Friday, 8 June 2007

COTs Twiddle Thumbs

Well, it sure was a crazy week in the markets. Yesterday's horrible close, then today's rebound. I hope you got through it okay. What did the Commitments of Traders say this week? Not much. No new signals for my system. Which I think in fact may say a lot.

I'll ponder the data some more over the weekend and post an analysis Monday or Tuesday. Starting this week, I'm going to concentrate Friday afternoons on putting out a timely report on the latest COTs data, then work on an analysis piece for the following week.

Meanwhile, I've been re-reading Edwin Lefèvre's classic Reminiscences of a Stock Operator and came across a poignant passage: "No, sir, nobody can make big money on what someone else tells him to do. I know from experience that nobody can give me a tip or series of tips that will make more money for me than my own judgement."

I strongly agree. It got me to thinking about a new format for reporting what I think the COTs data is saying. You'll see no more "buys" or "sells." Instead, it will be "bullish," "bearish" or "neutral."

This is to underline my view that, while I think the COTs can provide valuable market insights, you should do your own research when it comes to your hard-won bucks. It's also a reflection of the fact that my trading system is still in development. I'm refining and validating it as I'm trading it, but that's a risk I'm willing to take with my risk-management strategy. I'm in no position to tell anyone how they should trade their own money.

New Signals*


Renewed Signals**

-Soybean Oil
-Natural Gas***

Existing signals (date of original signal in parentheses)****

-30-Year Treasury Yield (3-Jan-07)
-10-Year Treasury Yield (20-Mar-07)
-S&P 500 (20-Mar-07)
-NASDAQ (27-Mar-07)
-Semiconductor Index, symbol: SOX (30-Mar-07)
-Dow Jones Industrial Average (20-Dec-05)
-Russell 2000 (25-Mar-03)
-Nikkei (3-Feb-04)
-TSX (20-Mar-07)

-S&P/TSE Canadian Energy iUnits ETF, symbol: XEG.TO (3-Apr-07)
-Oil Service Holders ETF, symbol: OIH (18-Apr-06)
-Soybean Oil (2-May-06)
-US Global Investors Funds US Gold Fund, USERX (30-Jan-07)
-S&P/TSE Canadian Gold iUnits ETF, symbol: XGD.TO (30-Jan-07)
-Gold Bugs Index, HUI (13-Feb-07)
-Canadian Dollar (10-Apr-07)
-U.S. Dollar Index (3-Oct-06)
-Gold (13-Feb-07)
-Copper (10-Apr-07)

-S&P 400 Midcap (3-Jan-07)
-Crude Oil, Light Sweet (3-Apr-07)***
-Natural Gas (27-Mar-07)***
-Silver (21-Nov-06)

* For an explanation of what I do after a new signal, click “How It Works” to the right.
** A “renewed” signal is when a market is already on a bullish or bearish signal, and traders again register an extreme net trading position in the same direction. I normally ignore renewed signals unless I don't already have a trade on in this market. I haven't studied the profitability of trading on renewed signals.
*** See my special caveats for my Crude Oil and Natural Gas setups (click “Profit/Loss Results” to the right and check the footnotes).
**** The date in parentheses is the date of the COTs report that gave this signal - not the date I executed the trade (which can be up to four weeks later). For details on how I trade this system, including trade delays and portfolio allocations, click on "How It Works" and “Profit/Loss Results” to the right. These "existing signals," which are mostly several months old, are listed here as a reference, not a trading recommendation.


Anonymous said...


I see your system ranks BULLISH the
"30-Year Treasury Yield (3-Jan-07) and the 10-Year Treasury Yield (20-Mar-07)". Do this mean your system is bullish on bonds and notes, or does that mean it is bullish on yields (i.e., that rates will go up and bond and note prices will go down).



leon t said...

alex: i for one still like to call a spade a spade. ther's nothing wrong with your buy and sell signals. if it makes you feel more comfortable, post a disclaimer.(educational purposes only badabim...) and there, you are covered. if your mind is fixed on the matter , not much i can do to change it. but i'd much prefer the buy/sell format.

an alternative if you insist in the meek bullish/ bearish format. post bullish since 6/8/07, bearish since 2/3/00. not as effective but at least more precise. you could also indicate for the less gifted members of the flock, such as myself, that positions could be initiated when the report indicates a bullish or liquidated when bearish or neutral readings are given.
i see a lot of ambiguity in the new approach. but then it could be me. and no you are not telling me what to do if you keep the buy/sell format.

Alex Roslin said...

Hi Jim,

Thanks for your question. It's the latter - yields going up, note prices going down.

Take care,

gks said...


It is my understanding that the COT report data is 3 days late.

So we will not know what the commercials etc.. did during the 2nd half of this volatile week.

Is this correct?

That is a bit of a limitation.

