Felix Salmon

Why the quants won’t take over Hollywood

Felix Salmon
Feb 9, 2013 07:50 UTC

Andrew Leonard has a very odd column about Netflix and House of Cards, under the headline “How Netflix is turning viewers into puppets”. Netflix, you see, has lots of data, and it used that data in the commissioning process for the series:

Netflix’s data indicated that the same subscribers who loved the original BBC production also gobbled down movies starring Kevin Spacey or directed by David Fincher. Therefore, concluded Netflix executives, a remake of the BBC drama with Spacey and Fincher attached was a no-brainer, to the point that the company committed $100 million for two 13-episode seasons.

It should go without saying, of course, that dropping $100 million on a 26-episode remake of a great TV show is never a no-brainer. For one thing, for all that the original series is extremely good, it was also very timely, coming as it did at the end of Margaret Thatcher’s transformation of the Prime Minister’s office into something much more powerful and Presidential than the UK had ever seen. The BBC series tapped into Britain’s fear of the possible implications of that power, as well as the fact that Richard III and Macbeth are deeply rooted in the national psyche.

More generally, remakes are inherently dangerous things: what producers think of as a “proven formula” more often turns out to have been a unique and inimitable confluence of creative electricity. And it goes without saying that the better the original was, the less likely it is that the remake will surpass it.

But Leonard doesn’t see any of those risks, he just sees science, quoting a Netflix flack waxing implausibly about how the company is “able with a high degree of confidence to understand how big a likely audience is for a given show based on people’s viewing habits”. And then Leonard takes that PR fluff and turns it into a lesson about the fearsome implications of Big Data:

The companies that figure out how to generate intelligence from that data will know more about us than we know ourselves, and will be able to craft techniques that push us toward where they want us to go, rather than where we would go by ourselves if left to our own devices. I’m guessing this will be good for Netflix’s bottom line, but at what point do we go from being happy subscribers, to mindless puppets?

We’re never left to our own devices, of course: billions of dollars’ worth of marketing, programming, and other expenses are designed precisely to make us watch this rather than that. But at the same time, we humans somehow stubbornly refuse to become mindless puppets, and our tastes tend to evolve in wonderfully unpredictable ways.

One example of this is Netflix’s own first foray into production, Lilyhammer, which turned no one into puppets mainly because no one actually saw it. But the best example isn’t Netflix at all, but rather Relativity Media. If you think that Leonard is overly credulous about the power of Netflix’s Big Data, wait until you see Chris Jones, profiling Relativity’s Ryan Kavanaugh in 2009. He opens with Ron Howard cooling his heels in Kavanaugh’s waiting room, and then explains just what it is that gives Kavanaugh the power to keep the Hollywood A-list waiting like that:

Before Relativity commits to financing a particular movie — either through its slate deals with Sony and Universal or on its own — it’s fed into an elaborate Monte Carlo simulation, a risk-assessment algorithm normally used to evaluate financial instruments based on the past performance of similar products. Enough variables are included in the Monte Carlo for Wilson and his team to have reached the limits of their Excel’s sixty-five thousand rows of data: principal actor, director, genre, budget, release date, rating, and so on. After running the movie through ten thousand combinations of variables (in marathon overnight sessions), the computers will churn out a few hundred pages that culminate in two critical numbers: the percentage of time the movie will be profitable, and the average profit for each profitable run.

In fact, of course, what gave Kavanaugh all that power is exactly the same thing that gives any other Hollywood producer power: ready cash. In Kavanaugh’s case, the money came from Elliott Associates, the New York hedge fund. Which expected to get hedge-fund-like returns from its investment in Relativity, and instead lost money.

For some reason, there seems to be a huge amount of appetite for anybody saying that Netflix is being incredibly clever here. Rebecca Greenfield’s column desperately trying to work out how spending $100 million on this series could possibly make sense has now racked up more than 100,000 views. But the base case scenario for Netflix is exactly the same as the base case scenario for any other rich outsider walking into the shark tank that is Hollywood. Stars like Kevin Spacey and David Fincher will happily take Netflix’s money for however long Netflix is willing to spend it — as will the studios charging Netflix top dollar for the rights to stream their back catalogues. It’s a lovely new revenue stream for the industry, but it doesn’t mean that Netflix knows what it’s doing.

The truth of Hollywood is no mystery: as William Goldman famously said, nobody knows anything. Sometimes, people have hot streaks, and when that happens, David Carr will write a gushing column about what might be called the anti-Netflix approach: ignore the numbers and the heuristics, and just go out there and take creative risks. And in general, the biggest rewards always accrue to the properties which came from nowhere, doing something startling and new. Conversely, formulas only work until they don’t, and the problem with the Relativity approach is that it’s pretty much guaranteed to hit that inevitable failure, if it keeps on churning out formulaic movies.

