What Wall Street Learned from Willie Sutton

Willie Sutton was arguably America's most famous bank robber (until now, when execs of FNM, LEH, WM, MER and others have stunningly overshadowed his accomplishments). During his bank robbing career in the 1920s and 30s his estimated take was $ 2 million, even though he spent more than half his adult life in prison. When asked by a newsman why he robbed banks he reportedly said, "because that's where the money is". Although that anecdote was denied by Sutton in his autobiography, he did say the following:
"Why did I rob banks? Because I enjoyed it. I loved it. I was more alive when I was inside a bank, robbing it, than at any other time in my life. I enjoyed everything about it so much that one or two weeks later I'd be out looking for the next job. But to me the money was the chips, that's all." Go where the money is...and go there often."

Seeking Alpha recently profiled some of the better banks.
IMHO, worth a close look in these times when bankers are generally viewed as liars, scalawags and scum of the earth and when the shameless looting of depositor accounts and betrayal of investor's has become the norm not the exception.

Got an Hour?

I mentioned Nouriel Roubini last week and if you haven't had a chance to check him out here's the link to his blog. His spin on the markets is not something you're likely to hear from the shills in the adminsitration, but his perpsective and his prognosis make a lot of (scary) sense.
If you click through to the above link and then scroll down to his hour long Oct 14th interview video with Charlie Rose, you'll get Roubini's full analysis of the financial situation.

Visiting the Cheaphooker

This is an update of my Cheaphooker system that's posted on the side panel of the blog as part of my Dirty Dozen systems. The code is the same as the original, I've just focused on a shorter time frame (14 months) in order to see how the inputs have changed. Obviously I've had to do some tweaking of the input values in order to reflect the downward market bias for the past year The system has a great balance of longs and shorts and has produced excellent max consecutive loser ratios. Although the system generates a good risk/reward return, another way to use this system is to monitor the equity curve and pay attention when it starts to flatten out.

TS2000i code shown below:
I've left the code with programmable inputs so you can run optimization tests on other ETFs, stocks, etc. that might better suit your trading interests.

Inputs: Len1(3), Len2(10), Len3(18), Len4(1), Len5(10), Len6(10);
Optimized inputs: Len1(3), Len2(6), Len3(6), Len4(1), Len5(8), Len6(12);

If DayOfWeek(date)=Len1 and Close < Average(C,Len2)
Then Buy This Bar on Close;
If BarsSinceEntry = Len3
Then ExitLong at Market;

If DayOfWeek(date)=Len4 and Close > Average(C,Len5)
Then Sell This Bar on Close;
If BarsSinceEntry = Len6
Then ExitShort at Market;

Weekly Pivot Update

As expected, we got a range contraction this week centered on the PP. Rather amazingly, all 4 of the ETF basket closed at the PP and all four are showing momentum to the upside as the likely path of least resistance. While the upcoming week's range and % range remain almost identical to last week's, absolute delta has dropped significantly, especially in the XLF.
Also, for the first time in many weeks the upcoming pivots are very similar to last weeks, clearly a change in the otherwise virulent downtrend momentum. This leveling off may be an indicator of a bottoming process in progress, although if you viewed the Roubini video posted on Monday, you know his argument for the bottoming process to continue for some time.

Side note: I made a quick scalp in GE today, deciding to take my own advice from yesterday.
In at 19.34, out at 19.97 for a quick 30 minute trade. I should have stuck around longer, but had projects outside the office to take care of, so I walked off early.
But the thing really interesting about GE was in the Nov options, which I mentioned yesterday. Those same Nov 20 calls that closed at 1.83 yesterday closed at 1.30 today, down some .53 while GE was only down .19. Talk about premium decay. . .and the VIX was actually up 2.72 to 70 today (coming off an intraday high of 74.5). Go figure.

