Final Numbers In – Updates on California Grape Prices
To start this blog post off, I want to rewind to the very first post I put up here. The following excerpt basically sums it up:
“So far, this year, I’m hearing two different stories. The first story comes from the grower reps from the big wineries and from Silicon Valley Bank. The story goes something like this: “High grape prices are compressing our margins and with two big harvests leading to a lot of product to go through our sales channels, grape prices have to come down.”
All the farmers I’ve spoken to, however, are reporting that they do not intend to reduce prices. Allied Grape Growers has put out a chart showing that bearing acreage growth will not exceed the long-term average trend line until around 2018.
Now, I’m a big fan of SVB and I’d be hard-pressed to identify a better source of information on the market, but they, like the grower reps, have a vested interest in signaling to the market that winegrape prices need to drop. They back a lot of wineries. Why is it that when the going is good for growers and the wineries are still profiting that we hear about the need for a correction, but when vineyard owners can’t pay their bills we just hear that times are tough for everyone? But I digress. And I’ll admit growers have their own biases.
So where does the truth lie?
The average price per ton for winegrapes from California are estimated to be: $740.64, a decrease from $746.49. That’s less than a 1% decrease. Basically, look for prices to remain steady.”
As the season progressed, I stuck to my guns and, quite frankly, I began to needle SVB’s Rob McMillan a bit. I’m sure he thinks I’m pretty annoying, if he has even made note of who I am. For the record, let me reiterate that I actually think he’s been great for this industry and really admire what he’s built at SVB. In response to my needling him, he did clarify that he saw North Coast prices rising, but Central Valley prices pulling grape prices down overall. I always stated that my forecasts showed prices holding steady, an equilibrium between the bottom falling out of the Valley and continued price rises in coastal regions.
I was right. In fact, prices technically rose, but by only .63%. Of course I make my living making these predictions, so it’s not just about being right, but about how right was I?
So, how did I do? I had said California winegrape prices would remain essentially flat and they did. However, they did inch up a bit, whereas I felt they would inch a bit down. In that sense, both I and Rob were wrong. In fact, my predicted price ended up 2.38% below the actual prices for 2014. I shoot for getting everything within 2% and, until this year, every forecast I’ve done was within 2.5%, so that’s not too bad. But, since I will be evaluating all of my public predictions, I’ll start with this one and work through all the ways in which I analyze my own performance.
How Far Off Was I?
Like I said, I was off by 2.38%.
How Many Standard Deviations Was I Off By?
More important to me than the nominal difference between my predictions and the actual numbers, are the probability distributions for those numbers. In this case, my prediction was off by 0.66 standard deviations. If I get things right, 68% of my predictions will come within one standard deviation of my prediction. This is more important than the percentage amount because I can’t ever tell a client that prices WILL be such and such. I tell them that prices are X likely to fall within Y and Z, so the client can plan around that. So, being within one standard deviation is good news. What does it mean, though? It means my model predicted a probability of about 25% that prices would rise by an amount equal to or greater than the amount it actually rose by. Perhaps an easier way to explain this is to state that my model implied a roughly 41% chance that prices would actually rise, instead of dropping. So this is a good result.
Plain Language: Was I Right about Where Prices Were Going?
Yes, I said they would be basically flat and they were.