73% of journalists [*] sometimes or always use blogs in their research.

No Gravatar[*] = newspaper, magazine, TV, radio, and web journalists

Editors &amp- Publishers

Via Henry Blodget (who is hilarious, as always)

Implications for the field of prediction markets (InTrade-TradeFair, BetFair-TradeFair, Betdaq, HSX, NewsFutures, Inkling Markets, etc.):

  1. The P.R. arm of the prediction market firms should also reach the bloggers &#8212-not just the journalists.
  2. Prediction market firms should monitor the Blogosphere for rumors, and deal with them in a subtitle way. :-D
  3. A long (&#8220-too long&#8221-, some will say), balanced, detailed blog post about your product is more useful to reach out than a media kit. :-D
  4. Our industry needs a blog network of reference, all focused on prediction markets. :-D

Previous blog posts by Chris F. Masse:

  • Last year’s best April Fool’s Day Joke had something to do with the Wisdom Of Crowds.
  • Will HedgeStreet USA, the hypothetical InTrade USA, and the hypothetical TradeFair USA, be regulated in the future by a merged SEC+CFTC regulatory structure?
  • WORST THAN ELIOT SPITZER (if it were possible): Formula One boss, Max Mosley, had sado-masochist sex with 5 prostitutes, for 5 hours (!!), reenacting a concentration camp scene (!!) in which he played the role of both Nazi guard and inmate.
  • Is BetFair Poker a booby trap for the gullible novices? Does The Sporting Exchange (the operator of the BetFair brands) help gangs plucking down innocent recreational poker players?? To get an inkling, don’t read The Guardian, seeded by the BetFair spin doctor- read Midas Oracle.
  • The video that the technologically retarded BetFair spin doctor should watch.

Googles Bo Cowgill takes a swipe at the prediction market software vendors.

No Gravatar

Using Prediction Markets to Track Information Flows: Evidence from Google – (PDF file – PDF file) – by Bo Cowgill, Justin Wolfers, and Eric Zitwewitz – 2008-01-06

Bo Cowgill:

[&#8230-] Trade-by-trade data can reveal characteristics of specific working groups: What they know, how they feel, how they process and share information and how all of that changes over time. I didn&#8217-t try to put any of this in the paper because the conclusions would be sensitive, and I thought this application was pretty obvious to anybody who understood our methodology. [&#8230-]

Bo Cowgill:

I&#8217-ve also heard that other companies would find it impossible to analyze the interaction between their market and the organization. Why? Lack of data. [&#8230-]

Bo Cowgill:

Some more remarks about applications that combine prediction markets and organizational data (org charts, social networks, seating locations). The obstacle to these applications is not a lack of data. Jed mentions privacy concerns &#8212- and if he thinks this is a big obstacle then I&#8217-d be interested in discussing his thoughts.

A bigger problem is that that current prediction market vendors and consultants cannot support these applications. At heart, these vendors are software engineers and salespeople at heart, not statisticians or data miners. They want to write one system that can support lots of clients. At conferences, one hears PM vendors complain about having to do &#8220-customization&#8221- work for clients.

This approach would not work for the applications I describe for two reasons:

  1. The inputs for different clients won&#8217-t be the same. Each client&#8217-s organizational data will likely take a different structure. This makes it difficult for prediction market vendors to architect a single system that can served many clients (yet another challenge with integrating markets with other corporate IT services).
  2. The outputs for different clients won&#8217-t be the same. The business relevance and statistical power of each analysis will differ with each client&#8217-s data.

Prediction market vendors may also need to familiarize themselves with the statistical learning methods necessary to fully utilize these rich datasets. So what&#8217-s the solution? First, move to a software-and-consulting model. By &#8216-consulting,&#8217- I don&#8217-t mean &#8216-consulting on how to implement the market.&#8217- I&#8217-m talking about helping the client solve its problem using a variety of data, including prediction market data.

Second, the vendors also need to pitch prediction markets as more than a forecasting tool. People in the business world commonly identify as data junkies &#8212- probably more so than they identify with the &#8216-wisdom of crowds&#8217- ethos. It is unclear how much companies really care about accurate forecasting anyway.

