Uncharted Territory

May 25, 2016

Marketing Live Chess Broadcast Rights

Background: Agog at Agon’s Candidates Broadcast Monopoly

The moves of chess games have always been in the public domain.  Anyone can quote them in any medium including whilst games are in progress.  Indeed, in recent years live internet broadcasts of commentary on games at chess tournaments and in matches have become very popular with the chess community.

It was in this context that several chess websites geared up to broadcast commentaries on the most significant chess tournament of 2016, the Candidates tournament in Moscow which opened on March 10th.  The winner of the Candidates gets to play the current World Champion in a match for the title of World Chess Champion, so we’re all looking forward to watching Sergey Karjakin challenge Magnus Carlsen in November. But two days before the Candidates started, the organisers, a company called Agon, working with the worldwide chess federation, FIDE, forbade anyone else from broadcasting the moves until 2 hours after each game.  They did this by requiring anyone accessing the official website to sign a “Clickwrap Agreement” agreeing not to retransmit the moves.  Presumably onsite spectators and journalists were subject to similar restrictions.

Malcolm Pein, the editor of Chess magazine, noted in the May 2016 issue (p.4-5) that Agon’s attempt “to impose a new order on the chess world” were “cack-handed” and it is indeed very unfortunate that FIDE has been involved in preventing a number of websites from supporting what I would have thought is its core objective, promoting the game.  The sites are likely to have incurred costs as a result, perhaps having committed to pay commentary teams.

Furthermore, as Pein notes in his Chess editorial, aspects of the Agon Candidates commentary left something to be desired.  He highlights an unfortunate incident when the moves were inadvertently shown swapped between games at the start of the last round.  I noticed that too, and was also confused for a moment by the disconnect between the commentary and what actually appeared to be happening, but much more serious was the quality of the commentary itself.  I felt it was interrupted much too often for breaks, usually to show the same couple of ads, plus what I’ll describe as “pen-portraits” of the players (cartoon-style drawing accompanied by commentary).  These were quite entertaining the first time you saw them.  Not quite so much the tenth time.  And, although the commentary team obviously worked hard to help the audience understand what was going on, I’ve enjoyed other commentators somewhat more.  The commentary is much more important in chess than say football, since (as non-players will appreciate!) there are periods of a game when there’s nothing much to see happening on the board.  I would have liked the choice to watch another broadcast.

Steve Giddings writes, also in the April edition of Chess (p.8-9), that preventing unauthorised broadcast of chess games is in “the commercial interests of the game”.  That may be so, but it seems to me that monopoly broadcasting is not the best way forward.

Given the goals of promoting chess by maximising the number of viewers of chess matches and tournaments and maximising revenue from the broadcast of elite events, simply in order to pay for them, a better option would be to license multiple broadcasters, if they’ll pay collectively more than would a single exclusive media outlet.  I outline in this article how the revenue-maximising number of broadcasters could be established by a simple process of bidding for a share of the rights.

First, though, let’s consider how other sports rights are sold and then whether times have changed – perhaps other sports might want to reconsider granting exclusivity – and how chess is different. I focus particularly on the case of domestic rights to broadcast the English Premier League.

The Football Precedent: English Premier League Live Broadcast Rights

When I was a kid, the FA Cup Final was always shown live simultaneously on both BBC1 and ITV.  So much for consumer choice – at the time there were only 3 channels (the other one being BBC2).  Nevertheless, there was competition, of a sort.  Sometimes we’d switch over to see what they were saying on the other side, though when the ads came on we’d switch back.

One might wonder why ITV would bother broadcasting the Cup Final when it was also on BBC (without ad breaks) and, indeed, why the FA would sell it to two broadcasters rather than just one.  I can think of two considerations:

  1. There was some product differentiation between the broadcasts on BBC1 and ITV.  The channels employed different commentators and pundits.  This produces what I would argue is healthy competition for viewers between broadcasters.
  2. Strange though it may seem to many younger readers, back in the day many – perhaps most – households watched either ITV or BBC almost exclusively, even though they both were (and still are) free-to-air.  It could be argued that the choice between watching BBC and ITV used to be very much driven by social class or at least the social class households identified themselves as belonging to, but that is actually irrelevant to the argument.  The point is that broadcasting the FA Cup Final on both ITV and BBC ensured that the product reached more people – ITV viewers and BBC viewers – than it would have done had it gone out only on BBC or ITV.