What a difference a day makes

With today's little 800 point swing and the VIX hitting 81 and closing at 67, we have recovered from new lows, shown volume above both the 10 and 20 DMSAs in 3 of our 4 ETF basket (XLF is above the 20, and at the 10), and seen an upslope swing develop in the technicals.
The Qs chart is incorrect . . Telechart data is not updating.
The Qs closed at 32.24, and are now poised at the lower channel band of the daily LR30. While the RSI 2 remains downslope in all 4 ETFs, this could reverse on a small pullback Friday. As some of my previous studies have demonstrated, Fridays have generally fared better as days to sell rather than days to buy. Whether expiration dynamics will muddy those waters remains to be seen, but all things being equal we might see a ncie rally develop next week.
For the risk adverse, some interesting buy/write plays.
GE, for example. With GE at 19.87, the Nov 20 call pays 1.80 or 9%. The 17.50s pay 3.45 or 5% with almost $2.50 in insurance.
Lots of other similar situations.
Hey! If Buffett liked GE at 22.25, shouldn't I love it at $20????

Sell New Highs - Correction

OK, I messed up. Humble thanks to Ramon for catching the error. Yesterday's Sell New Highs system had some poor coding on my part. I think that derives from my use of the HighestFC function to identify new highs. I originally decided to use that function because the intent was to scan a larger universe of ETFs and optimize the lookback period. For the IWM that lookback period ended up being 1 day and as a result TS generated trades both on the next bar and on 9 bars out, hence the 4 day average holding period. I should have seen there was a problem when a sytem generates 50 trades in a year with a 9 day hold. The chart above reflects the error in the coding, which has now been corrected to read as follows:

If Close > Close[1] and High > High[1] and Low > Low[1]
Then Sell This Bar at Close;
If BarsSinceEntry=2 Then ExitShort this bar at Close;

The resultant system has the same frequency of trades as the original faulty system, but the holding period (and risk exposure) is reduced to just 2 days . . a holding period I can live with. This is actually an attractive system to me as with the corrected code, the system has a relatively low intraday drawdown, a good max winners/max losers ratio, and trades on an average of once a week. If nothing else, the system should serve as a warning to momentum trades who still like to buy new highs.

Thanks again Ramon.

This could get ugly

With today's multiple closes on the lows or near lows of the day, odds strongly favor further decay tomorrow and perhaps beyond. Although odd and scary reversals can happen right at expiration it's hard to fathom what possible catalyst could lead to such an event. That being said, the highly anticipated mega volume washout may be close at hand, or we may just be basing at these levels before a drop into the real abyss.
I see Adam made an offhand reference to Nouriel Roubini, whom I referenced in Monday's post . . .I'm sorry Adam didn't explain his comment a little clearer. If you haven't watched the video yet, it's definitely worth a look and listen.

Despite the somewhat muted volume today, these are not bullish charts.

Sell New Highs

In my continuing search for market momentum opportunities, here's a very simple approach that sells daily bars when the close, the low and the high are above the previous day's values.
The time period under investigation is only the last year. Extending the test period back 5 years requires a little tweaking of the hold period and the lookback period for highest values, but still proves robust.
I like the max consecutive losers . . 1 . . and the 82% profitable is a nice risk/reward ratio. Max drawdown is a bit of a concern for this size account, and I'll explore possible ways to scale it back in future posts.
This is the optimization report for a 10 day holding range. Not sure what's magic about 9 days, but that's the number to hit for the past year. Holding this short anywhere from 3 to 9 days pays out a respectable amount. The other lesson to be learned from this little system is that new highs have been followed by extended selloffs, not continued new highs.

TS200i code shown below:

If Close=highestFC(close,1) and Low=highestFC(low,1) and High=highestFC(high,1)

Then Sell This Bar at Close;

If BarsSinceEntry=9 Then ExitShort this bar at close;

XLF and Lag

On the above chart pair . . Qs on the left, IWM on the right in 5 minute bars. At a quick glance 3 feet from your monitor, the charts are virtually identical. They both fade off R1 on the opening bar, bounce off the PP 30%, then plunge back through the PP to hit S1 at 14:45, show a little hairy bottom at 15:00 and then return to PP levels in the last half hour of the day. While the range volatility was more pronounced in the Qs, the pivot action was a mirror image.