On a related note, there is something that only the prediction market software vendors could do, at this time, for those who are in capacity to do so: setting up inter-industry prediction markets &#8212-or at least, handing over (with everybody&#8217-s agreement) anonymized prediction market data on industry topics to anyone else in the industry who is a client of that PM firm. I don&#8217-t know about NewsFutures or Inkling Markets, but if you look at Consensus Point&#8217-s list of clients, you&#8217-ll see that David Perry&#8217-s firm is strong in the (consumer) electronic industry &#8212-Motorola, Qualcomm, Siemens, Nokia. Use your imagination, or ask David Perry directly, for more&#8230- (I can&#8217-t talk- otherwise, next thing, I&#8217-m a dead blogger.)&#8230-

Previous blog posts by Chris F. Masse:

  • Last year’s best April Fool’s Day Joke had something to do with the Wisdom Of Crowds.
  • Will HedgeStreet USA, the hypothetical InTrade USA, and the hypothetical TradeFair USA, be regulated in the future by a merged SEC+CFTC regulatory structure?
  • WORST THAN ELIOT SPITZER (if it were possible): Formula One boss, Max Mosley, had sado-masochist sex with 5 prostitutes, for 5 hours (!!), reenacting a concentration camp scene (!!) in which he played the role of both Nazi guard and inmate.
  • Is BetFair Poker a booby trap for the gullible novices? Does The Sporting Exchange (the operator of the BetFair brands) help gangs plucking down innocent recreational poker players?? To get an inkling, don’t read The Guardian, seeded by the BetFair spin doctor- read Midas Oracle.
  • The video that the technologically retarded BetFair spin doctor should watch.

How Decision Markets Work (with emphasis on InTrade) – by Robin Hanson

No GravatarVIDEO – (it&#8217-s a Guatemala-based website) – 2007-10-24

I have clicked on the second blue link, on that webpage, and Window Media Player opened, and now I am watching The Master Of All Universes lecturing a group of politics students&#8230-

The main difficulty in developing efficient public policies is misinformation and false beliefs. The traditional information institutions are the media academics, and informal communications. These institutions have limitations in providing reliable, efficient, useful and enough information to make decisions. Speculative markets base their profits on the prediction of price patterns and are proving to be an efficient information institution. Betting markets have been better and faster predictions than experts because manipulators develop more accurate information. Therefore, speculative and betting markets have to become the central institution on political institutions in order to have effective public policies.

Click here to read the chapter &#8220-Decision Markets for Policy Advice&#8221-.

Great. But the second part of the video is more painful to follow. The students&#8217- questions are sometimes inaudible.

Previous blog posts by Chris F. Masse:

  • Mat Fogarty’s Xpree needs a mini Chris Hibbert.
  • Did you know that Real Clear Politics (a political news aggregator in bed with InTrade) is now owned by Forbes?
  • Easter Egg made in Mathematica
  • MIDAS ORACLE POWER: People googling about BetFair’s new bet-matching logic are automatically directed to our group blog. BetFair’s SEO can return to the locker room.
  • David Pennock, a respected expert in prediction markets and market design, discusses some aspects of BetFair’s new bet-matching logic.

BetFair’s new bet-matching logic + BetFair Maltas trading on the multiples

No Gravatar

BetFair:

Bet Matching Forum Q&amp-A Session 19/03/08

Betfair Customer Services 17 Mar 11:50

As announced last week we’ll be hosting a Q&amp-A session on the forum this Wednesday evening (19th March) between 6pm and 7pm (UK Time). The purpose of this Q&amp-A session is to answer questions regarding BetFair’s new bet matching logic. To help us get through as many questions as possible you can send them in advance to [email protected]. Unfortunately it is not possible for us to respond to each Email individually but we will attempt to answer all questions raised via the live Q&amp-A session.

We realise that customers would appreciate the chance to have questions answered on other topics too, but we want to focus this initial session on just the new bet matching logic to ensure that we answer as many questions as possible. For those customers who have questions for Betfair that aren’t related to this topic we’ll be reintroducing regular forum Q&amp-A sessions over the coming weeks. We’ll post more information about those sessions nearer the time.

We hope you find this session helpful and informative

Betfair Customer Services 19 Mar 18:00

Welcome to the Betfair livechat.

Answering the questions this evening are Mathias Entenmann (MD of Betfair&#8217-s UK and Ireland business)- Mark Davies (Betfair&#8217-s MD Corporate Affairs) and members of their teams.

We have received a number of questions in advance which we will start to answer now. If you have any questions which you have not already submitted, please email [email protected] and we will attempt to answer between now and 7pm (UK time), when the session ends. Please note that you will not be able to post in the relevant forum section.

Betfair Customer Services 19 Mar 18:02
Chatname – Lee (Genuine Scouser)

Do Betfair employ traders to bet in their markets?

There is a trading team based in Malta which manages the risk around the multiples product. They have software which tells them what the risk is associated with a potential result is and suggests what hedge bets can be placed to mitigate that risk at current exchange prices. They then place the hedge bets to manage the risk. They place the bets using the same software as everyone else using the site and respecting any in-play delays.