Presumably ITV could attract enough viewers and sell enough advertising to make it worthwhile to broadcast the FA Cup Final even though BBC1 was showing it too.

Sports Broadcast Monopolies

Why, then, you might ask, is almost all football shown in the UK now, in 2016, indeed, almost all sport (and much other content besides), broadcast on just one channel?  That is:

  • why have sports broadcast monopolies developed?;
  • why do sports administrators tolerate and even encourage broadcast monopolies?;
  • and whose interests do sports broadcast monopolies actually serve?

Some years ago I had the dubious pleasure of a job interview with BT; actually they wasted an entire day of my time at their recruitment centre (and even more with some further interviews later on).  The question arose in discussion – I guess after we’d noted the ongoing convergence of internet and broadcast media – as to how BT could best grow their broadband market.  I suggested offering some exclusive movies.  Perhaps my interlocutor was playing Devil’s advocate, but I don’t think so; regardless, he seemed to be arguing that they should market on the basis of their whizzy new network.  No, no, no!  The vast majority of consumers care only about what appears on their TV; they don’t care at all about the underlying technology.  And if there is some exclusive content – I mentioned movies at my BT interview because Sky had already “done” sport – that is likely to be decisive in winning customers.

It seems clear from their enthusiasm to enter into them that exclusive deals for live sport transmission rights are in the interests of subscription broadcasters, particularly when  trying to build a customer base.  We have the example of the English Premier League (EPL) and much other sport (as well as films and other content) on Sky, now being contested by BT.  Netflix and Amazon are exclusively hosting supposedly must-see drama series.

As a consumer, I’m always wary when I’m told something is “exclusive”.  The very word suggests to me that someone is being ripped off.  Probably muggins.

But let’s not jump to conclusions.  Besides, what we’re really interested in is the health of the sport – that is, chess, when I get to the end of this preamble.

So, could exclusive sports rights sales be in the interests of the sports themselves?

Well, when broadcasters are trying to grow their business – think of Sky and the EPL – they may be prepared to pay what appears to be a substantial premium for a monopoly.  I say “appears to be a substantial premium” because at some point the broadcaster has to demonstrate income (advertising and/or subscriptions) commensurate with the expense.  Otherwise they go bust.

It’s not immediately apparent, and, indeed, somewhat counter-intuitive, that a single broadcaster of live events or TV series can unlock more advertising and/or subscription income than can multiple broadcasters of the same material.  Nevertheless, many sports administrators appear to believe monopoly broadcast deals are in the interest of their sport.   At least in the short term.

An example of what can happen in the longer term is provided by EPL broadcast rights in the UK.  Sky held the exclusive rights from the start of the Premier League in 1992 until 2007.  After the European Commission ruled that Sky should not have exclusive rights to all matches, they had competition, first from Setanta, who ran into financial difficulties, then ESPN, who took over Setanta’s rights and most recently BT who came into the market in 2012 prepared to bid aggressively against Sky for a whole range of football and other sports rights and apparently with equally deep pockets.  Guess what happened once there was competition?  The total paid for EPL live transmission rights went up.  Considerably.

Note, though, that what Sky and BT bid for is how much of the monopoly each enjoys.  They are not in direct competition, in the sense of broadcasting the same matches, as BBC1 and ITV used to be in the case of the FA Cup Final.

The only logical conclusion is that – given that live broadcast rights to the EPL have a definite value represented by the income they can generate – they were previously being sold too cheaply!  Who’d have thunk it?

Players on £50K a week 5 years ago should be a bit miffed.  They could have been on £60K!