Now let's look at a comparison of the same 5 minute charts today for the XLF on the left versus the IWM on the right. XLF comes out of the chute just below R3, slides through R2 down to R1 and then bounces back and forth between R2 and R1 for the rest of the day, closing just below R2 for the day. Clearly a strong day and a strong close.

Over the weekend I suggested that the LRs and the weekly pivot study argued for XLF to outperform the other components of my little ETF basket. That proved not to be the case yesterday, but was clearly the case today with XLF up 6.5%, while XLE was flat, IWM lost 2.8% and the Qs lost 4.3%. XLF's outperformance may have been driven by the adoption of the bank nationalization plan with the decline in other indices a function of the probable impact of that $250B drawdown on those sectors. Maybe . . . but I really have no way of knowing for sure.

What's interesting to me is not the underlying causes for XLFs gain, but the fact that it happened. For some time now I've been working on a timing project I call Lag, Sway and Slide and the relative action in XLF over the past few days has been perfect example of what I would term "lag". Traders who use RSI, %R and CCI based trading systems know well that profits can often be increased with reduced risk by entering the trade several bars after the actual trigger signal. Exactly why this lag works is not apparent, although I could conjure up a number of possible theories. I suspect there is no single simple explanation, but rather a variety of intervening market forces that can delay or slow down momentum in the markets. The trick is knowing when your trading signal is valid and on a short term pullback and when your trading signal is false, which can be costly. Daytraders often utilize fractal signals to both expose and manage lag. Traders who focus on 1 or 2 minute bars routinely toggle to 5 or 10 minute bars to confirm technical trigger signals. A multi-timeframe view of the markets helps avoid the chop inherent in 1 or 2 minute bars while at the same time lending confirmation to trade signals and providing logical stop loss levels. While the 1 or 2 minute bars may trigger a trade, it is often the lagged confirmation of the 5 or 10 minute bars that actually provides the higher profit, lower risk signal.

I'll explore lag, sway & slide concepts in future posts along with some practical applications.

Monday, Monday

Although the technicals looked somewhat bullish on Friday, they didn't look THAT bullish. And to add insult to injury to my market weekend prognosis, the XLF did not lead the charge up, but instead lagged most of the day.
Hey! it's a learning process and I actually learned a lot from today's market dynamics. The big argument against today's rally sustainability was the lack of volume, but given the lack of fade into the close, the most likely opening momentum tomorrow is, in fact, continued up. Negative earnings surprises or the usual revelation of of some new financial shenanigans could derail that momo, but today's strength was sooo unusual and big moves tend to carry overnight. Above is an update of this week's pivots as compared with last weeks. The yellow pivots highlight the weekly high/low range for the ETFs on the left matrix. The yellow pivots on the right matrix highlight the today's high/low range for this week's pivots. Today's pivot bounces off of the PP mean strongly argue for continued bullish momentum, and to show how expanded the ranges had become, today's 936 point gain only moved 2 ETFs of our basket 1 pivot and the other 2 ETFs 1.5 pivots. This behavior actually makes sense when we examine the %range values for the 4 ETFs. The 2 ETFs with a %range 19-21 rose 1.5 pivots. The 2 ETFs with a percent range 37-39 rose 1 pivots. Today's pivot range results correspond to last week's results, but with much more drama. Current values argue for short term range contraction, in order to bring the %range values back into mean territory.

Finally. . . a video link sent to me by fellow trader Dr. Carl Wyman. Because the video is an IDB product, the link will yield a video of J. Mazza (in Spanish). If you scroll down to the next video, you'll see the Oct. 1oth link to Nouriel Roubini. That's the one to check out before you get too excited about the longer term bullish prospects for the market.

