Sunday, September 30, 2007

Qs 3rd Quarter Weekly Data

My Qs end of quarter (EOQ) study concludes with this snapshot of weekly performance for 2002-2006. The data field is limited to only 5 years so I am reluctant to bet the farm on the results, but certain patterns do pop out here:
1. The week before EOQ3 has been consistently negative 02-06.
This year is different.
2. The week of the EOQ3 has been close to the zero line 02-06.
This year is different.
3. The week following EOQ3 has been solidly up 4 out of the last 5 years.
4. The 2nd week following EOQ3 has been solidly down 3 out of the last 5 years.
5. The 4th week following EOQ3 has been up 5 out of the last 5 years, perhaps benefiting from October/November cusp buying.

The 2007 data shows impressive strength in the Qs relative to the past 5 years and the historical data suggests that strength is likely to continue through October.
All this sounds great but past performance is no guarantee of future success so caution is ALWAYS advised.

Saturday, September 29, 2007

More on Deconstructing the Qs

This chart is a continuation of the study begun today. My goal was to look at how select ETFs perform 1 month after each quarter and to discern if there were any seasonal patterns revealed that might provide a trading edge. The initial study posted this morning suggested that October might be a good time to get bullish on the Qs and I wondered if the same pattern was reflected in the DIA and the IWM. Based on the above results, the Qs clearly stand out as the best bet, with the IWMs taking second place and the DOW showing good results, but not outstanding. One must also consider the the relative value of each of these instruments: a $9 gain in the $50 Qs reflects a much higher ROI than a $10 gain in the $138 DIAs.
Next I want to look at a breakdown of weekly performance after the quarter to determine if there is a surge pattern. I hope to present those results on Sunday in order to identify possible entry points.

Deconstructing the Qs

A piece by Bespoke on S&P behavior at quarter's end got me thinking. . . so I ran a somewhat similar study for the Qs. I've highlighted third quarter results just because that's where the action currently resides. I've also modified the study a little bit from the Bespoke version: whereas they looked at performance results over the entire next quarter, I'm kind of a short term guy and only looked at performance over the following month.

The results are quite stunning and support a high probability gain in the Qs for the coming month. The study also has important implications for those that prefer to hedge their bets with buy/writes as the OTM calls with maximum premium decay and minimal downside risk become more attractive. The study also suggests a strangle strategy selling puts and calls with the puts slightly below the current strike and the calls considerably OTM.
I haven't had a chance to look at the relative costs of simply buying ATM calls out 2 months. . . that will also be a topic for further investigation.

Over the next few days, as time permits, I'll be running similar analyses of post quarter performance for the DIA and IWM.

Friday, September 28, 2007

A modest backstep


The markets failed to follow through today, as was widely anticipated. We need to see a reversal, or at least a consolidation of recent September gains before blasting off in to the traditional power months of October-January. Also as expected, the Qs showed considerable resilience relative to the other indexes, especially into the close, and although the volume was about the same as yesterday, selldowns were treated as buying opportunities. Cucca does a great job of showing the trap door setup in the Qs this morning, and that was a classic example. I suspect this weakness may follow through on Monday but am also expecting a resumption of the uptrend fairly quickly. Tomorrow I'll post a little study of Qs behavior around the end of the quarter that may help put things in perspective.
The VIX finally got a nudge off the lower band with an initial target of the 10 DSMA which we may see on Monday. Based on the VIX study I cited yesterday, once the VIX departs from an 18% variance we can expect it to hit the 16 DMSA within 4 days 82% of the time. That would return us to a VIX reading of 21, approximately in the middle of the bands and set up for another run down.

