Friday, March 31, 2006

Weak Daily TICK: What It Means

What does it mean when the NYSE TICK hits a very low value during the trading day? I went back to March, 2003 and investigated all instances in which the TICK hit a value of -1000 or less. This occurred 108 times out of 775 trading sessions.

One day later, the average price change in SPY was .15% (64 up, 44 down), stronger than the average daily price change of .06% (432 up, 343 down) for the sample overall. It thus appears that a day of strong selling pressure results in favorable expectations the next day.

I took the analysis one step further, however. I broke down the strong selling pressure days into those in which the absolute value of the high TICK for the day was lower than the absolute value of the low TICK (N = 52) vs. those in which the absolute value of the high TICK for the day was higher than the absolute value of the low TICK (N = 56).

Thus, we're looking at high selling pressure days with low buying pressure vs. high selling pressure days with high buying pressure.

One day after the high selling/low buying days, the average change in SPY was only .01% (27 up, 25 down)--weaker than the overall market average gain. One day after the high selling/high buying days, the average gain in SPY was an impressive .29% (37 up, 19 down)--much stronger than average.

It thus appears that strong selling pressure without countervailing buying interest tends to carry over the next day, creating subnormal returns. Strong selling pressure matched by strong buying interest creates superior returns.


Jaewoo said...

BTW, we really appreciate your work. I am sure there are many who visit but leave no comment. Keep up the good work. I really enjoy your thoughts.

Brett Steenbarger, Ph.D. said...

Thank you, Jaewoo. That is very much appreciated. One of the benefits of a blog is hearing from other traders and sharing ideas. Some of my best ideas for analyses have come from reader comments and emails. Do stay in touch--


John Wheatcroft said...

I finally figured this out. It took some calculating but I think I have it now.

My thanks too, Dr. Steenbarger - yours is a truly intellectual site and always makes me think. I really appreciate that.

Brett Steenbarger, Ph.D. said...

Thanks, John. It's interesting...just the daily low value of the NYSE TICK appears to contain valuable information for near term returns.


Jeff said...

Hi Brett. I visit your site often and am always inspired to run my own numbers similar to the way you do it.

May I ask what data provider you use? I use Tradestation and only came up with 76 occurences out of 777 in which NYSE TICK was at or below -1000. That's a significant difference.

Thanks in advance,

Brett Steenbarger, Ph.D. said...

Hi Jeff,

Thanks for the heads-up re: data discrepancies among vendors. I obtain my data from RealTick and have cross-checked the TICK values with e-Signal data. It may be worth running the analyses using data values other than -1000 to see if the general patterns hold for other data sources.


D TradeIdeas said...

I wasn't aware we tracked TRIN and TICK values at Trade-Ideas until I was shown otherwise. Use the tool to scan and identify helpful patterns in the following:

$TRIN-A - AMEX Arms Indicator
$TRIN-N - NYSE Arms Indicator
$TICK-A - AMEX Tick Indicator
$TICK-DJI - DJIA Tick Indicator
$TICK-N - NYSE Tick Indicator

Thanks for the great post.

Brett Steenbarger, Ph.D. said...

That's excellent, David. It opens the door to a number of promising short-term screens. For example, I've found it worthwhile monitoring breakout high and low values for the TICK for price follow through.