The multiples product is run under Betfair&#8217-s Maltese bookmaking license and is regulated by the LGA there. Therefore the team has to be based in Malta. The operation is an arms-length operation &#8211- there is no special access to any functionality or data from the exchange. Betfair Malta is charged commission on winning bets in the same way as any other customer in order to comply with relevant regulation and law.

Betfair Customer Services 19 Mar 18:05
Chatname – NB

– Why did you choose not to announce this significant change to your clients? Was it in the hope that we just wouldn&#8217-t notice?

Betfair frequently makes changes to it’s software and we always have to weigh up the balance between keeping customers informed against inundating them with information. In this particular case we’d agree that we’ve done a poor job of assessing the reaction of some customers and communicating appropriately, for which we can only apologise. We’re committed to doing a better job of communicating with customers in future, and this Q&amp-A is the first step in that process.

– You appear not to be using the new bet matching engine on those events that run under the Australian wallet. Is there a particular reason as to why not?

Events run under the Australian wallet are processed on hardware located at our office in Hobart, Tasmania. If we make a change to the software on the UK exchange it isn’t just a case of choosing to switch Australian wallet markets on or off. We have to install the software on the Australian exchange itself. Keeping both systems in sync imposes an overhead, and it’s an unnecessary overhead if, as in this case, there are further changes imminent. We’ve made the decision that our efforts are better spent getting to the complete solution we want in the UK, with price improvements and those bets we could match across selections displayed, and then to look to implement that for the Australian exchange just the once.

– When the SP product was released you announced it in advance and couldn&#8217-t advertise it enough. Why then did you not announce this fundamental change with the same enthusiasm?

Announcing and promoting Betfair SP has had a significant effect on the amount of new customers we’ve been able to attract to Betfair. Once we have the ability to offer price improvements to bets matched across selections, and we can display all the bets we could match, we’re very confident that the vast majority of customers will recognise that as beneficial. Even then that’s still not going to be something that’s going to make a compelling advertisement.

Betfair Customer Services 19 Mar 18:07
Chatname : CLYDEBANK29

– My understanding of the new cross bet matching logic is that it offers neither best execution or common pricing in most circumstances. Is this correct? (By common pricing I mean that if one of my bets is matched using this method it will be matched at that price by another customer).

That’s not correct I’m afraid. The majority[b1] of the bets that have been matched by the new logic so far have been matched at prices where no price improvement would have been possible. While there’s inevitably a temptation to focus on situations that aren’t typical, most Betfair markets aren’t hugely volatile, for example soccer match odds markets. For those markets where prices are more volatile, for example in-play tennis, as I’m sure you’re aware the vast majority of the betting activity takes place on the favourite, so the proportion of bets matched across selections is relatively small.

– What commitment do you have to introduce best execution and common pricing on this new cross bet matching logic? and if you are committed to providing it why have you introduced this change before it delivers either best execution or common pricing in most circumstances? (By common pricing I mean that if one of my bets is matched using this method it will be matched at that price by another customer).

Our developers are already working on providing price improvements when bets are matched across selections. The only factor limiting when we’ll introduce this is how quickly we can develop and test it.

We introduced the current change even without the ability to offer price improvements because we considered it an improvement over the previous situation. One of the biggest barriers to becoming a regular Betfair bettor for new customers is the concept of an “unmatched” bet, an experience they won’t have had when placing bets with our main competitors. Anything we can do to address that and give them a better chance of getting a bet matched immediately helps the long-term growth of our markets. Clearly for customers who’ve been with Betfair for some time, and for whom unmatched bets and what to do about them are second nature, that isn’t going to be obvious.

– Can you understand why some customers think that because this new matching logic doesn&#8217-t offer best execution or common pricing in most circumstances that they view you effectively as a player in the market skimming overbroke situations and who is beating the in running delay and therefore cheating on your own exchange to achieve this? (By common pricing I mean that if one of my bets is matched using this method it will be matched at that price by another customer).

I’d suggest that focussing on the in-play delay is missing the point, although it’s easy to understand why a customer might mistakenly come to that conclusion

The purpose of the in-play delay is to prevent someone watching an event either live at the event, or using pictures with a shorter delay, from selectively matching orders on one side of the market or the other following a price changing event(like a goal or break of serve). As the process only matches opposing customer bets, and no bet is matched by this process selectively based on anything that’s happened in the event being bet on the in-play delay isn’t applicable.

