For anyone familiar with the continuous contract indices ($WTIC, $NATGAS, $GOLD, etc) on stockcharts.com, aren't they supposed to represent the front month contract at all times? Taking a look at the crude chart today, it shows oil closing at $58.16, or $2.50 higher than where the front month closed on NYMEX: Link to chart I've seen this issue happen a lot lately where large movements are not reflected in the continuous contract indices (and currency indices such as $XEU). Is there another free data source for the continuous contracts to compare the daily charts? Does IB offer the continuous contracts in TWS? Thanks. Edit: I tried google first before posting, then toyed with the search string and found the answer: "Like all of our commodity indices, our $GOLD symbol is not technically a gold price at all. It is an index value that corresponds to a theoretical, non-tradable "continuous" contract for gold that uses a weighted average all of the currently open gold contracts. It is provided on an end-of-day basis so that users can compare gold's mid- to long-term performance and trends to other markets. It should not be used to make trading decisions concerning gold itself." Hmmm, I know at least eSignal (and others) represent the continuous contract as the continuous front month, which seems like it would be far more beneficial than a weighted average.. I can't seem to find if IB has them as well.
With continuation charts, you have to properly understand what you're looking at. It's a bigger problem with inverted markets when the rollover occurs. It can seem as though there's a big decline when in fact it didn't "happen". Use the front month contract for shorter-term trading. The continuation chart is better suited for longer-term trading.
TA for $WTIC on Stockcharts.com will be completely wrong for patterns of size 0-4 months. Use CL1! or actual contract name(CLG07) on Futuresource.com(CL1! today is CLG07). Or use USO on Stockcharts.com for the ETF.
That's precisely the time frame I was looking at and I certainly won't be using the weighted-average continuous contract on stockcharts. Thanks for the input everyone. However, wouldn't a front-month continuous contract (provided by Futuresource, eSignal, etc) be sufficient? I understand the rollover effect could skew the technicals when dealing with futures that have large month-to-month spreads (natural gas, etc). For contracts that typically do not exhibit a significant spread between subsequent monthly/quarterly contract rollovers, wouldn't a front-month continuous index be suitable for backtesting purposes?
Indeed: Reuterslink (or whatever the Metastock data is called) and eSignal FAIL to backadjust their continuous contracts for gaps.
But isn't that indicative of the true cost of rollover? If you were holding December crude and then rolled over into January crude, you would have to pay the spread (assuming there was one).
It's a cost(interest rate) because you only put up a small part of the cost of 500 or 1000 barrels of oil that cost $55000(1000 barrels). But it's probably not so much as $1 per barrel. http://realtime.agedwards.com/barchart/sites2/pl/_agep/quote.asp?sym=CLJ5&mode=i&ageContext=public
Same link for QM(eminy) : http://realtime.agedwards.com/barchart/sites2/pl/_agep/quote.asp?sym=qmJ5&mode=i&ageContext=public