Good Afternoon,
The NWRFC released its 120 day forecast this afternoon and brought us no surprises, no blog fodder to feast upon, just a reasonable forecast (relative to last week, that is) that will fail to move the market or generate strong emotions in everyone but the very neurotic. As such, we’ll use this week’s update to share a few comments on the methodology, sprinkled in with our insight on this week’s forecast.
In this report we compare the block of energy, by month (flat, all hours) for each of the last five forecasts, and compare to the average of the five (gray bar) and to the five year average for all STP forecasts (red bar) for those respective months. The latter value is somewhat obtuse, let us elucidate further. For January, the NWRFC has been pumping out #s every week since August; they did the same in WY16, 15,14, etc. The red column is the average of every forecast released since 2012 for that particular month.
From the above you can then draw a few relative conclusions.
- Dec – Down 250 aMW: it is lower in today’s forecast than it was last week, but is higher than the average of all the forecasts – and higher than the average of the last five (including this week). As for its reasonableness, we are not surprised to see an above normal number given precip that is 2-4X normal since October, plus toss in the approaching cold which the utes will draft to meet load, and it should be higher. What is perhaps the “surprise” is that the forecast, for the balance of the month, didn’t go up given that the 10 Day Forecasts have been consistently higher than last week’s STP:
- Jan – Down 110 aMW; Dec and Jan have been positively correlated since the start, and they typically should be, so no surprise it is down; maybe the surprise is that it is not down more given that Dec includes more of the Oct-Nov precip effects and it includes one assured cold snap
- Feb – Down 150 aMW: same pattern, just walking down, reverting to the mean
- Mar – Down 34 aMW; rounding error, proves they actually run a model
The variance to the 5 year average is highlighted here (NOTE: I turned off the previous 4 weeks and am just showing current, last week, and the 5 year average, by day). The delta in the back-half of Dec warrants a zoom:
From Dec 20 to Dec 31 is off 500 aMW; that has to give small comfort to length given that it is also outside of most reasonable people’s temperature forecasts. Call it bullish, mildly.
In this plot we compare prior years to this week’s forecast but the twist here is that those prior years reflect the SAME week as the current forecast. We all know the NWRFC bakes in a lot of volatility in their 120 day, like some months will see 5000 MW swings over the 120 day life of the forecast, so matching weeks removes some of that volatility. Here we see that the current Dec forecast is actually in line with the prior years, on average, for this first week of Dec. Jan is dwarfing 2015 and 2013, but is less than 2014, which was a very big water year (NOTE: these are actual calendar years, not water years). Similar story in Feb but March sticks out like a thumb that was battered with a sledge hammer – higher than any year we’ve seen in recent times suggesting it is potentially over-stated. Of course, this is a regulation month and much of March’s volatility can be explained by flood control drafts, not all, but much. At this stage we think it (March) is over-stated and will be walked down.
Same as the YOY (Year on Year) report above, except plotted by DOY (Day of Year). Note how similar 2014 and this year are, until Feb. Also note how much additional energy is projected versus 2013 and 2015, like a couple nukes worth. And there is every reason to expect to see that energy hit the market given five year highs in total Mid-C storage.