With hindsight, the biggest risk that Netflix took with House of Cards was not getting Andrew Davies to write it. Stars and directors are all well and good, but if you’re aspiring to the highbrow, as Netflix is with this series, you need great writing first and foremost. The BBC series, written by Davies, was some of the best-written television ever, at the time; it was The Wire of its day. The remake, by contrast, has cringe-inducingly bad writing, from which the best acting and directing in the world could never recover. (Not that a great writer guarantees anything: even the incomparable William Goldman had more than his fair share of flops.)

In order to realize that a script isn’t up to snuff and needs to be comprehensively rewritten, you need a producer with more than just a Monte Carlo simulation: you need someone who can not only hire talent but can fire it as well. And in order to create the kind of television which will resonate and become a cultural touchstone, you need an impossible-to-formulate cocktail of creativity, inspiration, teamwork, and luck. The House of Cards remake is perfectly good, but it’s not that good. And, in turn, that’s why we, the viewing public, will never be puppets, dangling on the end of some TV quant’s strings. TV’s quants are clever, to be sure. But clever is easy to come by in Hollywood. And it’s never been remotely sufficient for success.


These people should really read Art De Vany’s Hollywood Economics – the algorithims are a complete waste of money – the statistics of Hollywood are too wild for models.

Posted by NotPredictable | Report as abusive

The private-gig kings

Felix Salmon
May 1, 2012 21:43 UTC

It doesn’t surprise me in the slightest that the Monday-night entertainment at the Milken Global Conference this year was Lionel Richie. At 62 years old, Richie is pretty much the median age of conference attendees, and is a perfect calibration of the familiar and the inoffensive, combined with a frisson of star power.

There’s another artist with much the same combination of vintage and talent, who’s also a staple on the conference circuit: Kenny Loggins. And I’ve heard that each of them has quietly become one of the most financially successful touring artists the world has ever known.

I’d love to know if there’s any public record of how much Richie and Loggins tour, and/or how much they get paid for a typical gig. As I understand it, both of them are consummate professionals when it comes to entertaining crowds of men in suits, and they pretty much exclusively confine themselves to events where they’re paid by a single organizer, rather than having to go to the trouble and uncertainty of selling tickets to the public. The Milken event was atypical, I think: Loggins tends to do more of the corporate-conference events, while Richie seems to have unrivalled numbers of invitations to perform at the parties of the international ultra-wealthy.

Either way, I’m told that both of them are bringing in absolutely monster incomes, while staying far from the limelight. One of the reasons that they can charge a lot of money for these gigs is the illusion of exclusivity — the idea that a Lionel Richie gig or a Kenny Loggins gig is a rare and special thing, from someone who doesn’t go touring any more. Even if they really do still tour — just not in public.

I’d also love to know, in the context of the Milken Institute in particular, whether Richie got paid for this particular gig, and if so, how much. The Institute is, after all, a registered non-profit organization, which solicits donations from members of the public to “improve the lives and economic conditions of people in the U.S. and around the world”. Does it count if they’re improving the economic condition of Lionel Richie’s bank account?

Update: Milken’s Conrad Kiechel clears this up:

The Milken Institute did not pay for Lionel Richie to appear at our Global Conference, nor did we pay speaking fees for any of the 560 people who served as panelists or moderators during its three days.  That included National Institutes of Health Director Francis Collins, who joined Lionel Richie on stage Monday night for a knock-out rendition of “Sweet Baby James.”


Lionel Richie just hit #1 on Billboard with an album that recycles his hits with country artists dueting, so the 1% are in tune with the 99% on that score.

Posted by T-man | Report as abusive

James Murphy’s role in the LCD Soundsystem ticket fiasco

Felix Salmon
Mar 24, 2011 21:14 UTC

James Murphy, of LCD Soundsystem, is not on Twitter a lot. In the past month, he’s tweeted precisely eight times. But when he was trying to sell tickets to his final show at Madison Square Garden back in February, he was very active. He started on Tuesday February 8, with two tweets to announcements of a ticket presale on February 9. And then after the presale released tickets onto the market, he started getting angry, with a series of eleven tweets expressing violent and profane anger towards scalpers in general and StubHub in particular. It seems his ire was raised by someone selling a single ticket for $1,500.

But there’s something very interesting going on here. I talked to Glenn Lehrmann of StubHub today — himself the subject of an irate Murphy tweet — and he said that when Murphy started sending his tweets out, there were roughly 1,000 tickets for the LCD Soundsystem show available on StubHub. Most of them were priced at about $130 to $140, with about 90% under $200. The tweets, however, “significantly raised demand” and the perceived value of the tickets. By the time that tickets officially went on sale to the public on the morning of Friday February 11, fewer than 30 tickets had asking prices of less than $200, and the average price was around $500.

