Perhaps Geoffrey Streatfield, a British lawyer writing in The Observer in 1850, summed it up best when he said, "Facts speak louder than statistics." Or you might prefer Mark Twain's prophetic, "He uses statistics as a drunken man uses lamp-posts - for support rather than illumination." Regardless of your preference, there is truth in these witticisms. Nowhere is that better illustrated of late than in the U.S. rig count, the North American gas price and economic recovery in the upstream sector. Last year, everyone said, "Wait until the rig count moves up. Then the recovery will start." Later they echoed, "Wait until the rig count reaches 1,000. Then we will see the momentum shift."

Well, the U.S. rig count has been over 1,000 for a while. Just yesterday at lunch I heard that the magic number is now 1,200. At the same lunch, we all allowed that things are beginning to look - maybe - a bit brighter. Then someone issued a challenge - "Show me," he said. We couldn't.

The same thing can be said about gas prices. Three dollars a MMcf was what we needed, we all claimed. Then, maybe it was US $4 or $5. Well, we have danced around those numbers, or above them, most of the year. And, guess what? Not much has happened besides the rig count going up a bit.

And it's not that things won't get better. Lehman Brothers Original E&P Spending Survey has concluded that the 330 oil and gas companies they surveyed will increase 2003 worldwide E&P spending by 5.4%, to $138.8 billion. That number is some $5.4 billion higher than Lehman's December forecast number. But it is not the wholesale industry rebound that many predict and we all await.

The bottom line is that the sets of statistics we have counted on to predict activity don't seem to be working. That is the trouble with statistics. They are, or should be, cold, hard facts whose utility lies in analyzing past activity, but seldom in forecasting future activity. Returning to Twain, it seems we have been using various statistical counts as supports rather than illumination. We should let them illuminate the problems of relying on statistics in an unfamiliar market rather than leaning on them for unwarranted support in the same market.

For those of you who have an abiding fascination with the development of technology, I would like to recommend two fascinating books. Both are by Henry Petroski, the first entitled The Evolution of Useful Things and the second To Engineer is Human. In these books, both by Vintage Books, N.Y., and two others (Beyond Engineering and The Pencil), Petroski explores the rather lengthy design and technical processes that have produced, mostly, ordinary devices such as the paper clip and the can opener. The process to perfection of these everyday devices is not nearly so straightforward or casual as one might imagine. Each took numerous iterations and years, if not centuries, to perfect. Tin cans (actually tin-coated cast iron originally) for food preservation, for instance, were first introduced by Londoner Peter Durand in 1810, following a progression of more primitive storage devices. However, there was no can opener. According to Petroski in The Evolution of Useful Things, "as late as 1824 a tin of roast veal, carried on one of the explorer William Edward Parry's Arctic expeditions, bore these instructions for opening: 'Cut round on top with a hammer and chisel.'"

These books are not only highly entertaining, they reflect on the way all things are engineered and built. They examine, among other things, the value of failure, the role of precedence in design and the importance of commercial potential. If you are fascinated by the development of technology and the engineering process, they are a good read.


Comments