Good Morning,
Bruce said it well when he wrote that song, it is always a good feeling to be back home. This summer I’m going to be working on my “APT” blog and today will be a kickoff of that effort. You can’t do a trading blog without looking at the fundies since a big part of my trade tactics are cash-driven. Occasionally, there are trades so worthy of owning you ignore cash, but rarely and that is not the case today, so let’s look at some quick fundies and then we’ll grind through my first crack at an APT-based blog.
Demand
Signs of new heat but those hints are short-lived and mainly fall on the weekend.
Those red circles are hot days, Spokane boasts a 105 and probably sets its CY17 highs on that day, but it’s a Saturday. Seattle and Portland also clock in near the mid 80s, perhaps low 90s. A waste of a hot event, plus it is lasts but a few days, so barely registers on the Richter scale (not Jeff, the earthquake one).
California has the same event and Sacramento will be very hot, even Burbank should post high 90s. Summer heat often has a tendency to do two things: Grow and Linger. Watch this in the 12Z later today and the 0Z for tomorrow.
Palo continues to burn for another week, then it crashes and burns. Well, not really, but Phoenix does post its first below normal stretch in a few months, we call that crashing and burning. Yet it remains hot this week and has an end of week spike into the mid “Teens”. Still a scary hub and one to short with caution, though shorting paid well at the end of last week.
Loads are mostly down, week on week, though the Mid-C is more sideways than not. Strange, because we saw some heat this weekend.
Burbank almost posted a 100, LV and PHX are still around 110., but the Northwest was just “comfortable”. We were surprised, kind of, that the ISO didn’t post some stronger numbers yesterday:
Especially at the BPAT node given very tight water and decent demand. Also note the Palo blip in HA yesterday, two ticks with 900 handles. Most telling was the lack of volatility in the HA market all weekend long.
Hydro & Generation & Transmission
We’ll just lump all of these together since none of them are doing a lot, though hydro is falling hard in the Northwest:
Coulee is shaping hard, all the big cuts are now at night; the project realized a sub-40k discharge through the turbines Saturday morning. You’d have to travel back to last October to see those levels. And the reservoir is filling quickly, now around 1289 and functionally full. That factoid is more bearish than anything else, pretty much what comes in now must go out, though less and less is coming in :
Inflows to Coulee are below normal, significantly so. Contrast current levels with the ten year highs, or even average. Water is officially tight in the northwest.
Noms are tight, too. Palo remains off the charts, possibly explaining those 900 ticks yesterday; there just isn’t much gen left to turn on, and now PV1 is at 80%, not sure what that is all about. That unit is scheduled for a fall refuel. The Mid-C noms are especially bullish – especially given how weak cash traded going out on Friday.
The ISO has a few gas units out, not much more than normal, which is actually bullish since a running unit has but one alternative state – not running.
The Northwest is an exporting machine, aside from the selling taking place at the Northern Border. Interesting that the Canadians did not buy a single MWH over the last week, every day was a sell. We think it speaks to their strong reservoirs and still significant snowpack, and weak loads. In other words, Canada will remain a gray, perhaps black, cloud hanging over Mid-C’s summer.
Conclusions
I created a couple of new dashbaords that I’ll be using in my “APT” blogs:
I’ve broken the traded products into Months and Quarters (Includes Cal 2018 and we’ll be adding Cals 2019-22 later this summer). In this table I restricted the months to a rolling six, so the table includes periods out through Jan-18. We are not suggesting you can trade Jan 18 in July 17, but think it can shed some insight into how the Q1 and Q4 interact.
The table is a 2X2 with APT summaries for the four derivative types. The first field is the internal ranking of that particular product against everything else. So Rank is rank against All Derivatives and All Periods. The highest ranked outright is MidC Sept Offpeak (#25), let’s see how the plots look:
APT likes the trade because it likes buying cheap and the market is near a contract low. Note the forecast is at a low too, but that is driven by STP and we think those govt forecasts will be backed off, especially in light of the rapidly declining inflows to Coulee and the vanished snow.
The second “best” is also a LL, Palo LL for Sep. We choose to ignore that, one LL is plenty to put on, so let’s look at the third spot – Shorting July Palo:
Now this is a trade not for the feint of heart, check out that volatility over the last couple of weeks. We put out a short rec last week here and got paid, am not sure I’d stay short, not after seeing a nice 10% reversal, and seeing where the hub’s noms are (and a possible nuke issue) and heat that stays in the 110s. This chart speaks to a key problem with any algorithmic trading strategy since it ignores some very important attributes which should be considered.
While on the subject of BOM, let’s look at MidC:
Like Palo, the northwest hub got sold at the end of last week, rightfully so, but now it is back to being a tempting long. We’d like to see a $25 handle, but doubt we will, and given the state of the state we’d be indifferent about going long here, though I don’t think we’d short it again at this level.
There are a lot more interesting charts, we’ll let you click. Let’s look at the Qs:
The top-ranked Outright is shorting NP Q2, here is the chart:
The Q is trading near a contract high, would be a contract high except for the irrational exuberance last October over the Wash Carbon Tax. We were a bit surprised to see it trade so high after a very big water year; often the market projects the current year onto the forthcoming one, but that doesn’t appear to be the case at NP. That said, we’d short it at those levels and not worry about it, though I’d disagree that it is the best Q outright and would suggest clicking through a few more plots.
The top ON|OFF is also at NP – NP Q3:
Easy to see why the model likes this, the product has bounced off a contract low twice and now appears to be rallying up. The forecast volatility is a fraction of the market, too, which is one of the critical ranking components.
TIP: Set the filters to exclude hubs you don’t trade, or Periods you choose to ignore. I’ve included all periods for Qs and Cals and all four hubs. If you need help contact Garrett or Bill to make your dashbaords.
The top ranked roll is at SP: Q3 18 to Q2
The model recommends a buy here because its near a contract low and has started rallying. Obviously, it was a better buy a week ago but there should be some more room to run and as California works through its water crisis (too much) the Q3 17 is going to post some strong numbers which will make this roll widen further.
The top ranked spread is shorting NP-MidC Q2 HL:
Contract highs but is now range-bound. Both hubs had huge WY17 water years in Q2 and ultimately this trade will be determined by the MidC’s WY18 year, which we know nothing about. The model likes shorting highs, which is why it picked this as the #1 spread trade.
That’s it for today, I’ll be working on this APT blogs all summer and start tracking the calls.