSI). The maximum value that it could possibly assume would be that of all the stocks it contains (so something like 3000). Similarly it has a minimum value. The ES doesn't have a maximum value (though it does have a minimum). Remember, the $TICK is just the number of stocks that are ticking up or down, not how much they're up or down by - the market obviously reflects how much they're up or down by. There are other complications connected to how the $TICK is calculated, but I can't remember them. There was a great Steenbarger blog post on TraderFeed about the $TICK though - try googling for it? Finally, I'd be very careful about how you use the $TICK in your trading - I've tried to build all sorts of strategies around it and never found anything that showed much of an edge. Hope that helps, BlueHorseshoe

Shouldn't these market conditions then become incorporated into the rules/principles? Nevertheless, I see your point - it certainly makes more sense to cease trading based upon a self-awareness of internal psychology rather than some arbitrary '3 losses' cutoff. For example, on a day of three consecutive losses following a prior day of three consecutive losses, one's emotional state would generally tend to be more impacted upon, I imagine. I don't daytrade, so maybe I'm not really qualified to comment. BlueHorseshoe

Ha! Yes, jokes are always a little dangerous in writing - I figured the hedgehog comment gave it away though . . . As for simple answers, like anything else, I find they range from trite and meaningless to perceptive and helpful. From time to time I post questions in the coding forum, and then pray for a simple answer! BlueHorseshoe

[ Heaves big sigh . . . ] BlueHorseshoe

If your trading is discretionary then there are a number of platforms that have 're-play' features, I believe. This would allow you to test how you would have performed in historical time periods outside of the six months you describe. If your trading is rule-based then you simply need to learn to program those rules into backtesting software to see exactly how they would have performed. Hope that helps, BlueHorseshoe

Worse - it's flawed maths. If each trade has a positive expectancy and the probability of outcomes are independent then it makes sense to continue trading. Of course, this only applies in so far as a trader is able to continue following the rules from which the positive expectancy is derived . . . I think the 'stop trading after X losses rule' stems from the fact that many traders are unable to handle consecutive losses emotionally, and start breaking the rules, doesn't it? BlueHorseshoe

Thanks! One of the reasons I posted this thread was because there is repeated suggestion that many funds made material changes to the design of their systems following the flash crash; I'm wondering how many retail traders did the same? BlueHorseshoe

Hi SIUYA, Thanks for replying. I can't honestly decide whether or not I agree with you. On the one hand it's surely the case that everybody trading must be aware of the risks involved, crashes and order entry errors included, and have chosen to take on those risks. While I appreciate that errors and misunderstandings would be unavoidable in the turmoil of a busy pit, isn't any modern day trader who enters too many noughts basically just pretty damn careless? Is providing a safety net for careless or irresponsible market participants really promoting fairness? And although greatly accelerated in pace, it could also be said that the loss of liquidity in the flash crash is no different from the loss of liquidity in any other kind of crash. On the other hand, I suspect my feelings might be rather different if I were to find myself on the wrong end of this kind of event By the way, which exchange did you trade on, the LIFFE? BlueHorseshoe

Is tick data filtered by TradeStation? If so, is it possible to turn this function off? Or view the filtering algorithm that TS use? Cheers, BlueHorseshoe

Hello, The bidsize and asksize definitions are shown in the TS dictionary as followed by a bracketed field, but no explanation is given as to what an entry in brackets would denote. Does anyone know? Thanks, BlueHorseshoe

Hi Predictor, Thanks for an interesting response! The more I read about it, the more it seems that the futures exchanges operate with much 'fairer' and more transparent rules than equities exchanges (if anyone knows otherwise then please post!). One of the main differences seems to be the relative simplicity of order types available for futures. Busted trades are another example. BlueHorseshoe

I guess it just shows the extent to which we interpret market behaviour relative to perceived norms. Have you changed anything about the way you trade since the crash, or do you just regard it as a completely unpredictable and unavoidable rare event? Bluehorseshoe

Where were you during the flash crash of May 6th 2010? Did you have a position open, or did you have orders in place that were triggered during the price decline? The flash crash occured in the same week that I had first discovered trading. Amusingly, being completely new to trading, I watched the crash in real time and thought it was a perfectly normal day (I had absolutely no frame of reference for volatility at that stage)! Also, because it occured on the same day as a general election here in the UK, it got no real coverage in the news. Has the flash crash (or any of the subsequent similar incidents in the oil markets etc) affected how you trade or manage risk at all? BlueHorseshoe

Hi Uexkuell, It is indeed the key to a fail safe strategy, often called 'Futures vs Basket Arbitrage". Many market participants exploit it hundreds of times per day, but not retail traders. As you correctly point out, the delay occurs in sub-second timeframes. Unless you have the infrastructure to take advantage of it this strategy is of no direct utility. BlueHorseshoe

Hello, Ideally what you need to know is neither the historical number of completed trades nor the current number of limit orders. What you want to know is the number of buyers who will, in the future, be prepared to pay the spread. Because of the more simple and reduced scheme of order types available to futures traders, the information you need is more easily derived from futures than stocks. Futures are also far more liquid, resulting in a greater natural 'smoothing' of interest. Consider the following scenario: BIDAAAAASK AAAAAAA84 AAAAAAA83 AAAAAAA82 81 80 79 There are seven key pieces of information that are now of value: The number of traders prepared to buy at bid with limit orders at 81 (or less) *** The number of traders prepared to sell at ask with limit orders at 82 (or more) *** The number of traders prepared to buy at ask with market orders at 82 (or more) The number of traders prepared to sell at bid with market orders at 81 (or less) The total number of traders who currently hold positions (open interest) *** The total number of traders who may want/need to trade at this price Where your own order sits within the order book. I have asterisked the information that is available in your data feed. Two important derivatives of the above are the following: The ratio of traders who are prepared to buy at ask (with market orders) versus those who are willing to sell at ask (with limits). The ratio of traders who are prepared to sell at bid (with market orders) versus those who are willing to buy at bid (with limits). If you can work out how to obtain decent estimates of the information above that is not disclosed by your data feed, then this should form the basis of the information you need to judge liquidity and very short term price movement. BlueHorseshoe

Depending on what Carlton wants to do, this may not be useful. Once orders are filled it is too late to capitalise on any opportunity they may have presented. It may be more useful to know about orders that have not filled before they have not filled, so to speak. BlueHorseshoe

Well, it's more telling of historical demand . . . But otherwise I agree with everything in your post. BlueHorseshoe

No. Stocks usually react slower than futures. Nevertheless, I think that your idea of looking at interest at bid and offer is an excellent one, and if you invest the time to properly develop this based on the actual market behaviour (as manifest in historical data) then it should reward you. What you should be able to move towards is a more quantitative answer to the question that many of the accomplished daytraders on here often ask - 'who is buying/selling at this price, and why?' Hope that helps. BlueHorseshoe

You're not on your own - a frivolous and self-indulgent waste of money without a doubt. BlueHorseshoe