Why are BT and Sky paying more than Sky alone did?

Is it a conspiracy against the consumer, as I once read a commentator claiming?  Apparently, he wrote (I think it was a “he”) EPL fans would now have to buy two subscriptions.  As someone who only buys one, it might be worth pointing out that, unless it’s your team playing, or a key fixture (in which case there’s always the option of going to the pub – a form of pay-per-view) it doesn’t make that much difference which match you watch.  You don’t know in advance whether a particular match is going to be exciting.  In other words, if you’re only going to watch 20 matches a season, there’s not much point paying for 200.

Are BT and Sky trying to buy or defend market share and therefore overpaying?  Well, there may be an element of this, but, first, from the point of view of the sport this is a good thing.  Second, companies can’t do this for ever.  BT is now established in the market.  I doubt they’d be paying so much for 3 years of broadcast rights if they didn’t think they’d make money on the deal.

Has BT unlocked market segments Sky wasn’t reaching?  Yes, I believe so.  I pay a small add-on to my broadband internet deal to receive BT’s sports channels, which I watch online, on a PC.  For the number of matches I actually manage to watch I can justify this cost, but not a Sky subscription (plus charges for set-top boxes and so on).

But it may also have been that Sky was paying less than the EPL transmission rights were worth and making excess profits as a result.  These have not necessarily all appeared as profits in its accounts, but may have also been reinvested, for example, in establishing a dominant position in the UK in the broadcast markets for other sports, such as cricket.

A market needs to be competitive to establish the real value of a product.  It’s in the long-term interests of sports themselves, I suggest, to maintain a competitive market for broadcast rights and not allow monopolies to develop.  Such monopolies might end up underpaying until a competitor eventually challenges them, as, I argue, appears to have happened for EPL live broadcast rights in the UK.

In addition,  it’s in the long-term interests of sports for as many people as possible to be able to watch them.  This is best achieved by a number of broadcasters with different business models reaching different segments of the market.  It’s worth pointing out that sports administrators sometimes ensure that at least some events are “free-to-air” in order to show-case their product, for example the World Cup and, this summer, the UEFA Euro 2016 tournament (at least in most European countries).  This month’s FA Cup Final was broadcast on the BBC as well as BT Sport.

Differences Between Chess and Football

After that somewhat longer discussion of football than I had intended, let’s get back to chess.  As I’ve argued, even football could consider selling live transmission rights to multiple broadcasters, but are there differences between chess and football) that make monopoly broadcasting a less attractive option in the case of chess?

I believe there are several relevant (though interrelated) differences:

First, live chess is typically broadcast globally, over the internet.  This means that the peculiarities of local markets are much less relevant.  For example, in the UK the playing field for broadcasting football was uneven when Sky entered the market.  Sky had to have content that was not available to the free-to-air channels ITV and the BBC or no-one would have subscribed; and it needed subscriptions to fund the cost of its satellites.  OK, there is at least one place where chess appears live on TV: Norway, home of the World Champion, Magnus Carlsen.  But given the general reliance on the internet for broadcasting chess, it makes sense to simply leave distribution to the broadcaster and not sell rights separately for different platforms (TV, internet, mobile devices etc).

Second, and related to the first point, the world has moved on in the quarter-century since the EPL broadcast model was established.  To some extent sports channel subscription revenues funded a dramatic increase in the number of channels available by enabling satellite and cable TV.  But with the growth of TV over the internet, the potential number of channels is vast, and the entry-cost considerably lower than in the past.  Broadcasters don’t need huge guaranteed revenues to justify their business models.  Furthermore, given the flexibility of advertising charging that is possible on the internet – essentially payment depends on the actual number of viewers – advertisers do not need historic broadcast data.  They’ll just pay for what they actually get.

Third, the commentary and presentation is a more significant part of the overall package in the case of chess than it is for football.  Personally, for normal tournament commentaries I’m as much interested in who’s commentating than who’s playing.  I’d be much more likely to tune in if Maurice Ashley, Danny King or Peter Svidler are explaining a game.