Strange goings-on down under

Dr.Brett is off to the land of kangaroos, dingos, great whites, Mad Max, Steve Irwin and other cool stuff to help our Aussie brethren in their quest for trading success. There certainty can be no better ambassador for the cause and I enthusiastically support his efforts to share the wealth of information and inspiration which he so uniquely offers.
I have often wondered what cultural differences mark the Australians and in an effort to educate me an alert reader has brought the following news item from Brisbane to my attention:

Thief rips off woman's underpants in public
By Robyn Ironside and Ursula Heger
September 26, 2007 01:00am
Article from: news.au.com


DETECTIVES are baffled by a brazen daylight attack at Newmarket, in Brisbane's inner north on Monday, in which a woman had her underpants ripped off and bag stolen.
"It's pretty strange. I haven't heard anything like it before," Det-Sen-Sgt Brad Rix said.
He said the 23-year-old was grabbed from behind as she walked home from Newmarket train station about 4.30pm (AEST).
The offender then lifted her dress up, and pulled off her underpants before grabbing her bag and taking off.
"It was absolute daylight, not far from a train station. This person must have felt sure he was going to get away with it," Det-Sen-Sgt Rix said. He said there was no attempt to sexually assault the woman, who did not realise her bag had been stolen until some time later.
"It's possible he may have stolen her underpants as a trophy, or perhaps he intended taking the attack further. We don't know," he said.
I can only imagine the underpants scene from Zoolander.

Thursday, September 27, 2007

Overdue for a hiccup


Today's action looked an awful lot like yesterday. The activity was a bit more subdued and the volume somewhat restrained, but both the TICK and the NYAD actually looked more bullish today than yesterday. For the last 5 hours of the day the VIX, NYAD and TICK were in sync. The financials (XLF) look like they might be getting some legs, but the gaming in these stocks is nothing short of notorious, so I have little interest in the sector for now.
I'm continuing to hold on any new positions. Although the last and first trading days of the month are typically bullish, fourth quarter portfolio realignment by the funds may create some selling pressure and I waiting to see how that plays out. I'm also devoting a considerable amount of time to various performance studies, one on the VIX and several on the Qs, not surprisingly.
The VIX is REALLY getting overdue for a turn. Just to prove it to myself I ran a little TradeStation study (optimizied) and found that over the past 64 months, the VIX has crossed 18% below the 12 day moving average 136 times. If you had bought each of those crosses and held till the VIX crossed over the 16 day moving average (average holding time 4 days), you would have been profitable 82% of the time. I think those are pretty good odds.
TS2000i code as follows:
Inputs: Pcntile(.18), Len2(12), Len3(16);
If Currentbar > 1 and Close Crosses Below Percentile(Pcntile,C,Len2)
Then Buy Next Bar at Market;
If Currentbar > 1 and Close Crosses Above Average(Close,Len3)
Then Exitlong This Bar at Close;
I'm sure you smart programmers out there can improve on this simple system by adding stops and/or conditional clauses.

Wednesday, September 26, 2007

Tech stocks get giddy


Just remember gentle readers. . .you read it here first!. Apparently the Wall Street Journal was so inspired by my comments about the Qs that they wrote a feature article about it. Unfortunately, the WSJ site is subscription based, but if you have an account, read away.
They Qs gapped up at the Open, made another new high and settled down slightly on the upper Keltner channel band to close with a doji. I've pulled back a bit on the chart scale to show the current indicator signals. Typically, the Qs would fade at this overbought level (RSI3>90 and MACD>1) although the pattern of mid-July shows the overbought condition can prevail for some time as price hugs the upper band. The Qs have been on a steady incremental rise since Sept.7th with fairly consistent daily volume as opposed to the volume roller coaster of July and August. Based on current momentum I don't think we'll see a dramatic retracement here, but a pullback is overdue. However, as long as EOD bars continue to stay above the 8 DSMA (plot the low, not the close), the Qs will continue to climb.
The VIX continues to hug the lower bollinger band while maintaining their distance from the 10 and 20 DSMAs. VIX and More also notes this disparity and adds some historical meat to the stew. As with the Qs, the VIX is flashing extreme signals. . .in this case . . .oversold readings . . . the MACD histogram is creeping towards the zero line with the MACD itself registering an whooping -2.5.