Is there a way to get daily updated date information on the opening closing of futures/options contracts?

Also it is my understanding that the COT is only a small window into world trading.

All kinds of other markets are trading the same instruments but only those that are required by law and domicile in the US report in the COT. (some even skip it by doing it OTC - but that market may be small)

What I am implying by this is that just becuase the COT shows -say commericals reducing their short positions on gold at a certain rate - this may not be particularly bullish if them or others are on another market (say in Japan) increasing Gold short positions at a faster rate.

So the COT in this respect has severe limitations.

Further - every pro trader reads the report - so how valuable can it really be?

What is widely known is already discounted in the market.

Also do you know why the US goverment tried to discontinue the COT?

Perhaps I am just being too tough on the COT or am I right.

Thanks for taking the time to answer my questions.




Alex Roslin said...

Hi Leon,

Thanks for your comment. I understand your point, but I feel a certain responsibility about making these kinds of calls - especially when some readers may not appreciate my disclaimer that past performance is no guarantee of future results.

Also, not everything will see the disclaimer in the first place that's at the bottom of each page on this site.

It all goes to the idea behind this blog - that we should investigate the markets ourselves. After all, look how often my system fades the small traders and large specs! The data is strong evidence that retail investors and professionals alike are getting it wrong more often than right.

I was starting to think that giving outright buy and sell signals goes against this idea of empowering investors.

But I will consider seriously your point about making the new format as clear as possible.

Thanks again,

Alex Roslin said...


Thanks for your comment. You're correct to say the COTs data is valid as of the Tuesday prior to the Friday on which it's released (except for holidays, when it's released Mondays).

It's not a perfect situation, but at least better than before when it came out every two weeks or monthly and cost money to get.

Your other points are valid too. It doesn't contain all global market data by any means. In some cases, it's possible to correlate COTs data with other markets, like my setup for Toronto's TSX index, which is correlated to S&P 500 COTs data.

Regarding discontinuing the COTs data, the Commodity Futures Trading Commission did put out a call for comments on whether the data was considered still valid and should perhaps be discontinued, but the overwhelming response convinced the commission to keep it around. The problem was the rise of index trading, and the commission's solution was to break out a new data series for these traders.

This new data is still too young to be of much use in terms of spotting trends, so far as I can tell.


gks said...

Alex thanks for the prompt response.

A couple of other points I wondered:

The commercials can balance their position in the cash market.

So for example this week they may have been short S&P in cash while maintaining their longs as shown on the COT.

Is this necessarily bullish?
It's more like neutral - I mean if something did change this week in the market - the commercials can't start dumping - so they hedged going short S&P cash - then they exit later.

I also wanted to ask you if you have any examples from past data of when the commercials (or whoever) were long the S&P but were wrong and had to reverse position?

Was the COT a good predictor of this reversal or was it late?
We know its early on the bullish signal. And we know its early on a bearish signal (since they exit slowing in a topping market).

But what if they are caught off guard? What the COT signal be late in this case?

I guess what I am asking is:
Is it not possible for the market to enter an extended correction even with a supposed bullish COT signal?

Thanks again


Alex Roslin said...


Great questions! The COTs signals for the S&P 500 and other markets are far from being spot-on all the time, but I've tried to use the data to find the setups that are most consistently profitable while having the smallest maximum drawdown and beating the markets by the greatest margins.

But you have only to check the maximum drawdowns and win/loss results for the setups on my "Profit/Loss Results" page to see that the signals aren't always correct.

I think the best answer to your questions is to suggest you look at the samplke S&P spreadsheet I've offered to send readers. (Email me at to get one if you haven't already.) It doesn't contain the S&P 500 weekly open prices, but you can easily obtain those from a site like Yahoo Finance.


Anonymous said...


Thanks for the clarification on the bonds/notes yield question. One more question: can you give us some insight on how your system uses the COT report for debt instruments. I've noticed for some time now that the commercials have been **very** long bonds, and short the notes (that situation alone confused me to no end). I assume your system grades the levels and triggers when a certain level of bullishness (long activity) kicks in, and I would guess we haven't reached that threshold yet.

Thansk again,


Alex Roslin said...

Hi Jim,

I know, it's a strange situation. Oddly, I found that the most profitable setup for the commercial traders for the 30-year Treasury Bond yield involved fading them at historic extremes, while the most profitable setup for the large specs involved trading the same side as this group. That's the opposite of the usual practice of fading the large specs and following the commercials.

I think this shows, again, why there needs to be more research into this valuable data, instead of reliance on the popular, yet untested notions out there of how the COTs work.

I rely on a combination of these two setups for my trading signal for the bond yields, considering a trade only when both signals concur.

My best setup for the 10-year Treasury Note yield is more "normal": It trades the same side as the commercial traders.

Best regards,