Fourth, chess is still at the experimental stage, still trying to explore what works best in live transmission.  It doesn’t make a lot of sense to stifle this process by restricting the number of broadcasters to one.

Fifth, interest in chess is global.  Viewers might appreciate broadcasts in their own language as well as English.

Sixth, there are a limited number of marketable chess events.  To promote the game, as well as maximise revenues, it makes sense for these to be available to as many viewers as possible.

Seventh, I don’t believe there is a pot of gold waiting for someone able to sell advertising round chess events.  Compared to football, it’s always going to be a niche market.  Indeed, for many of the chess sites – Chess.com. Chess24.com, the Internet Chess Club (ICC), Playchess.com and so on – that broadcast (or might broadcast) elite chess events, covering live events is, unlike in the case of football, only part of their offering to visitors to their site (who may pay a subscription).  These sites also allow you to play online, host articles and instructional videos and so on.  Unlike the sports channels of Sky or BT, losing live transmission rights is not an existential threat.  They are therefore unlikely to pay huge sums for monopoly rights.  Collectively, though, they may pay a decent amount for something that is “nice to have”.  The resulting choice for viewers would also be beneficial to the game and raise broadcast standards.

For all these reasons it seems to me that it makes sense for chess events to be hosted by multiple broadcasters.

Price Discovery for Chess Broadcast Rights

Before considering the mechanics of an auction for chess broadcast rights, let’s first establish a principle: all broadcasters will pay the same price.

Live sports transmission rights are generally sold territorially.  That’s messy already – people cheat by importing satellite dishes from neighbouring countries and so on- but in the age of internet broadcasting its unworkable.

One might also consider language restrictions.  Why should a broadcaster be able to reach the whole Chinese population or the English, Russian, French or Spanish-speaking world for the same price as an Estonian native-language broadcaster?  Well, don’t worry about it.  The market will take care of things.  Broadcast auctions will be a repeat exercise and, if the price is low compared to the size of the market in a specific language, that will simply encourage more broadcasters.

What if some broadcasters are mainstream TV channels, in Norway, for example?  Again, don’t worry about it.  Just leave distribution up to the broadcasters.  TV channels are competing with internet broadcasters.  The only restriction should be that a one licence – one broadcast rule.  If a broadcaster wants to transmit to multiple audiences, in different languages, say, or by producing different versions tailored to experts and the general public, then they have to buy two or more licences.

What would the broadcasters buy?  An automatic feed of the moves (top events nowadays use boards that automatically transmit the moves electronically) is obviously essential.  Since you don’t want numerous video cameras in the playing hall, the organisers (or a host broadcaster) would also provide video feeds of the players, often used as background to the commentary (generally in a separate window).  Post-game interviews or a press conference are also usual and these could be part of the package, as could clips from the recent innovation of a “confession-box”, where players can comment during their game.  Broadcasters would edit these video feeds together with their own commentary to produce their final product.

Let’s make one other thing clear about the objective of the auction process.  The goal is to maximise revenue.  This is not in conflict with the goal of maximising the online audience and thereby promoting the game.

So, how would the auction work?  How can we maximise revenue from an unknown number of broadcasters all paying the same price per transmission stream?

Here’s my suggestion.  The broadcasters would be required to submit a number of bids each dependent on the total number of broadcasters.   That is, they would bid a certain amount to be the monopoly broadcaster, another amount (lower, assuming they act rationally!) to be one of two broadcasters, another amount to be one of three, and so on, up to some arbitrary number, for example to be one of more than ten broadcasters.

The chess rights holder – FIDE, for example – would simply select the option that generates most revenue.  All bidders would of course pay what the lowest bidder offered to be one of the specific number of bidders chosen.  E.g., if 2 bidders are successful, one bidding $70,000 to be one of two broadcasters and the other $60,000, both would pay $60,000.  In this case, neither broadcaster, nor any other, would have bid more than $120,000 for exclusive rights and no 3 more than $40,000 to be one of 3 broadcasters, nor 4 more than $30,000 to be one of 4, and so on.