Tuesday, September 25, 2007

More of the same


The Qs continue to show strength while the rest of the market wallows in worry and is in fact up .07 after hours on solid volume. That's probably due to at least 2 factors:
1) The Qs have no financial (except net banking), home building or retail merchandising exposure.
2) Microsoft has been on a run up the past two days (Halo 3 and ????)
While supporting the NASDAQ, MSFT's 75M share .48 gain also helped the ailing DOW stay above the zero line. Other than BA, HPQ, HON and UTX the DOW was not showing much enthusiasm today.
Although the Qs are now solidly overbought, that doesn't mean they are going down any time soon. Just look at the July chart reading. We haven't even touched the upper band yet and from a strictly risk management point of view, what are the options for the mutual funds that have to be fully invested and the hedge funds who don't want to be short because of the impending (maybe) Fall market surge. The Beta is with the NAZ and the Qs are the ETF of choice. Expect volume to pick up in the coming weeks as the field gets crowded.
The VIX had a 9% range and flashed a solid red bar into the close. This may be the precursor for the reversal as the VIX is solidly oversold. When we do get the turn I expect the IWM and SPY to give up the most ground and the Qs the least.
The portfolio remain 80% net long and I'm still on hold for Thursday before initiating any new positions.

When math skills fail

My wife and I always enjoy the Del Mar racetrack and we generally manage to at least break even based on a betting system my wife has developed that places little emphasis on the horses and great emphasis on the jockeys. I have learned that being clever with your betting seldom pans out, so I focus on the 30 to 1 long shots and bet them across the board. I typically hit at least one of these on each visit to the track and that's how I make my nut.

Here's an amusing little item from the Sports Column Blog about a guy in Rochester, NY (my hometown) on how not to bet the Kentucky Derby.
We love betting the trifecta. For a buck or two, you can win hundreds if not thousands of dollars on a race. Hell, sometimes we'll even go crazy and go with the trifecta box. However, we also know enough about statistics to know that you can't bet every single permutation in a race and expect to come out on top. But that's exactly what a man in Rochester, NY did. On the day of the Kentucky Derby, a man walked into an OTB and asked how much it would cost to buy every possible trifecta combination. And now we break for a quick math lesson-- If you have 20 horses in the field, the possible outcomes for the first three horses would be 20! / (20-3)! = 6,840. Likewise, a superfecta (first 4 horses) would be 116,280. And now back to your regularly scheduled programming-- The man was told the answer and came back to place the wagers. It cost him a total of $13,680 for his $2 trifectas and he wound up winning... wait for it... wait for it... $440. If you're gonna make a bet like this, you better know what you're getting yourself into. And since the payout odds are terrible on favorites, you should just eliminate those trifecta combinations. But alas, the man had more cash than math skills so now he's $13,240 poorer for his trouble.

Monday, September 24, 2007

Carbon Copy


Today's Qs and VIX performance were pretty much mirror images of Friday in reverse. The Qs continue to look overbought and the VIX looks oversold as it continues to hug the declining(and expanding lower band. Some strange behavior on the VIX today at 11:20 when we got a .50 gap down in 2 minutes, with just little blip in the TICK and no impact on the NYAD.
I continue to track order flow on the NYSE per my Sept 1-6 posts and note that the FICAPs have assumed a 4% larger presence in the markets over the past 2 weeks, although I'm not going to try and explain why.
A likely emerging scenario might include the VIX popping off the lower band and heading back to the 10 & 20 DSMAs in the next few days, with the markets showing some weakness and consolidation. This scneario would provide a jumping off point for an end of month rally that would carry through DOW 14,000 and Qs 52.
Althought these reversals are typically resaged by wide range bars, this market has shown itself time and time again not to be "typical", and the muliple dojis and hammers currently on display may be the tell for the next turn.
Meanwhile, I'm cautiously waiting until Thursday before getting frisky.

Saturday, September 22, 2007

Keeping up


One of the things I do during the typical trading day is backtest various EOD trading ideas on TradeStation to refine my swing trading strategy and tactics. Some of the most profitable and robust systems are fairly rudimentary. Here are the results and code for a simple CCI support/resistance system for the Qs that's profitable over 80% on the long side and 70% on the short side over a 5 year test period. The code is in TS 2000i format, because that's the version I have. Those with 8.2-8.3 should be able to copy the code to create the signals and then dump into the strategy builder. The system works best with select indexes and I leave it to the ambitious daytraders to optimize the inputs and backtest on 5, 15, 30 and 60 minute bars and apply stops appropriate to their risk tolerance.
Note: "Pyramiding" is turned ON in Strategy Builder.

