Friday, October 31, 2008

Late-Day Market Surges

For the 2nd day in a row the S&P 500 saw a sizable move in the last 10 minutes of the day. This time, though it was to the upside as the index gained nearly 1% from 3:50 to 4pm. We saw yesterday an example of how overly strong reactions are many times overreactions. Tonight I looked at the sharp upside reaction as opposed to last night’s downside reaction.

SPX rises at least 0.75% in the last 10 minutes of the day. Buy on close. Sell at the next days close. $100k/trade. Last 25 years:

Ten out of 12 finished lower the next day. Below is a listing of all the trades:

The last 5 have been especially harsh.

As a commenter pointed out yesterday, overreactions haven’t been limited to intraday bars lately. To add to that, I would say almost every move the market has made in this highly charged environment has been an overreaction. It has in fact created vicious whipsaws in both directions for intraday and swing style traders.

Thursday, October 30, 2008

Overly Strong Reactions Are Often Overreactions

If you left 10 minutes early on Wednesday you missed a lot. The S&P lost over 3% from 3:50 to 4pm and was down over 4% before bouncing in the last minute. I looked back over the last 25 years to find other times the market dropped 3% or more in the last 10 minutes of the day. This was the 1st. Lowering the requirement to 2% unveiled 3 instances. They are listed below along with the next day’s performance:

10/19/87 – S&P rose 5.23% the next day.
9/29/08 – S&P rose 5.27% the next day.
10/27/08 – S&P rose 10.79% the next day.

This is too small a sample size to use for analysis, but a nice illustration of a simple adage. An overly strong reaction is often an overreaction.

Tuesday, October 28, 2008

CBI Moves Back Up To 13

The CBI spiked back up on Monday. After hitting 48 earlier this month it quickly dropped down to 7. It never made it back to the neutral “3” level, but the sharp move from 48 to 7 prompted sales in all CBI related trades on the morning of Tuesday the 14th in the Quantifiable Edges Subscriber Letter. The CBI is now growing again and posted a 13 reading at the close yesterday.

Like nearly all breadth indicators, the CBI has not been quite as reliable this year. Still, the move to 13 has prompted me to take some long exposure. For those not familiar with my CBI you may read an intro here. Plenty of additional information is available by using the CBI label here. Below is graph of S&P 500 along with the CBI for 2008:

(click chart to enlarge)

Monday, October 27, 2008

Back To The Extreme

A couple of weeks ago I showed a table detailing a number of extremes the market had reached. The sharp bounce on 10/13 relieved many of them. Now some are back – along with a few new ones: (click to enlarge - for some reason I've had trouble making the .png's larger in Blogger the past few days)

Friday, October 24, 2008

Rally Lacks Breadth - A Bad Sign In Recent Years

On the NYSE up issues accounted for less than 40% of the total issues traded. This only the 3rd time since 1970 that the up issues % was below 40% while the S&P 500 rose at least 1%. I ran a study tonight that looked at other times the market had risen at all when the Up Issues % was less than 40%.

(click to enlarge)

Thursday, October 23, 2008

Reversal Tendencies For 2% Gaps Lower

Last week I showed how gaps up of 2% or more tend to reverse over the next few days. Today I’ll show gaps down of 2% or more. In the table below I show all instances where the SPY opened 2% or more below the previous day’s close. The column on the right shows how long it took for the SPY to close at a price above the opening gap price. Including Wednesday’s failure, the SPY has closed above the 2% gap down open about 69% of the time. There has only been one instance where it didn’t close above the gap open at some point in the next 3 days.

Also, here’s an interesting tidbit from last night’s Subscriber LetterHow incredible has the market action been of late? Since 1960 the S&P has had 14 days where the market sold off 5% or more. The 1987 crash accounted for 3 of them. The 2000-2002 bear market only had 1 – on 4/14/2000. In the last month the S&P 500 has seen 5 days with losses of 5% or more.

