Mining the data: What film will you next see?

In the May/June issue of Intelligent Life, there is an article on algorithims which shows how the company, Epagogix, uses data to determine who is a bankable movie star. Brad Pitt, Johnny Depp and Will Smith are actors who have demonstrated over time to bring a return on the investment. This has nothing to do with whether the film is any good, or if it's even art. But algorithims help Epagogix guide studios in their decisions. (The article went on to say that a certain unnamed A-list actress is guaranteed to lose money. I wonder if it's Julia Roberts?)

The Hartford Symphony recently did a popular all-Mozart program, and it was difficult to get a ticket -- even with four performances. Later in the season, a program featuring two pianists in music by Gershwin also did well. Unfortunately, the season closer, featuring Stravinsky's Le sacre du printemps, did not draw audiences of similar magnitude.

The Hartford Symphony is right to perform The Rite of Spring, and like most orchestras, they wanted to schedule it when the rest of the world was celebrating the work's centenary. But honoring that date -- May 29, 1913 -- means that you must close your season with it. And Stravinsky is never going to sell like Mozart or Gershwin. However viable artistically it may be, closing your season with Le sacre is a calculated risk.

Enter algorithims. Again, we are not talking about the artistic level of these performances, but rather the interest these programs would likely generate within the community. (For what it's worth, I attended the Mozart program -- featuring concertmaster Leonid Sigal as soloist and guest conductor -- and it was very good. Sometimes box office and artistic merits can be on the same page.)

Creating a symphony season is quite a puzzle. During my years as music director of the Hartford Symphony, there were always checks and balances to what I planned. Initially, the process was a bit arduous. The year before I arrived -- without a sitting music director -- the Program Committee had planned the entire masterworks series. With their newly minted success, they wanted to continue having an active role. What they did not understand is that programming is one of the most important things a music director does. I wanted to play to my strengths, and to that of the orchestra, and who would know better than a conductor?


Once you have a first draft of a season, then it's time for careful scrutinizing. And one of the first questions will be, "will it sell?"


Mozart always sells. Vivaldi's Four Seasons and Beethoven's Ninth Symphony are guaranteed box office hits. But you can't fill your whole season with these works, otherwise. . . . what are you going to do the following season? You must be judicious, and plan ahead.

There is also the matter of concertos -- for an orchestra that does several masterworks programs over the course of a season, the HSO always features at least one or two pianists, sometimes more. Then you must decide which concerto will be played -- will it be one by Rachmaninoff? Audiences love just about anything by Rachmaninoff, but they also have preferences: the Second or Third concertos, for instance, or Variations on a Theme by Paganini. But if you schedule the First or Fourth concertos -- both wonderful, if neglected works -- audiences may not come (and the less discerning ticket buyers may feel cheated).

Sometimes you get a surprise hit. With the help of Christopher Stager, a marketing whiz who continues to guide me and many other orchestras and opera companies, he suggested that if I wanted to do Berlioz's Romeo and Juliet, it would be best to position it near Valentine's Day. Now, there are lots of people who love Beethoven, Bach and Brahms, and maybe even one or two works by Berlioz (Symphonie Fantastique, for example), but few would admit more than a passing knowledge with this great work. Well, Valentine's Day arrived, and sure enough, just as Chris had predicted, scores of young men brought their beautifully-coiffed dates to the Bushnell, hoping to get tickets. Some did, but many others were denied entry, as the box office staff were caught completely off guard by the last-minute onslaught. Chris's suggestion had been a master stroke; if we had put Romeo and Juliet on any other month of the year, it would have gone by with nary even a dull roar.

If you need a sure thing, put Brad Pitt in the movie, or put Carmina Burana on the docket. But for the rest of the year, you'd better have the data on your side. Without it, you can only hope to get lucky.