Inputs: CCILen(6), OverSold(-130), OverBought(148),
OverSold2(-130), OverBought2(150);

If Currentbar > 1 AND CCI(CCILen) > Overbought
Then Sell Next Bar at Market;

If CCI(CCILen) < OverSold2 Then ExitShort at Market;

If Currentbar > 1 and CCI(CCILen) < OverSold
Then Buy Next Bar at Market;

If CCI(CCILen) > OverBought2 Then Exitlong at Market;

Friday, September 21, 2007

Volatility falters


Qs followed through to the up side, contradicting yesterday's technical signals.
The VIX had a 7% day, continuing to hug the lower band while simultaneously maintaining a 15% gap below the converged 10 and 20 DSMAs.
I was frankly surprised and caught a bit off guard by today's strength, since the TICK readings and a flatline NYAD were not reflective of market activity. I faded the Qs and XRX just after the noon hour on reversing 8 minute bar parabolic signals, but closed out both trades at 2 pm with modest gains in both.
The strength today was moderate and narrow based, perhaps reflecting the multiple head fakes and feigns that transpire during expirations, and I continue to bide my time until the later part of next week before initiating larger scale trades (think more risk).

Thursday, September 20, 2007

Market slides



Qs gave back a little today, but not without a fight. Technicals (especially RSI) point to further declines. First stop is 49th parallel. Historically, September closes poorly (Stock Traders Almanac) with 13 out of the last 16 years posting negative readings for the week following triple witching. Statistical odds favoring the long side pick up considerably starting next Thursday and I'm cherry picking my trades until then.
The VIX had another narrow range day, flashing a doji. The VIX is still over 15% below both the 10 and 20 DSMA and the RSI has now turned up suggesting more of the same to come.
The Qs/VIX RSI signals are in sync and seldom reverse at these levels.

Wednesday, September 19, 2007

Qs looking overbought


We are getting RSI signal confirmation between the Qs and the VIX suggesting the markets may fade the Friday expiration. As expected, the Qs did carry through with a pop at the open and then faded down for the rest of the day.
Dr. Brett has a different spin on current conditions and makes the case for continued strength.
FYI - Buying yesterday's close and selling at the open today was pretty much a sure thing given the substantial momentum of yesterday's last hour. CXO ran an interesting little study of buying the close and selling the open for the Qs, DIA and SPY. Adding a couple filters like a 3 DSMA, a 10 DSMA, a STO on the rise above 60 or an RSI on the rise above 60 can increase the results substantially. This is an EOD system that doesn't require a lot of maintenance, but you need to be prepared for those days when the news goes against the grain and you get a gap down at the open. More variations of this system this weekend.
The VIX had a narrow range day, flashing a doji and closing precisely on the lower band and hinting that a reversal is imminent (amplitude and duration unknown, but 20 DSMA is likely target). VIX and More also makes a case for higher VIX readings in the next few sessions.

Tuesday, September 18, 2007

VIX plunges


The market liked the news, except for the short sellers, who got snookered. Qs are back to 50 on moderately large volume. Let's see how things shake out by the end of the week, when triple witching transpires. The week following TW tends to be less that stellar and that's when the short sellers may make another run at it.
A whooping 23% drop in the VIX today brings it down and through the lower band and 20% below the 10 DSMA. I don't believe this condition can be sustained short term, although we may see a little follow through tomorrow prior to a reversal.
I went delta neutral on 30% of the portfolio this morning and ended up leaving some nice potential gains on the table. Nevertheless, the residual portfolio produced a good return today and because of my risk adverse attitude, I don't regret my hedging. I'm in a holding mode for the next few days until I see how the current pop plays out.

Monday, September 17, 2007

Drum Roll.....please



The Qs gave up some ground today on less than enthusiastic volume. This was pretty much as expected as we have pulled back to intermediate resistance at the 49th parallel.
The VIX performed today exactly as VIX and More had predicted on Sept 12th, with a 8.5% range and a 6.5 % rise, finishing the day with a classic doji. The concept is the pre news, pre event, pre earnings and pre FOMC volatility spikes as uncertainty of the outcome nears.
Interestingly, although the VIX is expected to revert back to the 10/20 DMSAs in the next few days, it does not follow that the markets will rise.
The general tone of the market today was negative and more than a few commentators have noted that 25 to 50 basis points have already been priced into the markets in the past few days. Regardless of what the FOMC announces, this may be a case of buy the rumor, sell the news, so I have adopted a strong defensive position against my otherwise long portfolio. Extreme caution is advised for tomorrow as the action will be fast and furious.