Wednesday, October 22, 2008

Stretched VXO Not Bearish

A common misconception among traders is that an extremely low VIX or VXO compared to its short-term moving average has bearish consequences. On Tuesday the VXO closed over 15% below its 10-day ma for the 2nd day in a row. Below is a test that shows results of shorting the SPX when the VXO is stretched below certain levels for at least 2 days in a row. It covers once there has been a reversion to the short-term mean:

The system as designed shows a slight edge when there is a mild VXO stretch. As the stretch gets larger the edge disappears. To understand why this occurs, consider what would cause the VXO to get extremely stretched, as opposed to slightly stretched. A mild stretch might be the result of a typical move higher in the market. An extreme stretch is more likely to be the results of a strong thrust higher in the market. If the market is truly strong, then the result will be more upside – not downside.

When trading below the 200-day moving average the difference is even more pronounced.

Here we see the mild stretches have more bearish implications. A drift higher in a downtrending market can offer short opportunities. A strong move higher, on the other hand, can lead to a vicious short-covering rally – not something you normally want to step in front of with short trades.

I used some other studies back in May to illustrate these concepts as well.

There are indicators suggesting short-term downside, such as the volume study I posted yesterday. The extremely low VXO is NOT suggestive of a selloff, though.

Tuesday, October 21, 2008

Volume Disappoints Again

Last week I looked at the extremely light volume on the Nasdaq rally and suggested it had bearish implications. The plunge came soon after. Monday’s rally was also suspiciously light. This time the setup occurred in both the Nasdaq and the S&P. Below are the stats for the S&P:

Volume action continues to disappoint. I’d be surprised to see a reaction as negative as last week, but volume is suggesting that perhaps buying enthusiasm isn’t as strong as would appear based on price alone. A pullback may likely be in order in the next few days.

Monday, October 20, 2008

Monday Morning Observations

Just a few quick observations this morning. Although the market rose last week, several measures of breadth and volatility remain at extreme levels suggesting more of a bounce is likely. For instance, the VIX made a new intraday high on Thursday and a new closing high on Friday. The CBI remains elevated at 13. Worden Bros. T2114 also remains extremely elevated. T2114 measures the number of stocks trading at least 1 standard deviation below their 40-day moving average. T2114 closed at over 90% on Friday despite the fact that the market rose last week. The current 90% reading still exceeds all other periods other than the Crash of ’87.

The market is looking to gap up this morning. The SPY is up over 2% a little before the open. Traders may recall my post from last Monday that suggested 2% gaps tend to pull back at some point over the following few days. It happened twice last week. The market has been extremely choppy. If this pattern of choppiness is to be broken there will need to be a strong move right from the start which utilizes the extreme conditions mentioned above to mount a sizable rally.

As violently choppy as the market has been it is also possible that both the short-term bullish and short-term bearish indications are satisfied over the course of the week. In any case, it should be another interesting one.

Friday, October 17, 2008

VIX New Highs Without SPX New Lows

The VIX put in a huge spike yesterday to another new high. The SPX manage to hold above its recent low before rebounding. Below I looked at long-term VIX spikes that were not accompanied by long term SPX lows and performance moving forward:

In the past there this particular setup has provided traders an upside edge.

Wednesday, October 15, 2008

Scary Pictures

Here’s a chart of the Dow Jones:

Here’s another chart of the Dow Jones:

The 1st one is 1929. The 2nd one is 2008.

They sure look a lot alike to me.

On Wednesday the S&P dropped 9%. It was the 3rd time in less than a month that it dropped over 7.5%. Since my S&P data only goes back to 1960, I checked out the Dow to see if it had ever dropped 7.5% 3 times in one month. It had. Once. In the 1929 picture you see above. While history never plays out exactly the same I’m sure everyone is wondering how the current picture resolved itself in 1929. Was the initial crash low broken?

