Re-cap of Friday’s whipsaw moves in stocks and commodities

  • Posted on: November 28, 2009

This is a brief re-cap of today’s market action. It may have been Black Friday for US retailers, but early in the AM of Friday, it looked like it might be Black Friday for US investors. The shock of The Bank of Dubai asking for a delay in its interest payments til may, on their 60BB, along with the Yen move to a new multi year high against the US dollar, were the two catalysts which set off corrections across the board in both the stock indexes and the commodities.

The bears were rewarded, if they were awake and didn’t get stubborn, with 38 and in some cases 50 percent retracements of the recent 6 week rally. Every day, bearish traders couldn’t believe there wasn’t a correction. Typically it came when most US traders were in a tryptophan induced coma. Most of the move was finished by about 2:30 AM CST, and then, we actually began the creep back in the face of bears who woke up late and decided to look for the back-draft lower. By and large, the stock indexes, metals and grains all rebounded nicely considering the whipsaw nature of the correction.

The Fibonacci numbers were uncannily accurate. The issue is where to start the retracement from, and than comes just with experience and practice.

For the record a quick recap of the retracements and rebounds across the market.

In SPZ, we had a high At 1098.25 and a low at 1067.00… On the charts, beginning with a Nov 2nd low at 1026.00 and moving to the November 16th high at 1,112.25, the 50% retracement was 1069.50.. As is usually the case, the Fibonacci is not the stone cold low of the move, but it is usually very close. So many people look at them, its the level that’s more important than the specific chart point prediction.
SPZ rebounded nicely, to settle 23 handles above that low, at 1089.25.

In the DJZ, we had a a target 38 percent retracement at 1,143. The low at 10,116 was within 30 ticks. In fact, just enough to probably stop out some people who bought the 1040 area with too tight a stop. And watching the trade jump around, 20 or 30 ticks, due to the lack of liquidity, was kind of scary. A couple of large orders came in 2 times, sweeping the screen lower. You could see the quantities on the side of the screen as every 2 and 3 and 1 lot was swept as the programs were dropping sell bombs. For the next wave down, if we have one, there is support at the 50 percent level at 1004, basically 10,000. No surprise there. Below that the 62 percent retracement is at 9937. If we get there, that will certainly create some interesting headlines.

Next, moving on to the Metals and Crude Oil. Gold had a hellacious move last night, as first people did an early flight to quality, punching up to the 1,195 level, and then the bears got a hold and crushed it down to the 1,130 level. A 65 dollar break.
Looking at the Oct 29Th low at 1,026.90 low and then the rally up to that high print at 1,195.00 it was impressive that the correction 1) took place so blindingly fast, and 2) stopped uncannily at the 38 percent retracement level at 1,130.00 level…
If we have another shoot lower, there is support at 1,110.60 ( 50%) and then at 1,091.30 (62%)

The crude already explored the Fibonacci retracemnt. Friday’s low at 72.39 was right at the 62% retracement at 72.33. You must take the Sep 25th low at 66.10, measure up to the Oct 21 high at 82.58. Even though the crude oil rebounded to settle at 75.97.. we could look forward to more volatility. But again, usually, we have some time to trade over a week or so. The move we had Thanksgiving Day Eve through Friday Morning, was, in one word, freaky… I can’t remember a move like that, except for possibly the Chernobyl disaster in the 80’s, where gold and grains rocketed like that. It takes a certain set of circumstances to get that level of volatility.

The Grains took it on the chin in sympathy with the break in the crude. Jan beans completed a 38 percent retracement, just nipping through the exact point at 923, by moving another 2 cents, posting a low at 921. It crept higher through the night rallying up to the 935 level. When 9:30 CST came by on Friday morning, the contract posted an early low, ended up rallying back up to the 1056 3/4 level, to settle at the 1052 1/4. All in all a nice almost 40 cent rally back off the low at 921. SF will have support at 1008 3/4 (50%RT) and 994 3/4 (62%).