Backing one selection is (and has always been) equivalent to laying the other selections in the market. In-play bet matching takes place as soon as the in-play delay has expired on a newly submitted bet request. If we have a bet request to back a tennis player, say, then it would clearly be unfair to match a request to lay that player as soon as the in-play delay expires while imposing a 2nd delay on a customer looking to back his opponent.

Betfair Customer Services 19 Mar 18:14
Chatname : askari1

– How much money has the engine / arber made for Betfair so far? If commercial confidentiality prevents you from quoting a figure, could you give some indication in terms of the total turned over on a typical event e.g a televised tennis match and / or the typical post-game commission taken by Betfair?

The biggest market the new code has operated on was the televised tennis match between Andy Murray and Roger Federer a couple of weeks ago. Approximately ?6.85 million was matched in the market, and the amount accrued as a result of our inability to offer a price improvement when bets were matched across the two players was ?882.91 . Obviously in well-traded markets like a big tennis match the amounts won and lost by customers are much less than the headline volume figure too, but as a percentage of what’s won or lost in each market the amount is small.

Betfair Customer Services 19 Mar 18:18
Chatname : askari1

– Please could you give the company&#8217-s working definition of the terms &#8216-best execution&#8217- in cross-matches and &#8216-of holding a position&#8217-? According to these definitions, does &#8216-holding a position&#8217- differ from &#8216-holding a liability&#8217-? Do you apply the same definition to &#8216-best execution&#8217- in the case of cross-matches as you do to matches in the case of single-runner sub-markets?

“Best execution” means never less than the price you requested, with the prospect of a price improvement where we can do so, if we can deliver that.

By “holding a position” we mean taking an outright position against a customer, rather than matching opposing customer bets.

The definition is the same for bets matched across selections. As and when we have the means to provide a price improvement we will.

Betfair Customer Services 19 Mar 18:20
Chatname : astonvillain

Hi,
how can betfair users be assured that in the case of a market which suspends with a delay such as football, that betfair owned bots will not match bets which are out of line at the time of suspension? there is a big trust issue here.

The times at which Betfair will match bets across selections are identical to the times at which regular matching (backs vs. lays) takes place. If the market is “active” (not “suspended”) and there are opposing customer bets that can be matched then we’ll match them. If the market is “suspended” then no matching takes place, either backs vs. lays or across selections.

Betfair Customer Services 19 Mar 18:21
Dear BetFair,

I hope you will answer the following:

Q. Why did you feel it unnecessary to inform your vast customer base of the changes to the bet matching algorithm in advance and take feedback- or accept that a trial period may have been a more appropriate way to introduce the changes?

It’s Mark here. I made that call and the judgment was based on the fact that I saw this as a product enhancement which was a benefit to users, and we do not announce every one of those every time. It was doing what we have always said we do – using our technology to match demand between customers – and it was just doing it more broadly than directly backer to layer. We have frequently made the statement that we are a bookmaker which is using technology to match demand, and to my mind this fell directly into that.

I think that if you consider a situation where a 100% book saw backers all sit and look at each other (for example, in a two-outcome event, I think it is daft that two backers, one of each outcome at 2.0, should not be matched), that is easy enough to understand. I also feel that in a situation where there is, say, a backer at 2.0 and a backer at 1.98, it is silly for us not to match those bets: we were doing so at the price requested, not worse- and therefore the customer matched was getting the best price that we could offer them, bearing in mind the limitations of our technology currently preventing us from giving a price improvement. If we left a backer at 1.98 and a backer at 2.0, people would think we were daft- and equally, if the price was matched by another customer seeing the arbitrage, the initial customer would only be getting the same price.

I accept that I did not consider the difference in-running, as it relates to people mistakenly posting the wrong price, which would take the book over-broke by a significant margin. But I did not think we needed to make an announcement about the fact that we were using our technology to match bets. That is the business we are in.

Betfair Customer Services 19 Mar 18:22
alexe
19 Mar 18:13

Didn&#8217-t anybody at BF imagine this will drain quickly the markets?

Absolutely not – we expected this to increase liquidity and make it easier for our customers to get bets matched. On markets where we’ve had this in operation, that’s exactly what we’ve seen happen – the markets have been more efficient.

Betfair Customer Services 19 Mar 18:23

Andy Fuller: I will email this question but if you read it here first please reply BF &#8211- what have you done with the money you have collected from clients unfairly as you admitted to earlier today?

Under our UK bookmaking licence, these revenues are legitimate profits and have been treated as such. The amounts involved are not what some people were speculating and some people have suggested that the money made should be donated to charity. However, be assured that we will be donating to charity this year far in excess of what the bet matching process has made!

Betfair Customer Services 19 Mar 18:24
Chatname : lippy

Are these changes an attempt to boost betfairs profits for an impending IPO?