Answer: Yes. The current period compares to the beginning of November in the 1929 chart. There was one final leg down before a sizable rally ensued that lasted well into 1930.

Of course if we zoom out a bit more…

This is not a quantitative study. It is not a commentary on the state of the economy or the action of the government. It is definitely not a prediction. But if you thought there was no way it could get much worse from a long-term standpoint…well…it could.

Tuesday, October 14, 2008

Volume Clues

The markets wild ride is virtually unprecedented when looking at price action. Determining the markets next move based price bars alone is difficult to dangerous at this time. My current focus is on volume. On Monday the Nasdaq rose over 11% yet the volume was the lowest in 5 days. Previously, the largest percent gain accompanied by a 5-day low in volume was under 6%.

To get a sizable number of occurrences I had to lower the % gain to 2%:

Indications are for weakness in the 5-10 days following such occurrences. Below is a table that appeared in last night’s Subscriber Letter which looks at gains of 3% accompanied by the lowest volume in 5 days:

The larger % gains showed even worse performance.

As I type this morning the Nasdaq weakness is playing out. The Dow and S&P remain higher though. NYSE volume was also fairly unimpressive on yesterday’s rally. I’d be wary of further gains on low volume. I’d also continue to watch volume for clues at a time when price action is incredibly volatile.

Monday, October 13, 2008

2% + Gaps And Their Tendency To Pull Back

It’s late Sunday night and the S&P futures are up over 4%. Typically when the market gaps up by massive amounts it tends to pull back at some point in the next few days. Below are listed all instances where the SPY gapped up by 2% or more. In 19 of 20 cases it posted a close below the gap open within the next 5 days:

Friday, October 10, 2008

Two Possible Outcomes

It seems the market is getting ready to either:
a) Complete this capitulation and reverse upwards very hard, or
b) Drop to zero and close its doors for good.

At this point I wouldn’t rule out either one. If you’re of the belief that it is more likely to reverse hard than go to zero then you’re probably in the minority. You may also find the following charts interesting. They’re 5-minute charts of the legendary reversal bottoms. Whether you’re long, short, or sidelined you may want to study them for a few minutes – in case it doesn’t drop to 0.

September 1, 1998

July 24, 2002

October 10, 2002

They didn't all happen in 1 day. Here's 3/12 and 3/13/2003.

One common theme I see is that once the bottom was established the uptrend stayed in tact. October 10th 2002 saw a brief and relatively minor break of price support just before noon. Other than that all the rallies held their ground throughout the day. In other words, if you're trading intraday, a loose trailing stop may not be a bad idea. A sharp break of support seems unlikely if the market is going to put in it's next legendary reversal.

Thursday, October 9, 2008

Extreme Conditions

The market is hitting massive extremes with regards to breadth, price, and volatility. Below is an incomplete list that illustrates just how severe current conditions are:

Take just about any of the conditions above, slice it in half, and under “normal” circumstances it would be extreme enough for a bounce a high percentage of the time.

Note: The top 3 breadth statistics come from Worden Bros. - T2106, T2114, & T2116.

Wednesday, October 8, 2008

CBI Finally Spikes

The CBI (Capitulative Breadth Indicator) finally spiked above 10 today and hit 12. For those who are unfamiliar with my CBI indicator, it basically uses a proprietary calculation to determine how much capitulation is evident among large-cap stocks. Spikes of 10 or higher in the past have led to market bounces on a fairly consistent basis. Those who would like more info on the CBI may want to read the intro post here or the full post history here. Until Tuesday the CBI had sat relatively dormant. I had thought the market might rebound before the CBI ever hit 10 this time, but this market seems bent on marking every extreme.

Until July, buying the S&P any time the CBI hit 10 or higher and selling it on a return to 3 or lower had a perfect record. The July trade turned out to be a loser. Below are statistics going back to 1995, which is as far as I was able to accurately reconstruct the indicator. The CBI has been tracked live for about 3 years now. All trades assume $100,000 into the S&P 500.