For example, it may be the case that one bidder bids more to be the sole broadcaster than any two bid to be dual broadcasters, any three to be the only three broadcasters and so on.  In that case, one broadcaster would secure a monopoly.  Or, at the other extreme, 12 broadcasters might, for example, bid more to be one of “more than ten” broadcasters than any sole broadcaster bid for a monopoly and so on, and more than 13/12 times what the unlucky 13th highest bidding broadcaster bid to be one of “more than ten” broadcasters, 14/12 times what the 14th highest bidder bid, 20/12 times what the 20th bid and so on up to the total number of bidders.

It’s my guess that revenue will be maximised for a World Championship match by a relatively large number of bidders.  And the crucial point is that the more broadcasters, the larger the audience and the greater the choice for viewers.



September 4, 2012

Skype Video Call Bandwidth Usage

Filed under: Internet, Skype, Technology, Telecomms — Tim Joslin @ 3:57 pm

For once my title avoids any of my usual no doubt misguided attempts at wit or humorous alliteration. For this is a serious business. I mentioned a couple of days ago my confusion at having to deal with one particular supplier, Thames Water, as a result of having to move home. My preferred mode of behaviour is to set up Direct Debits with these guys and forget about them, apart from filing away bills in case there’s some kind of a problem – which is why I prefer paper bills sent to me in the post rather than online ones to which the supplier controls access. But let’s not digress too much as that would defeat the object of the title chosen for this post.

My Thames Water induced confusion has been as nothing compared to the ordeal of having to switch internet provider. I’ll save the long story for another post, but against my better judgement – nay, naively, or perhaps through mis-selling – I’ve chosen a usage-limited package, the heavily advertised BT Total Broadband Option 1. I thought 10GB per month would be plenty since the household very rarely downloads anything at all, and downloads were all BT happened to mention:

And 10GB/month would be plenty if it weren’t for that pesky Scooby-Doo, sorry, Skype.

For it rather seems Skype video calling uses as much bandwidth as it can get its hands on, at least up to around 1GB/hour – which is quite a bit when you’ve only got 10GB to use a month.

Now, I’m not such an idiot that I didn’t think of this before signing on the dotted line with BT. I assumed Skype would use about a 10th of what it is using. My logic was essentially that Skype is nowhere near movie quality and BT suggest on their website that you could download 15 movies a month with your 10GB allowance (seems a bit optimistic – more on that another time) and random sampling of BBC iPlayer content suggests a 60 minute programme takes up 640GB.

And indeed when I tested Skype video calling over a relatively slow work WiFi connection to the BT ethernet-connected desktop PC at home it did seem to use only 30kBps bandwidth (big B for byte, not little b for bit!) in total for both upload and download, which does indeed equate to around 100MB/hour or 0.1GB/hour.

But then I tested Skype at home from the desktop PC to a laptop operating over WiFi. And – besides shattering all the windows with feedback (best turn the radio off before you try this yourself!) – I did indeed manage to consume bandwidth at a prodigious rate, equating to the 1GB/hour I’ve already mentioned. Though the beauty of this experiment was that the usage seemingly didn’t count towards my 10GB monthly allowance – Skype is peer-to-peer, so I guess the video calls never left the flat!

How did I determine my bandwidth usage?

I’m presuming the answer to this question will be the most useful part of this blog entry – at least for those, like me, using Windows 7 Home Premium (and possibly those with other recent Windows releases) – so I’ll go into some detail and provide some pretty pictures.

In fact there are at least 3 simple ways of determining the same information (besides BT’s oh so useful daily updated Broadband Usage “Monitor”). I used them all and the results were consistent.

1. Skype’s Call Technical Information

Skype maintain a set of answers to FAQs, such as FA10415, Can I see how much bandwidth a video call is using? which I first came across in the answer to a query about a household exceeding a 250GB a month usage limit(!). The procedure FA10415 is slightly confusing in two respects.

