Daily Commentary - Posted on Thursday, October 28, 2010, 1:01 PM GMT +1

3 Comments


Oct Thursday 28

COT: Commercial, Non-Commercial and Small Traders

Most recently the Commitments of Traders (COT) reports got (and still get) a lot of press around the blogosphere, regularly picking out the rapid increases in the net short /net long positions of non-commercial and commercial traders (respectively) in major market index futures, regularly pointing to the bearish (rarely the bullish) implications of the current elevated readings.

Three weeks ago I presented an analysis concerning the Commitments of Traders (COT)’s (questionable) quality of forecast due to the fact that on October 5, 2010 commercial traders held a net short position of -30,762 contracts in Nasdaq 100 futures (large contract, not the E-minis), the lowest reading on record (since 4/16/1996), see here. Three weeks later, the Nasdaq 100 has gained (and not lost) 5.00%.

(If you’re interested in a short summary only, please click here)

On , October 19, 2010, the U.S. Commodity Futures Trading Commission (CFTC) reported net positions in ES E-mini (S&P 500) futures as follows:

  • Commercial Traders: 116,804 contracts
  • Non-Commercial Traders: -169,869 contracts
  • Small Traders: 53,065 contracts

Currently non-commercial traders hold their highest net short position in ES E-mini (S&P 500) futures since September 8, 2008, while commercial traders held a net long positions more than 1.75 standard deviations above its 6-month mean (but no record net long position in absolute terms). I thought this would be the right time to check who is the better follow (commercial, non-commercial or small traders) – if any – and their respective ‘quality of forecast‘ (and precision in timing the market).

In order to account for absolute and relative level, I checked for the following setups (what happened if you would’ve followed the respective group of traders either on the short or long side of the market, e.g. one went short when a group of traders switched their net position from long to short, or one went short when positions were reported at an elevated low level, and vice versa):

  • the group (commercial, non-commercial or small traders) switched from a net long to a net short position and vice versa,

and net position in ES E-mini (S&P 500) futures were reported

  • in the bottom / top percentile of the up to then current readings (no hindsight bias)
  • at the lowest / highest level during the last 6 months (6-months low/high)
  • at least 1.75 standard deviations below / above the (running) 6-months mean.

Table I below shows the SPY‘s historical performance over the course of the then following 5 , 21 (regularly 1 month), 42 sessions (approximately 2 months), 63 sessions (approximately 3 months) and 126 sessions (approximately 6 months) sessions assumed one went short on close of a session where the respective group (commercial, non-commercial or small traders) switched from a net long to a net short position (e.g. ‘CM: L -> S‘ = commercial traders switched from net long to net short), and long on close of a session where the respective group switched from a net short to a net long position. In this event ‘Profitable Close‘ x sessions later represents the respective group’s quality of forecast (winning percentage).

Here is the link to the stats in a more ‘readable’, original size: Statistics 1

Whenever any group of traders has switched their net position from net long to net short, historically this has never been a good point in time to close out one’s long position and / or to short the market. The probability of a profitable close over the course of the then following 6 month always had its difficulties to make it above the 50% mark (you’d be better of to flip a coin), leave alone to beat the random chance for a profitable close if you were always long the market (buy and hold approach). And even if any group of traders has switched their net position from net short to net long, none of the groups were able to qualify as above-average market timers (historically they were not able to beat the random chance for a higher close x sessions later).

Table II below shows the SPY‘s historical performance over the course of the then following 5 , 21 (regularly 1 month), 42 sessions (approximately 2 months), 63 sessions (approximately 3 months) and 126 sessions (approximately 6 months) sessions assumed one went short on close of a session where the respective group’s (commercial, non-commercial or small traders) net position was reported in the bottom percentile of the of the up to then historical readings, and long on close of a session where the respective group’s (commercial, non-commercial or small traders) net position was reported in the top percentile of the of the up to then historical readings.


Here is the link to the stats in a more ‘readable’, original size: Statistics 2

Conclusions remain unchanged (and partly not only unimproved, but even worse): Historically elevated levels (bottom or top percentile) in net positions in ES E-mini (S&P 500) futures did not provide – for none of the participating group of traders – an (considerably) above-average (leave alone a reliable) signal to go short  or long the markets.