Sunday, September 16, 2007

Figure the odds

At the 1994 annual awards dinner given for Forensic Science, AAFS President Dr. Don Harper Mills astounded his audience with the legal complications of a bizarre death. Here is the story:
On March 23, 1994, the medical examiner viewed the body of Ronald Opus and concluded that he died from a shotgun wound to the head. Mr. Opus had jumped from the top of a ten-story building intending to commit suicide. He left a note to the effect indicating his despondency. As he fell past the ninth floor, his life was interrupted by a shotgun blast passing through a window, which killed him instantly. Neither the shooter nor the deceased was aware that a safety net had been installed just below the eighth floor level to protect some building workers and that Ronald Opus would not have been able to complete his suicide the way he had planned.
"Ordinarily," Dr Mills continued, "Someone who sets out to commit suicide and ultimately succeeds, even though the mechanism might not be what he intended, is still defined as committing suicide." That Mr. Opus was shot on the way to certain death, but probably would not have been successful because of the safety net, caused the medical examiner to feel that he had a homicide on his hands.In the room on the ninth floor, where the shotgun blast emanated, was occupied by an elderly man and his wife. They were arguing vigorously and he was threatening her with a shotgun. The man was so upset that when he pulled the trigger he completely missed his wife and the pellets went through the window striking Mr. Opus. When one intends to kill subject "A" but kills subject "B" in the attempt, one is guilty of the murder of subject "B."When confronted with the murder charge the old man and his wife were both adamant, and both said that they thought the shotgun was not loaded. The old man said it was a long-standing habit to threaten his wife with the unloaded shotgun. He had no intention to murder her. Therefore the killing of Mr. Opus appeared to be an accident; that is, assuming the gun had been accidentally loaded. The continuing investigation turned up a witness who saw the old couple's son loading the shotgun about six weeks prior to the fatal accident. It transpired that the old woman had cut off her son's financial support and the son, knowing the propensity of his father to use the shotgun threateningly, loaded the gun with the expectation that his father would shoot his mother. Since the loader of the gun was aware of this, he was guilty of the murder even though he did not actually pull the trigger. The case now becomes one of murder on the part of the son for the death of Ronald Opus.
Now comes the exquisite twist. Further investigation revealed that the son was, in fact, Ronald Opus. He had become increasingly despondent over the failure of his attempt to engineer his mother's murder. This led him to jump off the ten story building on March 23rd, only to be killed by a shotgun blast passing through the ninth story window. The son had actually murdered himself so the medical examiner closed the case as a suicide.
A true story from Associated Press

Friday, September 14, 2007

VIX Contraction Continues


After an initial drop, the Qs and the rest of the market quickly recovered up to yesterday's close level at the PP pivot. . .and then did basically nothing for the rest of the day. . . to finish down .02. The Qs are flashing a peek-a-boo overbought signal, continuing to show strong resistance at 49.28. All the buying was in the first hour on 24M shares after a completely manipulated drop of .42 on 3.5M shares IN THE FIRST 4 MINUTES ! So much for random motion in the markets.
The VIX popped up at the open then faded down with the rising market and then traded in a narrow range for the rest of the day to close up .16. The Bollinger Bands are continuing to contract and hug both the 10 & 20 DSMAs. I expect more of the same on Monday.
The NYAD was mildly bullish with a .97 median reading, while the TICK showed wide range 5 minute bars but no clear slope.
This was not a day to initiate new trades and the markets will probably continue to flounder until Tuesday afternoon, when you know who will open the flood gates. Caution is advised at this point, since the trading logic that previals around these events is best described as bizarre.