First, FA10415 starts off by detailing how to switch on the feature to display technical information during calls, before noting that this has been unnecessary since Skype 5.2. As I’m on it would seem there have been a lot of updates since version 5.2. So, unless you have an old version of Skype, the first part of FA10415 is redundant. You can check your Skype version by using the drop-down menu to select Help, About Skype. You can keep up to date via Tools, Options, Advanced, Automatic Updates, presuming you have Administrator access.

Second, the menu item Call Technical Info only exists during calls! – it would be more consistent for it to be greyed out when not available.

So, to determine their video or voice-only call bandwidth usage most Skype users will simply be able to call up the Call drop-down menu:

Selecting Call Technical Info should result in something like this (personal information redacted for reasons of paranoia):

The figures we’re interested in are at the bottom of the Basic section, i.e. in this case upload at 56kBps and download 146kBps (B for byte in this case, b for bit being more usual). This represents a lot of bandwidth, and in fact both PCs I used settled down to just under 150kBps for both upload and download after I took the above snapshot.

A total of 300kBps represents bandwidth usage of 300*3600kBph (hour), i.e. 3*360MBph – call it a round 1GBph!

2. Connection Status information

You can verify what Skype is telling you, and determine bandwidth usage over a period of time by viewing your network Connection Status Information. If you’re using an ethernet cable to access the internet you need to right-click on the network icon (two terminals at the right side of your taskbar – sorry no pic handy) and then select Open Network and Sharing Centre. You can also call up the Network and Sharing Centre from the Control Panel. From the Network and Sharing Centre you just need to click on View Status for your network to call up the required data. You might then see something like this:

If you’re using WiFi you can access the Network Connection Status directly (i.e. without visiting the Network and Sharing Centre) by clicking on the network icon (the 5 bar signal strength indicator at the right side of the task bar) then right-clicking the relevant network (the one that’s Connected!) and selecting Status.

3. Task Manager information

Finally, if you want a snapshot of second-by-second network bandwidth usage, you can use the Task Manager. Holding CTRL and ALT while pressing DELETE calls up a menu. Selecting the last entry, Start Task Manager fires up a window allowing you to view what’s going on in your system in real time. Tab Networking is the one we want. Calling it up then initiating a Skype call might result in something like this:

The grid of subdivisions of axis units are not the ones I would have chosen – each box seems to represent 6 seconds (10 per minute) and dividing 5% into three, with one label is odd, to say the least – but it’s clear enough what’s going on.

In this example you can see that Skype (nothing else of significance was using the network at the time) managed to use around 2.5 to 3% of the 54Mbps bandwidth available. Note that we’re now working in b for bits, rather than B for bytes. Irritating, isn’t it? Anyway, 2.75% of 54Mbps is 1.5Mbps, WLEE (within the limits of experimental error) as they told me to write in O-level science. Dividing by 8 to get bytes from bits results in 200kBps, WLEE. This is not so different to the 56kBps upload + 146kBps download speeds in the snapshot of the same call I took a few minutes later using Skype’s Call Technical Info (see section 1, above).


It seems all three methods of determining real-time bandwidth usage by Skype on a Windows 7 machine yield consistent results, compatible with the getting on for 1GB per hour of bandwidth usage during Skype video calls deduced from BT’s daily bandwidth usage summaries. Task Manager provides only one figure for upload and download (though maybe more information could be gleaned were one to break the hacker’s code and actually RTFM, read the manual), but otherwise the choice of which method to use depends merely on convenience, or whether one prefers to deal in bits or bytes!

Issues arising may (meaning I don’t always keep my promises!) be dealt with in subsequent posts. These are unlikely to include how to reduce Skype’s network bandwidth usage, since investigation suggests that such a capability does not exist for users (perhaps they’d consider introducing it) – though please feel free to make a comment disabusing me of this notion. I do have plenty to say, though, on BT’s marketing of their 10GB BT Total Broadband Option 1, so watch this space!

Create a free website or blog at WordPress.com.