Thursday, February 6, 2014

Our Trading Data Lies to Us

We are only as good as our data source and our data source is often inconsistent if not outright misleading.  I’m not 100% sure if realizing this earlier would have improved my trading but I understand it now much more vividly and it has altered my approach, especially in stock scanning.  One would think that the most simple of data points, a stocks float, would be constant across multiple providers, but it’s not.  Ex: SNCR- 18.2M from TeleChart, 34.4M from FinViz, and 21.15M from Yahoo.

What about volume?  Well, TeleChart’s volume is provided through BATS and often TC rounds, so 2.4M today, 2,363,466 from FinViz, and  2,364,181 from Yahoo.  And keep in mind this is just the known volume on the session and does not include dark pools so we can assume it’s probably much higher than any of these numbers. And Price?  Our charts deceive us here as not every price tick gets recorded and this results in the fact that the days high and low may not be accurate not only because of dark pools, but sub-100 lots will not register a tick on the end of day chart.

What about earnings? Today FLDM had earnings and ThinkorSwim has an estimate of -0.18 with an actual of -0.15. Briefing has and estimate of -0.13 and an actual of -0.04. Now I understand how analyst estimates can be site dependent, but the actual result should have no wiggle room what so ever.

Trading is a game of incomplete information and so is our trading data.  This can effect our entire trading plan from our vehicle selection and which stocks enter our universe and which do not based upon our scans as well as the criteria we use to build them.  This can effect our entry signals based upon which source has the most current and up to date information let alone the most accurate.  This can effect what type of alerts we use and what combination of criteria they are based upon.  This can effect how much we choose to micro-manage our trading.


What I’ve taken from this realization most is acceptance that there is an inexactness to trading and to release the tension of perfection in making sure I have the best candidates to trade. I no longer worry that I have the most precise entry requirements.  When price runs through an alert but it never triggers and I lose profit as a result of not exiting at my target or my stop runs slightly and my loss is larger, I’m much more at ease in letting it go.  In other words I’m learning remove precision, allow for sloppiness, and be OK with keeping it simpler.

No comments:

Post a Comment