Thursday, September 13, 2007

VIX Contraction



The markets rallied again today, but volume was not convincing. Compounding market contortions, the NYAD managed to hold a relatively narrow range but TICK was downslope for the last 2.5 hours and the VIX was decidedly upslope. The program traders' footprints were all over the place today and I'll be dealing more with that issue this weekend.
The Qs are now showing under performance relative to the DOW. This may be a case of catch up, as the reverse has been the situation for the past 2 weeks. The Qs are also slamming against substantial resistance at 49.50, but are not overbought at this level. If we do get an intraday pop to 50, I'm inclined to view that as an opportunity to fade.
On a 6% range day the VIX went nowhere (.25), so we are seeing a volatility contraction at work here. This may be a a precursor for next week's triple witching, which according to the Stock Trader's Almanac has a truly dismal performance history (think DOWN), or the pre-FOMC contraction noted by VIX and More and referenced yesterday. I'm watching this situation closely to see how it plays out tomorrow. If the VIX continues to exhibit contraction in the face of a wide range market move I would be inclined to sell the VIX on Monday.

Wednesday, September 12, 2007

Rally stalls


Oil was the news today, along with umpteen prognostications on what effect the FOMC meeting will have on the markets. Vix and More profiles an interesting little study on VIX behavior pre and post FOMC that's worth a look. I took the Qs data and looked at FOMC meetings back to Feb 2005 to see if there was any confirmation. The results were statistically insignificant. Tomorrow I'll run the same test on the SPY to see if there's any correlation.
The Qs finished just pennies off the open after a relatively low volatility day. We are currently mid-signal with no clear direction.
The VIX had a 7% range day and finished 1.01 below the open.
Neither the TICK nor the NYAD showed much enthusiasm today, so it was difficult to find good trading candidates. Today was also curious in that the VIX was down, along with the IWM, $DJII, $SPX and $COMPX (not expected), while the Qs, the DIA and the SPY were all UP.
I welcome any thoughts on explaining this market-wide divergence.

Tuesday, September 11, 2007

Markets surge


The Qs got a big lift today along with the rest of market, continuing the uptrend begun yesterday. At current levels (48.91) the Qs are back to upper resistance, although there's no sign of an overbought signal in the RSI. The last 10 minutes today saw a tremendous surge of volume and price across the board, suggesting more of same to come.
The VIX had another 10% range day, descending to the 20 DSMA, which is now about to cross an upsloping 10 DSMA. I failed to notice yesterday that the VIX was 10% above the 10 DMSA, which is often a tell for imminent reversals. It sure played out that way today and the technicals suggest more short term downside in the VIX.

Monday, September 10, 2007

VIX back at upper band


The Qs took off like a bronco out of the chute, promptly hitting R1 before descending down to S1, and then taking off to hit R1 again, only to fade to the PP pivot at the close. That was a wild ride accompanied by volume slightly above the 2 week average. The Qs reached for 47.5 support but didn't quite make it. I think we'll see that level broken Tuesday, as the market finished weakly.
It never ceases to amaze me how the pivots act like magnets for many of the bigger ETFs and if the Schwab/CyberTrader charts could be displayed clearly on Blogger I'd be more than happy to illustrate the point.
The VIX had a 10% range day, blowing through the upper BB band, then retreated to stick there. In sync with the Qs signal, I think there are higher levels to come as the VIX is only marginally overbought and the current markets are exhibiting both accelerated and exaggerated moves relative to the Spring.

Do you have what it takes ?

Reuters has a pictorial hint on what it may take to succeed as a trader in today's volatile markets.

Futures & Options Trader

A couple interesting articles in this month's issue. The Trader Interview with Pedram and DeVore (p.34) puts a little different spin on selling naked options and the article on Rolling Leaps Calls (p.22) is something for conservative investors to consider in lieu of buy and hold. This is a monthly e-magazine and subscriptions are free. Some good regular authors and features.

Friday, September 07, 2007

The herd heads South


Storm clouds broke this morning and the herd headed South in a hurry. When the NYAD opens at .09 and the VIX gaps up 10% in tandem, you better know it's gonna be bad. The VIX is now back above the 20 DMSA, and although it finished on a doji the RSI is strongly hinting at a further rise.
The Qs opened at S2 and then fell to S4 in the first 2 hours on heavy volume before attempting a midday rally back to S3 before falling back to finish near S4 on another volume surge. That's a BIG drop and the Qs closed on the highest volume in 15 days. The 20 DSMA at 47.50 looks like the next target with both the RSI and MACD heading down. In afterhours trading the Qs are up .13, sharing a modest afterhours lift in the DIA, SPY and IWM.
Nevertheless, the closing TICK was -540 and the NYAD never made it past .29, so I am suspicious of the aftershours lift as indicative of anything meaningful. It's still a long ways to Monday's open and the premarket Globex will tell the real tale.

