Will Oil Hit $135 a Barrel In 2008?

A reader, under the belief that I am bearish on the future prices of oil, on April 22nd made me the following wager:

I'll bet you $100 that the price of oil will hit $135 by the end of the year. You can write about it in your column.

He misunderstood my objections to people who suggest the price of oil is going to skyrocket in the very short future. I disagree with them, not because I believe the price of oil is necessarily going down. Rather, anyone who really, truly believed that oil prices were going to skyrocket would be buying up oil futures and options - not wasting their time e-mailing me.

Although I had no interest in the bet, I thought it would be interesting to take a look at how likely this is to happen. I derided the bet by saying that given the price of oil at the time was $118.28:

A bet that sometime between now and 8 months from now (since there are just over 8 months between April 22 and the end of December) the price of oil will be $17 different than it is today? I can't imagine a bet that goes less out on a limb.

But that statement is a poor substitute for doing research. I decided to try to determine roughly how likely this is to happen. I could have done something fancy using but I decided to keep things simple. I decided to answer the question:

If someone had made this bet in the past, that is bet that oil prices would hit a level $16.72 higher in the following 8 months, how likely would they have been to win?

I took a dataset with six years worth of oil prices (2001-2006) that I used to develop MERT. There were 1304 trading days between January 2001 and April 2006 where one could have made this bet.

Results of $16.72 Increase Bet

In only 188 days of 1304 (14.42 % ) would this bet have paid off - that is in only 188 of the days would the price of oil be $16.72 higher at some point in the next 8 months as it was that date.

Interestingly there was exactly one date out of 1304 where a bet that oil would drop by $16.72 or more would have paid off - April 21, 2006. The fact that this paid off far more on the "up" bet than on the "down" bet should not come as a surprise, as the price of oil went from $27.29 on January 1, 2001 to $60.85 on December 29, 2006.

By this measure, it turns out that for this period a $16.72 rise in the price of oil in a 8-month period was quite unlikely. I quickly realized, however, that I was probably asking the wrong question. The price of oil on November 15, 2001 was $17.50 - so oil prices would have to double for there to be a $16.72 rise. A $16.72 rise (or fall) would seem to be far more likely to occur during periods of high oil prices than low ones. So perhaps what matters is the percentage change.

I then examined the bet from the following point of view:

If someone had made this bet in the past, that is bet that oil prices would rise to a point 14.1 % higher in the following 8 months, how likely would they have been to win?

Results of 14.1 % Increase Bet

In 989 days of 1304 (75.84 % ) would this bet have paid off - that is in 989 of the days would the price of oil be 14.1 % higher at some point in the next 8 months as it was that date.

Although the price of oil more than doubled over this six year period, in 442 of the 1304 days (33.90 % ) the price of oil dropped by 14.1 % or more. And if you had simply bet that the price of oil would rise or fall by 14.1, you would have been right 1266 out of 1304 instances - 97.09. This goes to show the volatility of oil prices, and we haven't even included the roller coaster which was 2007.

A very rough estimate would suggest that it is about 76 likely that oil will hit $135 at some point in 2008. This is a very rough back-of-the-envelope type calculation - 2008 will not necessarily have the same level of volatility as 2001-2006, not necessarily have the same upward trend, and the sample size involved is not particularly large. However, as a rough estimate, it seems plausible. $135 oil this year would appear not to be quite the slam-dunk I assumed it was, but it does appear more likely that not.

Predictions for Other Price Levels

With this method we can construct a crude estimate of the likelihood of oil hitting other price levels between April 22, 2008 and the end of the year:

$69.09: 0.08 % - (a price fall of 41.6 % ) - 1 time of 1304
$80.00: 8.05 % - (a price fall of 32.4 % ) - 105 times of 1304
$100.00: 28.76 % - (a price fall of 15.5 % ) - 375 times of 1304
$135.00: 75.84 % - (a price rise of 14.1 % ) - 989 times of 1304
$150.00: 49.00 % - (a price rise of 26.8 % ) - 639 times of 1304
$175.00: 8.97 % - (a price rise of 48.0 % ) - 117 times of 1304
$199.34: 0.08 % - (a price rise of 68.5 % ) - 1 time of 1304
$200.00: 0.00 % - (a price rise of 69.1 % ) - 0 times of 1304