Table III below now shows the SPY‘s historical performance over the course of the then following 5 , 21 (regularly 1 month), 42 sessions (approximately 2 months), 63 sessions (approximately 3 months) and 126 sessions (approximately 6 months) sessions assumed one went short on close of a session where the respective group’s (commercial, non-commercial or small traders) net position was reported at the lowest level during the last 6 month (6-months low) , and long on close of a session where the respective group’s (commercial, non-commercial or small traders) net position was reported at the highest level during the last 6 month (6-months high).


Here is the link to the stats in a more ‘readable’, original size: Statistics 3

Two notably observations:

  • with net positions in ES E-mini (S&P 500) futures reported at a 6-month low, commercial and non-commercial traders were regularly positioned on the wrong side of the market over the course of the then following 6 months. With commercial traders net positions in ES E-mini (S&P 500) futures at a 6-month low, the SPY was trading at a higher level 21 sessions (1 month) later on 3 out of every 4 occurrences (winning percentage on the short side of the market of 23.29% only), and with non-commercial traders net positions in ES E-mini (S&P 500) futures at a 6-month low, the SPY was trading at a higher level 3 and 6 months later on 3 out of every 4 occurrences as well (winning percentage on the short side of the market of 25.42% and 25.86% only),
  • with net positions in ES E-mini (S&P 500) futures reported at a 6-month high (and with 86 occurrences a statistically significant and remarkable sample size), and looking 1 month ahead, small traders were conspicuously frequent (70.93% of the time) positioned on the right (the long) side of the market.

Table IV below now shows the SPY‘s historical performance over the course of the then following 5 , 21 (regularly 1 month), 42 sessions (approximately 2 months), 63 sessions (approximately 3 months) and 126 sessions (approximately 6 months) sessions assumed one went short on close of a session where the respective group’s (commercial, non-commercial or small traders) net position was reported at least 1.75 standard deviations below the (running) 6-months mean, and long on close of a session where the respective group’s (commercial, non-commercial or small traders) net position was reported at least 1.75 standard deviations above the (running) 6-months mean.


Here is the link to the stats in a more ‘readable’, original size: Statistics 4

Two notably observations again:

  • with net positions in ES E-mini (S&P 500) futures reported at least 1.75 standard deviations below the (running) 6-months mean, commercial, non-commercial and small traders were regularly on the wrong side of the market over the course of the then following 6 months. With commercial trader’s net positions in ES E-mini (S&P 500) futures at least 1.75 standard deviations below the (running) 6-months mean, the SPY was trading at a higher level 21 sessions (1 month) later on 7 out of every 10 occurrences (winning percentage on the short side of the market of 30.14% only), and with non-commercial and small traders net positions in ES E-mini (S&P 500) futures at least 1.75 standard deviations below the (running) 6-months mean, the SPY was trading at a higher level 3 and 6 months later on 3 out of every 4 occurrences as well (winning percentage on the short side of the market at or around 25% only),
  • with commercial trader’s net positions in ES E-mini (S&P 500) futures reported at least 1.75 standard deviations above the (running) 6-months mean , the SPY was trading at a higher level 3 and 6 months later on 4 out of every 5 occurrences and 3 out of every 4 occurrences respectively (winning percentage on the long side of the market at 80.77% and 75.00% respectively)

But one thing to keep in mind:

Non-commercial trader’s held a net short position in ES E-mini (S&P 500) futures of -150,000 contracts (or even more, like on October 19, 2010) over long periods of time in 2007 and 2008 (financial crisis), and since 05/2007 looking ahead 3 and 6 months later they were positioned on the right (the short) side of the market almost all of the time, while they were positioned on the wrong side of the market on almost all of those occurrences before 04/2007, see table V below:

Conclusions:

  • Historically switching from a net long to a net short position and/or elevated levels (net short position in the bottom percentile, at a 6-months low or at least 1.75 standard deviations below the 6-months mean) in net positions in ES E-mini (S&P 500) futures did not provide – for none of the participating groups of traders – an (considerably) above-average (leave alone a reliable) signal to close long positions or short the market, but instead sometimes provided a (contrary) significantly above-average chance (above the respective at-any-time / random chance for a higher close x months later) to go long,
  • when net positions in ES E-mini (S&P 500) futures were reported at an elevated positive level, commercial traders were significantly more often positioned on the right (the long) side of the market than non-commercial or small traders, and especially when commercial trader’s net positions in ES E-mini (S&P 500) futures were reported at least 1.75 standard deviations above the (running) 6-months mean (as of October 19, 2010), the SPY was trading at a higher level 3 and 6 months later on 4 out of every 5 occurrences and 3 out of every 4 occurrences respectively (winning percentage on the long side of the market at 80.77% and 75.00% respectively),
  • but when non-commercial trader’s held a net short position in ES E-mini (S&P 500) futures of -150,000 contracts (or even more, like on October 19, 2010), they were almost always proved correct 3 and 6 month later at least with respect to those occurrences since 04/2007, but positioned on the wrong side of the market on almost all of those occurrences before (fortune or timing ?).

With 3 setups triggered in favor of the long side of the market (non-commercial net short position at 6-month low and at least 1.75 standard deviations below 6-months mean – regularly a contrary signal to go long – ; commercial net long position at least 1.75 standard deviations above 6-months mean – historically an intermediate- and longer-term buy signal) and 1 setup (non-commercial trader’s hold a net short position in ES E-mini (S&P 500) futures of at least -150,000 contracts) triggered in favor of lower prices ahead, it will be interesting to see who ultimetely will be on the right side of the market with respect to the most recent occurrence.

Successful trading,

Frank

Disclosure: No position in the securities mentioned in this post at time of writing.

________________________________

If you might want to be instantly notified about what’s happening in the markets and at TRADING THE ODDS, I encourage you to subscribe to my RSS Feed or Email Feed, and (or) follow me on Twitter.

________________________________

Remarks: Due to their conceptual scope – and if not explicitely stated otherwise – , all models/setups/strategies do not account for slippage, fees and transaction costs, do not account for return on cash and/or interest on margin, do not use position sizing (e.g. Kelly, optimal f) – they’re always ‘all in‘ – , do not use leverage (e.g. leveraged ETFs), do not utilize any kind of abnormal market filter (e.g. during market phases with extremely elevated volatility), do not use intraday buy/sell stops (end-of-day prices only), and models/setups/strategies are not ‘adaptive‘ (do not adjust to the ongoing changes in market conditions like bull and bear markets).

________________________________

Disclaimer

The information on this site is provided for statistical and informational purposes only. Nothing herein should be interpreted or regarded as personalized investment advice or to state or imply that past results are an indication of future performance. The author of this website is not a licensed financial advisor and will not accept liability for any loss or damage, including without limitation to, any loss of profit, which may arise directly or indirectly from use of or reliance on the content of this website(s). Under no circumstances does this information represent an advice or recommendation to buy, sell or hold any security.

I may or may not hold positions for myself, my family and/or clients in the securities mentioned here. Actions may have been taken before or after information is presented, and any opinions expressed in this site are subject to change without notice.

(Data courtesy of MetaStock , and for data import, testing, surveys and statistics I use MATLAB from MathWorks)

________________________________

Comments (3)

 

  1. […] See original here: COT: Commercial, Non-Commercial and Small Traders […]

  2. […] This post was mentioned on Twitter by Frank Hogelucht and ukarlewitz, ukarlewitz. ukarlewitz said: RT @TradingTheOdds: New blog post: COT: Commercial, Non-Commercial and Small Traders URL: http://bit.ly/bwP4gz ($$ $SPX $ES_F $NDX $NQ_F) […]

  3. Michael says:

    As I commented on your last COT study, you cannot use absolute levels. You must use relative levels. It’s much more complicated than that. I usually do a weekly COT analysis on my blog and what I discuss is the relative movement of the trader groups in relation to their past levels and in relation to the price movement. If it was as simple as going long when they flip from short to long then we’d all be millionaires. :)

Leave a Reply

Your email address will not be published. Required fields are marked *