Hey guys, a lot of my intraday automated trading system stopped working this month across indexes and commodities. anyone experiencing similar things like me ? Any idea which historical market period is best to compare to the current one ?
Considering we're really talking about "unusual" or "extreme" volatile time period... Why don't you just pull up a weekly or monthly chart of the VIX. It's easy to see similar volatile historical market periods although their reasons for the extreme volatility are different. Some of these extreme volatile historical market periods have similar like reasons while others do not...all of these reasons has absolutely nothing to do with technical analysis. Its mainly just geopolitics, war (armies or currency), debt crisis, FED/IMF/ECB decisions, global economics, temporary trading rules that causes fear amongst those that have the greatest impact on the markets, illegal trading activities and many other none technical reasons. Most data sources go back as far as year 2000 while others go back as far 1986 when the VIX first started although the professor that designed it actually started computing it I think in 1984 or 1985. I remember because when I was a kid my old man was a floor trader at the time, a few of his buddies were over at the house playing a big game of high stakes poker...they were calling the VIX a game changer. Just in case you are curious, the biggest historical volatile period (since 1980s) also began in Asia. I'll let you pull up a historical VIX chart to determine the dates for such (your homework). Oddly, in the recent situation, many were concerned about Greece while forgetting or underestimating what was brewing in China. I know a few algo traders that begin adapting (making adjustments) about 2 - 3 months ago. P.S. I'm not an algo trader. P.S.S. I know a PhD guy that calculated his own custom VIX going all the way back to the late 1800s (1890) when exchange data became more accurate due to changes in technology. Reasons I mentioned above are primary causes for dramatic (volatile) changes in the markets.
Not a good year at all. The obvious comparisons to this would be January 2008, October 2008 and the 2011 volatility period but then again my experience doesn't go beyond 2003. It might be one of those "pattern breaker" years, even the best things that work seem to have these, market's cleansing in order to get rid of the riff-raff!
You got to also look at the other situation. These types of volatile trading conditions are top profit periods for some traders...best trading months. That's because their trading plan is geared towards these types of trading conditions or the traders have different trade methods in which one is very suitable for these types of trading conditions. Thus, its not uncommon for successful traders to have different trading systems for different types of market conditions. In contrast, too many traders have ONE system only that's designed for a certain type of markets and not designed for market conditions like this. The latter is easy to reveal via someone making a historical performance chart (hypothetical/backtesting) of their trading system (even algos) going back to at the minimum to year 2000 but preferably back to the 1980s and then overlapping it with the VIX chart to see the performance in those years or months the VIX was spiking... If the system designed performs poorly when the VIX is volatile, they'll at least know they need to shut down their system when the VIX spikes like that and they'll know they need to have another system that's designed for situations that we saw last week when volatility went crazy if they want to trade successfully in conditions like last week. I just don't believe ONE system works for all market conditions as if price action doesn't change. Too many folks run around saying markets do not change but those same folks are awfully quiet during trading conditions like last week.
LOL, as I pointed out in my previous posts, there are only two type of trading : mean reversion or trend following depend on the market conditions - spike (aka trend), channel (aka slightly trend with some pull-back) and range/chop (barb wire) It is easy to say this and that in the HINDSIGHT. If you are able to know the market conditions in real time, you will be the richest man in the world. No one can do this even the most advanced/informed institutional in this earth today and this is why their main money making machine still the brokerage/commission/financial product.
His question didn't have anything to do with trade signals. Simply, I think you're stuck on trying to find trade signals as the only way to profitability. In contrast, his question is about what other historical market period compares to last week ? Its a simple answer to find. To find it...he'll need the VIX chart or anything similar (e.g. VXX). Thus, I do not believe his question was about price patterns as in what prior price pattern is similar to a particular trading day last week. Lots of statistical new records occurred last week and none of them had anything to do with price patterns. As to market conditions, you don't need to know them in real time. In contrast, you need to recognize what they are in hindsight after they occur so that you can adapt and make adjustments just in case the conditions continue in reference to extreme volatility. Yet, you don't want to recognize it in hindsight too late because by the time you figure out what has happen...market conditions have changed again. For example of hindsight...prior to the big U.S. market sell-off on Monday and Tuesday...the week before contain some of those reasons I mentioned in the prior messages (e.g. China) and new records in some markets were occurring one week prior to last Monday and Tuesday. That's hindsight info to be prepared for Monday and Tuesday market conditions. Once again, you don't need to know the info in real-time...hindsight is ok but not too late. Yet, there are several threads here at ET talking about the market problem A FEW weeks prior to last week. Too many folks think some price action pattern all by itself will be enough warning about Monday and Tuesday market conditions last week. It just doesn't work like that.
I don't know anything about CNBC and I think you have clearly missed the point. What has happen in China was not normal. It occurred prior to Monday's and Tuesday's market volatility. The last time I saw prior coverage like that was early summer of 2008 just before the global financial fiasco started late 2008. Once again, nothing to do with price patterns and nothing wrong with being prepared if it happens again. If it never happens again...no biggie. A trillion dollars was lost prior to Monday and Tuesday in China alone. Sooner than later that will impact global markets and it did and if it happens again...we'll see another week like last week. You don't need any real-time info...you just need to be ready via the hindsight info and understand prior if your trade system works or doesn't work in conditions like last week. Any smart trader that got burned last week will probably have done some statistical review and make adjustments just in case it happens again and it will happen again.
The thing to note about the China thing is that it was entirely possible to foresee all of this actually happening had some critical outside the box thinking been applied. I applaud all traders who positioned for this possibility appropriately as they were ahead of the curve.
Guys ..., I wish it was as easy as pulling up a VIX chart but, the correlation between asset prices such as indexes, PMs and Treasuries are a bit different then 2008 and going up further back for intraday patterns will be moot. as algo trading have enormously impacted price patterns on intraday trading in the last few years. most obvious example is CL. So I might be forced to sit the next few months out for my algo intraday until I collect sufficient data for this market period. except for some options weekly play. long gamma maybe ? Any suggestions on fun vacations or activities for september-october ?