It looks like this spread becomes narrower whenever it comes close to expiration. If I sell at 0.01000 and close this sell position at 0.00500 each month, can this be a consistently profitable strategy? I will accept drawdown up until 0.03000 which is historically rare. What do you think? /HH spread has limited historic data, so backtesting is difficult.....
I have clients with Natural Gas trades - and right now, May '21 is the most prompt month I have seen. From my experience, when you get prompt on physical futures with intra market spreads it becomes too delta directional with the cash spot market and the spread just becomes a proxy for the spot. It ceases to become a legit spread play on divergence or convergence and that totally defeats the purpose IMO and opens you up to way too much risk. That's why I'm not aware of any clients who got caught the wrong way on May '20 Crude - we stay away from the prompt months for intra market spreads on physically delivered futures for exactly that reason: the delivery wildcard.
Yes, that really makes sense to me too. However, my understanding is /NG-/HH spread is somewhat different from conventional /NG intra spreads. As Feyri explained on page 6 in this thread, /NG-/HH spread has been range bounded within 10 /NG ticks for years until recently and I don't imagine such a huge divergence exceeding 30 ticks happens often (then I can cut loss. My stoploss will be 0.03000). Unfortunately, it's very difficult to find historic data of /NG-/HH spread to justify my strategy for some reason even though /NG-/HH spread trading is one of the most popular /NG spread tradings. Also, many brokers (including AMP or TD Ameritrade) don't clear /HH.
You're on the outside looking in. I don't see the room for arbitrage that you see. You need to be looking at the actual bid/ask - not the synthetic spread. I get this all the time from chartists. The latency between two products produces a "mirage". It's not what is actually achievable.
Thank you very much for your advice. I really appreciate it. By the way, that Barchart image was not synthetic spread chart. I compared it with exchange supported /NG-/HH spread DOM and they are identical. /NG-/HH spread DOM is very liquid (actually, even more liquid than /NG). Margin is only $126 at this time. Otherwise, it couldn't become such a popular product..... I took these images right before (Tokyo open):