Weekly Pivots Update

Above, the 3 LRs study on weekly bars.
Results for the past week: Qs down 13.5%, IWM down 20%, XLE down 14.6%, XLF down 3.7%.
The XLE did retrace to 2005 levels, as suggested last Monday. I frankly did not expect it to occur so quickly, but then expectations about recent market behavior have surprised more than a few traders.
The Qs and IWM retraced to 2003 levels, while XLF sank to its lowest level ever. Although its not at all clear that the weekend global economic summit will produce any meaningful results, perception is the key.

Once again our little pivot and range spreadsheet proved accurate in forecasting market range expansion, and once again I was surprised by the magnitude of the % range gain, punctuated by a VIX of 75 on Friday.

What is interesting to me is that this rather simple statistical approach to mean reversion was able to find some direction in the volatility of the markets. Last week I talked about the relative range contraction in XLF as indicative of a basing pattern. While this week's range activity was clearly in an expansion phase, the fact that XLF returned the best performance of the basket demonstrates the potential of closely examining range expansions and contraction patterns. Looking forward, what's different about this week's pivot matrix is that 3 of the ETF basket components managed to close above the low of the week by one half pivot and the XLFs managed to close above the low by a full pivot. Contrast that performance with the previous week, when all the weekly closes were on the low half pivot of the weekly range.

Things are looking up with XLF the ETF most likely to outperform this week again.

I read several articles on Thursday suggesting that unleashing the short sellers would actually support the XLF (and other financials) by providing liquidity. Whether that argument holds up to close scrutiny is subject to some debate, but the net effect was, as noted, out performance by the financial sector this week.
Above is the next generation of the weekly pivot study. . . a work in progress that is designed to analyze the probability of a single stock, ETF, index, etc. for mean reversion momentum. Clearly, we have a downslope trend in progress accompanied by an expanding weekly range, both in absolute values and percentage. I've added a graphical interface to demonstrate the mean version parameters, but blogger has done strange things and locked up when I've tried to post it, so I'm working on that little problem and hope to update this post early next week.

No Bottom Yet

This is a little lesson is why buying plunging closes can be dangerous. Going into the last 15 minutes yesterday, I suspect many traders were looking for a bounce this morning. Although I entertained similar thoughts going into the close, I ultimately decided that the probabilities favored more decline, based strictly on the technicals. Here's why.
First of all, I don't know about your trading platform, but mine started to lock up about 13 minutes pre close. This is always a bad sign in the last 20 minutes of the trading when a platform starts acting squirrely because it typically means there are massive MOC trade imbalances that market makers are having trouble resolving. For the last 10 minutes of the day the quotes were essentially locked and the charts blank. If you can't get a reliable quote, you don't want to trade and if you do, you had better use a limit order. Ultimately, I couldn't get a final print until 8 minutes after the market closed. The other issue is simply time related risk management. In the midst of earnings season, with weekend coming up and a global financial meltdown in progress, what are the odds that favor holding positions over that time frame with no exit? Pretty slim.

But equally important as the platform problems was the condition of the VIX, a historic high in the 60s. Although the VIX started going parabolic the last hour, it wasn't at all clear that an exhaustion peak was close at hand and with the 8/16 MAs wide apart going into the close, and I noted on VIX and More that VIX 80 looked like a real possibility.

But perhaps the last nail in the coffin for the bulls was the complete lack of enthusiasm in the NYAD. Although we saw a little pullback in the equities and the VIX, the NYAD remained steadfast in downslope, which I why I place such a high level of confidence in this market tell.

What the downslope NYAD was telling me was that the little green surge in the last 10 minutes was the result of shorts covering, rather than longs buying.

Buy Bonds Friday Close

So the flip side to selling the Qs on Friday is buying TLT, the 20 year treasury bond fund.

This system is even more abbreviated that the Qs/IWM Friday sell system . . . here we buy TLT on Friday's close and exit the position on Monday's close.

TS2000i code is:

If DayofWeek(Date) = 5 Then Buy This Bar at Close;

If BarsSinceEntry = 1 Then ExitLong This Bar at Close;

As with the Sell Friday system, you don't even need a computer to figure out this trade.