Impressive stats, but it’s important to keep in mind that the current environment is unlike anything we’ve seen since before data exists on this indicator.

Tuesday, October 7, 2008

An Off-The-Charts Example

As an example of the kind of extremes I was referring to in my previous post, below is a chart of (NYSE New Highs - NYSE New Lows) / Total Issues. New Highs are in the top panel. News lows are in the 2nd panel. Total issues are in the 3rd and the net percentage is in the bottom panel. Over 50% net of issues hit new lows yesterday.

According to my data, the last time the reading was under -50% was 10/19 and 10/20/87. Prior to that was 5/21, 5/25 and 5/26/1970, which is about as far back as my data goes.

Edit: It was pointed out in the comments section that Dr. Steenbarger also noticed this. His data went back prior to 1970 and he found an additional instance.

Everything Is Off The Charts

Most every indicator I look at with regards to breadth, volatility, and price action is strongly suggesting a strong short-term bounce should be at hand. Below is a short excerpt from Sunday night’s Subscriber Letter which puts some of my thoughts on these extremes into context.

What needs to be kept in mind is that the price action over the last week has been more severe than at any time other than 1987 and then back to the 1930’s. In other words, while extreme readings in breadth, volatility, price, and volume indicators of this magnitude have almost always led to short-term upside over the periods tested, the current situation is far beyond most everything tested. Measures need to be taken to control risk. Tight stops are a possibility, but difficult to implement with such extreme volatility. I’m controlling risk by scaling in with reduced position size.

Monday, October 6, 2008

An Elevated VIX Study

On Friday the VIX closed above 45 for the 2nd day in a row. This is the 1st time since the VIX has been measured back to in 1990 that this has happened. Meanwhile the VXO closed above 50 for the 2nd day in a row. The only other time readings this high can be seen were in a back-adjusted 1987 period during and after the crash. I ran some tests to see how the market has performed the week following back to back readings above other extremely high levels:

While the instances get low over 40, average profits of greater than 5% over the next 5 days across the board are quite impressive.

Friday, October 3, 2008

Another Example Of Unprecedented Volatility

Including Thursday there have been 18 days since 1960 where the S&P 500 has closed down 4% or more. Four of them have come in the last 3 weeks. The only other period to come close was the Crash of ’87 when it occurred on 10/16, 10/19, and 10/26. Of the previous 17 instances, the market finished higher the next day 14 times. All instances are listed below:

Recent volatility is tremendous, and this is just another example of it.

With such volatility comes opportunity. When looking to take advantage of edges during extreme periods such as this, traders need to make sure they are comfortable executing their plan. Otherwise they could end up as part of the panicked crowd. Trading while in a panicked state simply isn’t conducive to optimal decision making.

One last tip. Bailout news tomorrow could make for fast market conditions. Traders may want to place any stops they are planning ahead of the news. Otherwise execution may become difficult to impossible.

Wednesday, October 1, 2008

Strong Bounce, Weak Volume

A few weeks ago I showed how weak bounces have a tendency to quickly roll right back over. The good news is Tuesday’s bounce was far from weak. Tonight I decided to show a similar study examining strong bounces after a sharp move to new lows:

As you can see there tends to be an immediate and lasting edge when the bounce is sharp like we saw on Tuesday. There was a weak spot with Tuesday’s bounce, though – volume. It came in quite a bit lighter than Monday. Adding this filter changes the results to look like this:

While there is no negative implications in the first few days, the apparent lack of big buyers (volume) does seem to have a negative impact on returns after day 4.

Instances are a bit low, and many of them aren’t very comparable to current conditions. There’s a big difference between a rebound from a 2% drop and a rebound from an 8% drop. Therefore, I’d suggest the appropriate thing to do with this study is keep it filed for future reference. Then perhaps review some of the charts I posted last night.