When the tickets went on sale, no one got any. And so the demand moved naturally to StubHub — of the 1,915 tickets to LCD Soundsystem’s MSG show that StubHub has sold to date, roughly one third were sold on February 11, when prices were at their peak. Right now, prices are much lower; the average is $212, and the lowest-priced tickets are about $100.

Lehrmann confirmed to me that StubHub saw no increase in the number of tickets available for sale after 11am on Friday. The official James Murphy theory — that scalpers with bots had bought up all the tickets and were flipping them with StubHub — is simply not true: substantially all of the tickets which sold on StubHub that day came from the American Express pre-sale on the 9th.

“It’s not humanly possible to sell 9,000 tickets in one minute,” Lehrmann told me, adding that if MSG or Bowery Presents (the promoter) or Murphy himself simply published the manifest for the show, that would clear everything up, by showing to the public just how many tickets were sold on February 11 when the bulk of the tickets ostensibly hit the market. “The artists and promoter aren’t going to share the ticketing manifest, so they hide behind the bots theory,” says Lehrmann. “But if the bot theory was true, wouldn’t you be waving the manifest from the tallest mountain?”

The fact is that the number of LCD Soundsystem tickets sold on StubHub is entirely normal for the venue — the Lady Gaga show in February, for instance, saw more than twice as much activity on the site.

So what’s going on here? “I’m not revealing any huge industry secret,” says Lehrmann, “when I say that the majority of tickets are held back, and are sold either to local brokers or directly resold on a secondary site.”

Essentially, what happens is that bands set the face value of the tickets artificially low, so as not to look as though they’re ripping off their fans. But they only release a fraction of tickets to the public at face value. Lehrmann told me that a Taylor Swift show at National Arena last year sold just 13% of its tickets to the general public, with another 30% going to American Express and to the fan club. Fully 57% of the tickets were sold through some kind of back channel, presumably at a substantial mark-up from face value. In the case of MSG, it’s clear that’s going on: “at $40 face value,” says Lehrmann, “the promoter probably isn’t even paying the rent on the building.”

Between them, the band and their promoter build up long-standing relationships with ticket brokers, who then sell on their wares in a variety of ways. Some appear on StubHub and other secondary-market sites; others are sold directly to clients; others still are hawked on the street on the evening of the show. The risk is borne entirely by the brokers: the promoter has sold its inventory to them, and then leaves it up to the brokers to determine how, where, and when those tickets might appear for sale.

In the case of LCD Soundsystem, it looks very much as though the overwhelming majority of tickets went to brokers, and few if any were sold at face value on the public on-sale date. Murphy can rage against the scalpers as much and as loudly as he likes. But looking at the numbers from StubHub, it seems that Murphy himself — and/or his promoters at Bowery Presents — are exactly the people putting those tickets into the scalpers’ hands. If Murphy wants to go around blaming people, he should first come clean on how much his own behavior caused the very problem he’s complaining about.


Your math is wrong – 1,915 is about 9.5% of the capacity.

And to echo what was said above, using Lehrmann as your only source, not talking to Murphy and apparently not contacting Bowery Presents either: all of this looks like sloppy journalism.

Posted by MarcBrubaker | Report as abusive

The dynamic economics of LCD Soundsystem tickets

Felix Salmon
Feb 19, 2011 21:27 UTC

A clear narrative emerged pretty quickly in the wake of last week’s LCD Soundsystem ticket fiasco. Annie Lowrey tried and failed to get tickets when they went on sale at 11am on Friday, but was foiled:

Had something gone awry? I quickly checked Twitter. Nobody—really nobody, it seemed—had gotten through. Perhaps there was a problem with the site?

No. As it turned out, the show had sold out within seconds. It is just that professional ticket resellers, otherwise known as scalpers, had scooped up the bulk of the seats. Within minutes, hundreds of them were available on StubHub and other secondary markets where sellers can charge whatever they want. Tickets with a face value of $49.50 were going for 12 times that—with some coveted spots in the general-admission dance area going for thousands of dollars.
How did they do it? With bots. Computer systems—not particularly sophisticated ones, either—submit tens of thousands of requests for tickets the very instant they go on sale, crowding regular folks out.

This story seemed to be confirmed by LCD Soundsystem itself, with a profanity-laden posting blaming scalpers for the problem and presenting new shows at Terminal 5 as the solution. As Lowrey puts it, frontman James Murphy “realized he had an ace up his sleeve. He flooded the market, adding shows, upping ticket supply, and hopefully pushing prices down.”

For anybody who loves both music and teachable moments in microeconomics, the subject was irresistible. Lowrey’s post was followed up by Matt Yglesias, who drily declaimed that “optimal allocation of LCD Soundsystem tickets requires demand-responsive ticket pricing” if scalpers aren’t going to end up collecting rents. And Rob Cox, after looking into the matter, concluded similarly that what we’re seeing here “offers a strange insight into the laws of supply and demand”.