Algorithmic Trading links

I received a number of emails from traders asking for additional info on algorithmic trading, trickle filters and order defragmentation programs. Below are a few links to sites that may put this type of trading in perspective. Special thanks to Tom Henderson of www.camron.com.au/ for sharing his considerable knowledge of this cutting edge trading technology.

Thursday, September 06, 2007

Calm before the storm?


Qs opened with a pop, then a drop, then a pop, then a drop, then a pop, then a whimper finish. The black hammer is mildly bullish and while the 3/7 DSMA shows no immediate danger of crossing, the MACD is beginning to turn down. Somewhat odd was the fact that the $compx was up and the Qs were down.
The pop in the DOW was basically the result of 5 stocks (GE, KO, MRK, MSFT and UTX), so it wasn't exactly a wildly bullish day. Although the NYAD was upslope for the last 2.5 hours the TICK provided little confirmation to support additional long positions. There is still a palpable amount of uncertainty hovering around the financials along with the lurking fear that we may see a repeat of Aug 16th (looking for that double bottom W) and I suspect many folks are sitting on the sidelines waiting out the volatility chaacteristic of September.
The VIX had a quiet day with only a 2.5% range, still sitting between the 10 and 20 DSMAs and riding the MACD zero line.
This is beginning to feel like the calm before the storm and yesterday's caution advisory is repeated.

NASDAQ order flow

This is part 4 of the series on order flow begun on Sept. 1st and covers data available on the Nasdaq. The NAZ is an 100% electronic marketplace, which gives it a decidedly different character than the NYSE. The NAZ is driven by bid/offer spreads provided by market makers, which the NAZ does track and which numbered about 892 on a daily basis for August. But the reality is that with the anonymous electronic market, anyone can be a market maker, just submit a bid/offer.
While the NAZ does not segregate retail and program trading, it does identify "block trades", which it defines as trades over 10,000 shares. August averaged 8,783 daily block trades, which accounted for only 0.15% of all trades but 15% of total volume. Being home to most of the high beta stocks, the NAZ is the favored venue for day traders, so maybe the piddling NYSE retail numbers just mask the real action on the NAZ. But wait a second, looking back to our NYSE data we see an August daily average of 5.8M trades as compared to the NAZ's 6.0M trades. We also see NYSE average daily volume at 1.86B and Naz at 1.88B. Given the magnitude of the numbers in question, the correlation between them is rather surprising.
Discounting the block trades, we see that the remaining 99.85% of the trades account for 85% of the volume and yield an average of 265 shares/trade, a number that is midway between the NYSE retail and program trading trade share average. These numbers strongly suggest that program trading and algorithmic trickle filters are accounting for the bulk of trading volume but the nature of the electronic marketplace creates a veil that is difficult to penetrate.
As was the case with yesterday's study of WM on the NYSE, as I look at EBAY on the NAZ this morning I note 3.9M shares traded during the first 90 minutes and as I watch the L2 time and sales go streaming buy, the vast majority of the trades are 100 shares, a few 200s, a few 400s and an occasional 1000 share order, or about 14,700 orders = 163 trades/minute. The WM study produced 12,500 orders and 140 trades/minute and this further supports the contention that orders are being algorithmically partitioned and queued.
How awareness of these trickle filters can benefit the average trader will be the subject of upcoming research. My next series of studies will focus on NYSE volume tells and will begin in October.
Anyone wishing to further investigate NYSE or Nasdaq data can refer to the links on the right side of this blog under "Data and Charting links". Then click either "NYSE Data" or "Nasdaq internals". NYSE gives you 14 days of info free, Nasdaq gives you 5. I tried to collect AMEX data also since this is the domain of the large volume sector SPDRs like XLE, XLF, XLB, XLU, etc, but AMEX charges $30/day for a peek at their numbers so I will have to defer.
As I mentioned in Part 1, I'll be glad to forward my study as a consolidated Excel spreadsheet to anyone who wants to massage it further.