This might be a nice pair trade for the weekend . . short the Qs and long the TLT in equal dollar amounts. The Qs system gives you more bang for your buck, but the TLT side may smooth out drawdown when the Qs trade doesn't work.

Of course, if market momentum turns up and the Qs LR30 channel goes upslope, all bets are off and we need to reverse the buys and the sells. We could add a simple moving average filter to confirm the trades . . . in the case of the Qs if the current close is less than the 17 DSMA then performance improves a few percent, we cut total number of trades to 23 and overall return is knocked down by $1000, so my preference is to trade it without the filter. Nevertheless, market momentum does require monitoring to assure than the trades are reflective of the current market bias and the 3 finger system may be one approach to accomplish that goal. I'm going to fiddle with that concept over the weekend and will post any interesting developments.

Sell Fridays (and Mondays)

Sometimes you get this gnawing feeling while watching the markets in 2 minute bars that there must be an easier way. I've also noticed while fiddling with the Cheaphooker settings that Fridays haven't shown a lot of green for the past year so I ran this little test to confirm my suspicions. This data is for the last 50 weeks, since I was really only interested in seeing how market risk has been managed from the 2007 Fall highs (little play on words there).

Turns out Fridays are a great day to be short, at least in the Qs and the IWM.

I like this risk/reward and the # of consecutive losers for both examples, but the Qs model is the more attractive of the 2 for my money.

Feel free to scan a larger universe of indices and ETFs, or your trading favorites and post results as comment, if so inclined.

The code couldn't be much simpler. Here it is in TS2000i, but you don't even need a computer to trade this system . . just call in the order.

If DayofWeek(Date) = 4 Then Sell Next Bar at Market;

If BarsSinceEntry = 1 Then ExitShort This Bar at Close;

The comment keeps coming up, how do you accomplish "This Bar at Close" executions. With Schwab StreetSmart Pro I just set an Alert to place the trade at any designated time . . in this case 12:50 pst. Platforms without MOC orders (market on close) or a timed trading execution feature may require you to monitor the trade into the close and execute manually.

Reading the market

This an update of a previous screen posting of my Schwab StreetSmart Pro platform setup and may help you understand what I'm looking at to guide my intraday trades.
Front and center top to bottom:
The TICK, shown here on 2 minute bars, although I toggle to 1 minute bars for the open.
I overlay the pivots on the TICK and also use an 8 & 16 MA to gauge sentiment changes.
The three vertical columns under the TICK are a nice little feature of the platform that provide a tick by tick rolling scroll of new intraday high and low counts. I use this feature 3 times . . one each for the NAZ, NYSE and the indices. so I catch the full flavor of the market trend in real time. When the market's weak, all the scrollers will be running all red and rolling fast. When the markets are about to reverse up, the scrollers will roll very slowly and will be punctuated with periodic green tells of buying. In flat markets the scrollers will be a hodgepodge of red and green and typically scroll by slowly. Coupled with the NYAD chart, these 3 scrollers are invaluable in keeping me on the right side of momentum, even as the market makers are trying to head fake me in another direction.
Below the 3 scrollers, I watch charts of the NYAD and the IWM in 2 minute bars.
Both charts are overlaid with the pivots, 8/16 MAs and the 3/14/3 signal line, which is the moving average component of the MACD.
To the sides, I have 2 charts of the day, typically the Qs and XLF or XLE or IWM if I'm looking for a closer look at the technicals and fibs than the lower middle chart provides. Each chart displays the fibs, parabolics, pivots and 8/16 MAs as well as my technical suite of RSI, %D and StochRSI. Each chart has its own trading window to help old guys like me from getting confused when placing orders and I can instantly toggle from stocks to option string to level 2 options.
I have another SSPro screen that I watch with similar features, but that screen also includes, news, messages, account details, multiple watch lists and a programmable stock scanner, but the trading screen shown is really the core workhorse of the program for me.