CZ doesn’t have as clear a pattern for a good Fibonacci retracement. It opened lower on Thursday’s evening session at 386, slid to a low at 379 and then traded sideways through the rest of the night session. When Friday’s pit traded opened, the bulls took control, with CZ posting an early low at 382, and then trending higher up to 398 1/2, to finish the day up 6 cents at 397.

Finally we have WZ. It had more of a similar chart pattern over the past month as the Jan beans (SF), with a major low on Nov 2nd at 487 3/4, followed by a 3 week, one dollar rally up to the Nov 18th high at 583 1/2.

Its Fibonacci retracement resembled that of Jan beans. The 38 percent retracement was reached at 547. The 50% retracement is where it found traction. That point is at 535 3/4. WZ found support at 530 1/2. The next area of support for WZ is the 62% RT at 524 1/2. At the time of the move, however, it looked very possible that the market would move down to the 524 1/2 level.

Its only another 6 cents, and the market was jumping in 3 cent barrages. So, just knowing the Fibonacci numbers are not enough. One must have the order resting, with a protective stop above or below, in case, indeed, the market moves from the 38 percent to the 50 percent to the 62 percent, in a heartbeat.

The Use of Fibonacci retracements still requires a trader to take risk. Very often, its 4 or 5 minutes after the chart point has been tested, that it looks like it was either a great buy or a great sell… But when the market is trading there, one has to make a decision and live with it. And very often, the market is attracted to the Fibonacci area, and then does a head fake, first honoring the support or resistance level, then violating it, giving the trader the choice of getting out or sitting, depending on where he has placed his stop loss order.

Without a stop loss order, though, its very easy to buy the 50 percent pullback, only to watch the market continue on its merry way to the 62 percent area. And then at that point, the trader is faced with the choice of getting out or riding a loser to oblivion… Or worse, adding to the loser by re-buying the 62 percent break, looking for a bounce back for a push or a small profit. A very bad option, because it can snowball into a large loss for a trader who does that.

Finally, i think that very often if people have their stops set too tight in these areas I believe there are computer worm programs which go after those stops. I can’t prove it, but its the one glaring difference between open outcry and screen trading.

I think someone has written programs to fish out stops, based on standard deviation movements around new highs or new lows.

I also think these programs are the type where a 100 lot order might be triggered, but the program is written to pull the whole package as soon as 1 contract of the package trades.

Again, a human being would not be capable of doing that.
If a human trader had offered a large order, say 100 or 200, and another broker or local said “I’ll take 50 or 150, that other trader or broker could never say, “wait. wait. wait… one only…” No one would have ever traded with them again, because that person was not honoring his or her market.

On the screen, that’s the whole way business is done. No one honors anything, unless the program is written for the program to stay until filled. Most of the programs are designed to offer 20 contracts, but pull 18 after the first two get hit.

That makes for added ridiculous volatility, again, because a computer can be one thousand times faster than a human being bidding and offering with his or her voice, and hand signals. I think that half of the orders visible on the screen are meaningless, due to the fact it could be a program which offers a large order, but cancels the whole thing if one or two contracts of his order actually gets filled.

It makes for flighty markets, and it has changed the face of the spread markets. The spreads between different option months used to be a stabilizing factor, prior to the onset of computer trading. Very often the outright month would have a 20 cent trading range, while the spread market would only have a 2 cent move. Now, the spread market, moves like the out rights. The spread market which was once extremely deep and liquid has lost that trait. For example, July/Nov Spread in the Beans, basically old crop vs new crop, has radically changed the way it used to trade. It has morphed into a new creature, totally different than it was even 2 years ago.

Anyhow, there’s no going back. You just have to realize that when the screen is showing 200 on the bid or offer, there very well could be just 20 there, or it could be the first 200 showing of a 2 thousand lot. Instead of offering the whole 2000 lot, the program is designed to offer it 23 contracts at a time, to disguise the real order.

Again, that wouldn’t have been possible when business was done in the trading pits via open outcry.

Have a Great Weekend,

And Please Send Me a Comment or Complaint or Criticism, what ever. I look forward to hearing from you all.