12 replies to this topic

[arbman]

From NAV's thread below...

I am yet to discover again whether there is any edge left in the COT data as far as the commercials are concerned or perhaps when there is any edge to read...

Why do you think this huge shift by the commercials should be now considered bullish? Because they not only hedged, but also loaded up? I don't get it... If only, they probably hedged the heck out of their (short) positions...

Of course, the important is the small speculators and they are definitely at the wrong side...

Also look at OEX Put / Call ratios (short-term implications) and Commitment of Traders (long-term implications). It is not easy to justify a bearish case.

Denleo

You may be seeing some posts this weekend about the *massive*, absolutely massive, bullish changes in the large S&P futures contract in this week's COT. Then again, the bears will point out the bearish shift in the smaller contract.

Based on the above, I don't think the commercials have anything over the other S&P traders over the last year. They were net short from 1340 up to 1460, net long from about 1480 down to 1375. If anything it looks to me like the large specs fared a little better than anyone else.

Edited by kisacik, 28 September 2007 - 05:47 PM.

[Gary Smith]

Kisa, I am not all that much of a COT devotee. It worked well in the 80s through around 1996 where it became less predictive for several years. But it was spot-on during the 2000-02 bear, the birth of the 03 bull and gave some nice sell signals in May 06 and February 07. It gave a nice buy in March 07 but admittedly wasn't very accurate during the recent 10% decline. However, those with a longer term time frame may quibble with the later. It's becoming a bit too en vogue as some who had no clue to what it was a year ago now quote in on a regular basis in their analysis and there are several COT blog sites. I look only at the net difference between the three classes of traders and at least in recent memory, I can't recall where the commercials have been this net long while at the same time the small specs have been this net short in the large contract. The recent September quarterly expiration had some impact on the drastic switch and will be interesting to see if next week's report is more normalized. The current COT report reminds me of the old bull configuarations of the 80s and 90s. It's the simple that works best and the past year and a half the best bullish predictor has been Zweig's 9 to 1 signals which has been covered extensively on Fib's site. And of course junk bonds which cracked before the 10% stock decline and now have more than stabilized. As for your froth I tend to agree. China stocks are no difference than internets in 97/98, tech in 99 and early 2000. Then you have the natural resource stocks and the two stocks that saved my butt this year WYNN and LVS. Lots of froth in lots of sectors but the froth can go on for a long time before it is time to pay the Piper.

[arbman]

Thanks for your take. If there is any edge in the COT, it is probably the small specs, the commercials are always hedged with something, same with large specs, they are not really directional bets, I suppose. But the small specs are really doing this directionaly in the futures. If they are not leveraged, the trend continues, when they are at the wrong side, the trend accelerates and when they all pile on, the fast trend first slows down and then eventually reverses... The above chart kind of explains it... I think one must follow the action of the small specs first and then the other big guys in the COT, if there is any value in there. The position sizes must also make some sense, I like sentimentrader.com's COT charts too, Jason analyzes with some sort of a BB around it... Da small specs went long with the momentum last week and immediately reversed as they saw no follow through. I can not blame them, I am one of them, but I am neutral, not short. - kisa

Edited by kisacik, 28 September 2007 - 08:22 PM.

[arbman]

Unless they changed their methodology this is another indication of optimism... Some people that never gave a bullish vote during the past 3 years are now bullish...

[Gary Smith]

Yes, the Blogger Poll is worrisome and here's another poll showing lots of optimism for next week.


http://www.tsptalk.com/sentiment.html


And you also have Investors Intelligence showing an ever widening spread in favor of bulls.

[arbman]

Well, da_cheif calls this the recognition wave, I will call it the liquidity wave in a few months. This much devaluation in the USD can not happen without serious printing, and more printing is coming, the rates are lower. This will have an inflationary effect and it will be compounding for months to come. However, other than the immediate repricing effect of the earnings multiples, da_cheif also suggests the same, there is no compounded liquidity effect washing the markets yet, or the real growth... So, the market perhaps struggles here a little further, this much devaluation must be hard to swallow for some of the foreign investors. But I doubt this is a time to be (too) bearish, you will never know when they will start pouring the money into the markets, I guess the commodities and commodity stocks will be still a better place to be... It is clear to me, no deflation will be allowed and cash is trash! - kisa

Edited by kisacik, 28 September 2007 - 10:42 PM.

[jack]

From NAV's thread below...

I am yet to discover again whether there is any edge left in the COT data as far as the commercials are concerned or perhaps when there is any edge to read...

Why do you think this huge shift by the commercials should be now considered bullish? Because they not only hedged, but also loaded up? I don't get it... If only, they probably hedged the heck out of their (short) positions...

Of course, the important is the small speculators and they are definitely at the wrong side...

Also look at OEX Put / Call ratios (short-term implications) and Commitment of Traders (long-term implications). It is not easy to justify a bearish case.

Denleo

You may be seeing some posts this weekend about the *massive*, absolutely massive, bullish changes in the large S&P futures contract in this week's COT. Then again, the bears will point out the bearish shift in the smaller contract.

Based on the above, I don't think the commercials have anything over the other S&P traders over the last year. They were net short from 1340 up to 1460, net long from about 1480 down to 1375. If anything it looks to me like the large specs fared a little better than anyone else.

Looking at the net positions in the above graph looks dramatic.
Separate the longs from the shorts instead of subtracting them from each other and what we see is:

Comm long closed positions
Comm short closed positions
Large traders long closed positions
Large traders short closed positions
Small traders long closed positions
Small traders short closed positions
Spreaders closed positions

Everybody took it off the table, short commercials and small longs took more off than the rest but
long commercials closed about the same #contracts as small longs. Who got less geared and who got less
hedged, I dunno.

[jack]

BTW Not trying to point out interesting pattern in COT. Just the opposite. Regular quarterly cleanout.

[*JB*]

Kisa and Gary -- (Gary, good to "hear" from you, it's been a while. Kisa -- not to leave you out -- you always post very thoughtful and intelligent comments. I much appreciate them!) To keep it short, I did a "study" about 4 years ago that took the better part of 6 months of spare time -- a myriad of portions (and "ratios") of COT data that I hand entered into Tradestation and came to the conclusion that the best approach (BY FAR) to using COT data (in an IT trading style) was fading the small specs...IF the approach effectively filter out the noise. Interestingly enough, TASC published an article several month ago saying pretty much the same thing, that included a system (and it's results) that faded the small specs, using very recent data. I didn't follow up and test their system, but -- IMO -- it's worth looking at if you have access and are interested. BTW - these kinds of threads are THE reason I still stick 'round these parts. Great exchange of ideas and experience!

[jack]

To keep it short, I did a "study" about 4 years ago that took the better part of 6 months of spare time -- a myriad of portions (and "ratios") of COT data that I hand entered into Tradestation and came to the conclusion that the best approach (BY FAR) to using COT data (in an IT trading style) was fading the small specs..

LOOK DON'T THINK for sure
but applying this to emini cot or big contract cot, fading small longs or shrts looks like an expensive hobby.
I realize you are probably not looking for further comment from me but my interest in cot data
is genuine.