But in fact the story of these shows is much murkier than all this pop-economics punditry would have you think. Bob Lefsetz, who has real-world experience of how tickets are sold in practice, says that far from selling out 13,000 tickets at the public on-sale date, LCD Soundsystem in fact only sold 1,000. He notes:

James Murphy could publish exactly how many tickets go on sale to the general public, but he doesn’t want to. No act wants to, they’re afraid of the public outcry. This information is available to acts, but they don’t want to disseminate it.

After publishing his analysis, Lefsetz then mailed out a letter he received which lays out an intriguing counternarrative. What if the MSG show has not, in reality, sold out at all? The conspiracy theory goes like this: LCD Soundsystem’s promoter, Bowery Presents, owns Terminal 5. By holding back most of the MSG tickets, secondary-market prices would be sure to skyrocket. The way that MSG is structured, the coveted general-admission area in front of the stage is actually pretty small, which means that it’s quite easy to generate a handful of headline-grabbing offers of tickets for sale at $10,000 apiece or more. If they wanted, LCD’s promoters could even put those offers up themselves, and then encourage the band to complain in public about the exorbitant prices.

After getting everybody’s attention by artificially clamping down on the supply of MSG tickets, LCD’s promoters can then easily sell out four or more shows at their own venue, Terminal 5, which by coincidence just happened to be unbooked in the run-up to the MSG gig. Given all the buzz that this activity creates, the unsold MSG tickets can then be quietly disposed of on StubHub and other secondary-market sites.

I suspect that there’s more than a little truth in the conspiracy theory. For one thing, the number of tickets available on StubHub did not actually increase appreciably after 13,000 tickets were purportedly sold out in seconds. On top of that, we’re in mid-February already; it’s definitely weird that Terminal 5 was set to be completely dark from March 20 through March 31, with the exception of a single show on March 25. And it’s even weirder that no one — no one at all — got public tickets for the MSG show when they supposedly went on sale en masse: the only people who have gotten tickets in the primary market did so on the pre-sale dates or through tickets allocated to American Express.

The fact is that concert promoters, like art dealers, are fiercely protective of the asymmetric information advantage they have over the general public. Bowery Presents, the promoter of these shows, knows full well how many tickets were sold to the MSG show, and when. But they’re not releasing that information, because it’s very much in their interest for everybody to believe that 13,000 tickets sold out in a matter of seconds.

I don’t think that’s possible. Bots are sophisticated, to be sure, and anybody familiar with high-frequency trading on stock exchanges knows how quickly financial transactions can take place electronically. But Ticketmaster is not set up as a high-frequency exchange, and indeed puts up obstacles designed to make it harder for bots to buy lots of tickets quickly.

On top of that, bot-wielding scalpers had no particular reason to believe that LCD tickets would become hugely valuable on the secondary market, given that the band had never played a show of remotely MSG’s size in the past. I can see them buying a few hundred tickets over the course of 15 minutes or so; I simply don’t believe that they bought more than 10,000 tickets in the space of less than 15 seconds. I don’t believe they wanted to, and I don’t believe they’re capable of doing that even if they did want to.

People sympathetic to the band, like Rob Cox, claim that LCD Soundsystem and its promoters didn’t understand the economics of scarcity when they put the MSG tickets on sale. I, by contrast, think they understood the economics of scarcity all too well — and successfully used it to generate buzz and publicity. What really happened here, I think, is akin to the IPO of theglobe.com back in 1998, where the supply of new shares was so tiny that the price soared from $9 to $97 on the first day of trading. In turn, that generated lots of headlines, and ensured that the number of people who had heard of the website increased by orders of magnitude.

Supply and demand for concert tickets aren’t static numbers which then get reflected in prices. There are complex feedback loops here too: scarcity and price mechanisms can feed back into increased demand for tickets. Certainly this story has meant a large increase in the number of people who know that LCD Soundsystem is playing its last-ever gig at MSG in April. It’s surely naive to think that all the second-order effects here were completely unintended.


wow. tons and tons of conjecture, in the article and comments alike. basically all hot air here. why, people, why?

Posted by werdyo | Report as abusive

Deceased icon datapoint of the day

Felix Salmon
Jun 29, 2009 11:52 UTC

From the WSJ:

After the deaths of other major stars, including Elvis Presley, in 1977, and Kurt Cobain in 1994, around half of outstanding concert tickets were never returned for refund, according to people in the concert business, because fans preferred to keep them as souvenirs.

Is it possible that AEG will end up making more money now the shows have been cancelled than it would have had they gone ahead?


I would guess that most of the tickets to the upcoming shows havent been printed/mailed yet. Also, many tickets these days are electronic or will call. Thus, while the souvenir factor will undoubtedly influence some physical ticket holders I would think that this is not the majority of the people who actually paid for seats.

Posted by JM | Report as abusive