Monday channels

Last Thursday I looked at the daily bars kissing off the LR30 channel and suggested more of the same to come. Today's market action was a punctuated reinforcement of those channel breaks and I thought it might be instructive to view my ETF basket in a larger monthly bar time frame.
From the monthly bar perspective, we can see the early capitulation of the XLF and the last ditch capitulation of the XLE. Within my little basket, on a relative basis, XLE has the biggest gap yet to fill to hit 2005 levels currently being revisited by the Qs and IWM.
If the current economic meltdown was confined to US markets, there might be some high probability safe havens to crawl into until the perfect storm blows over. As most sentient beings are aware of, however, this is a global situation, with no immediate or fundamental relief in sight, despite panicky short term fixes various government entities are rushing to put in place.
The chart technicals are all decidedly bearish, and although we may have a few surprise short covering rallies (shorting ban ends Oct 17th), until further notice, the trend is down. To where is anybody's guess but selling volume on the daily bars is increasing in all 4 ETFs, suggesting a high probability that we'll revisit today's low on volume similar to the mid September's records.
As I've stressed many times, IMHO capital preservation is job #1 and I was amused to hear Cramer's latest piece of "must do" wisdom today was to set aside 5 years worth of cash necessary to maintain your lifestyle before investing any excess in the markets. That, of course, assumes you've got any left since Jim's advice is a little late to the party by about 12 months.
My plan is to sit on cash and wait for a washout since I believe current market conditions are unlikely to reverse course until such an event.

Weekly Pivots Update

This little spreadsheet has been uncanny in it's predictive power for (short term) upcoming range expansion or contraction. Per last week's update the indication was for significant range expansion, which is shown by the yellow range tracking bars. Unfortunately, 3 of our 4 ETF basket closed on the low bar of the weekly range, which suggests more negative pressure to come.
For the coming week, only the XLF has a 4 week range/pivot ratio that is greater than the current % pivot range, suggesting relatively muted volatility in that ETF for the coming week. For the past 2 weeks, the XLF has shown the least range expansion of the basket, which may be indicative of a longer term bottoming process in progress.
I'm going to add several more metrics to the study over the next week as well as a graphical interface to thumbnail the data. So far, the study has been a great tell for short term range volatility even in the face of deteriorating trends in all 4 ETFs.
It would be nice to have a coded algorithm for these computations so that a wider universe of ETFs and indices could be quickly scanned for maximum risk/reward potential, but I'm not sure when I'll get around to that little project. If any of you bored coders want to seriously take a crack at it, email me and I'll send the full expanded excel file for you to fiddle with.

Kissing the channel goodbye

Today the Qs and IWM decisively kissed the LR30 channel goodbye. The XLE looks like it might be next to follow suit, but one of my trading survival mantras is to never underestimate the power of big oil to foil the markets. The XLF is actually looking rather bullish relative to the other sectors, perhaps buoyed by positive expectations of the bailout. So far, reaction to the bailout plan has been very neagtive. With the ball in the House's court, expect renewed range expansion.
I didn't know whether to laugh or cry when I read that Al Fishman was paid $19 million for 3 weeks of work as WaMu's CEO, Sept 8 to 25th, while shareholders got zip when WM went down.
Also turns out WM did a little fiddling with the numbers before they went under, so their real asset base is even more dismal than JPM had figured. Hey!!!. Don't you guys ever use accountants to look at the balance sheets before you buy these turkeys?????? Amazing, and an insight into the rampant incompetence and scoundrel mentality that has finally come back to bite Wall Street, with you and I as the long term collateral damage.

Chop, chop

As expected, not a lot of momo in either direction, with my beloved Qs trading narrow range all day, closing almost perfectly on the daily PP pivot. The fib levels confirm the general lack of direction in the Qs today and the options MMs were having a hard time making any headway. The Nov OTM calls saw a boost in the last 10 minutes of the session, and after hours the Qs are up some .45. Odds are for a repeat performance tomorrow unless some late earnings surprises rock the boat.

Another view of the bailout