There are no plans for a Betfair IPO.

Betfair Customer Services 19 Mar 18:25
Chatname – Get On MASSIVE

Do you ever plan to stop the skimming and give best execution to your customers and if so when?

Betfair has always given a price improvement to bets placed wherever possible, and that approach has never changed. We’d like to be able to offer price improvements across selections too and it’s something we’ve been working on, but it’s much, much more complex than many people imagine , and we don’t have a way of giving customers that improvement yet. We’re hopeful that we’ll have that in place in the next few weeks.

Betfair Customer Services 19 Mar 18:27
Dear BetFair, I hope you will answer the following:

Under current UK regulations you are not required by law to inform your customer base of changes in advance, but under FSA regulations and European Law you would be. Given the strength with which you defended the GC&#8217-s consideration that exchanges should be FSA regulated when the Gambling Bill was first drafted, don&#8217-t you think it would have been wise or prudent to satisfy the most basic of FSA regulations too &#8211- and advise your customer base of material changes to the bet matching algorithm: the key component of the exchange software we all trade on, and one which matches wagers totalling billions of pounds in the UK today.

I am not sure you are correct here: what we were doing here is entirely in line with our licence and what we have always said we do. If you look at repeated statements made about what we are, I (this is Mark) have always stated publicly on behalf of the company that the best definition of a betting exchange is a bookmaker which uses technology to manage its risk perfectly. This is what we were doing here: matching bets in a manner which meant that we, as an operator, had no exposure to the outcome of the event. People have always described Betfair as P2P and told me that our description of the company in these risk-based terms was spin. The reality is the opposite: Betfair is a many-to-many system where demand between customers is matched such that the operator of the exchange does not have risk to the outcome of the event. It is precisely on this basis that we have always been licensed as a bookmaker.

Betfair Customer Services 19 Mar 18:30
Chatname : MoreTea

– Once it was established that best execution was no longer being provided in the new system (in conflict with your own T&amp-Cs and help area), why was it not turned off immediately and an apology made, and why is it still running now?”

– Why did you make a material change to your product which breaks your own terms and conditions and the description of your product in the help area without announcement or warning?”

There was no change to our existing matching process – if a bet could be matched against an opposing bet, that would be done giving the best price available. Adding cross-matching gives another chance to get a bet matched – something we thought, and still think, is an improvement that benefits the vast majority of our customers.

Betfair Customer Services 19 Mar 18:33
Chatname &#8211- Getting Better

Can you confirm that you will not be applying the new matching process to markets such as horse racing that have a reduction factor? If you did I fear that you would be open to abuse on occasions where there was a known or likely non-runner.

Yes, we can confirm that the new bet matching process will not be used on horseracing markets any time soon. If this changes we will let you know in advance.

Betfair Customer Services 19 Mar 18:37
Magician: Specifically did the GC APPROVE this change to the matching algorithm &#8211- or where they simply made aware of it and did not grant or reject formal approval

We did not seek Gambling Commission pre-approval for cross matching before we launched it because we don&#8217-t believe that this was required. However, we were in dialogue with the Commission in relation to the licensing status of cross matching and we would always be happy to address any questions the Commission has on any part of our business.

Betfair Customer Services 19 Mar 18:37
Feck N. Eejit
19 Mar 18:33
Have they answered mine yet? &#8220-Why do horse racing people all wear funny clothes?&#8221-.

John McCririck here… just stopping by Betfair towers, on my to the Ivy, looking for a new gig, what do you mean by funny clothes?

Betfair Customer Services 19 Mar 18:39
the man marcus
19 Mar 18:31

?882 on a 6 mill traded game?

yeah right

We expected to make more money as a result of this change because we believed it would make our markets more efficient and increase the volume matched – the amounts retained through the odds differential are much much smaller than all of the forum speculation would suggest.

Betfair Customer Services 19 Mar 18:40
Chatname Frog

Questions:

1. Is it true that Betfair have started matching bets across selections while not offering best execution? e.g. if two customers want to back different selections in a two outcome event at 1.9 Betfair would lay both outcomes at 1.9 and pocket the overround for themselves.

This is what we were doing but we have announced today that we would take it down until we can deliver best execution as you state it. However, I (this is Mark Davies) think that ‘best execution’ is moot in its definition here. Best execution for me means the best price at which we can execute the bet- and it is not trivial to offer a price improvement on a cross match.

Given the situation I have already suggested, where you have two backers of a 2 outcome event, one at 2 and one at 1.98, I believe that we ought to be matching those bets. Ideally, we should be matching them at 2 and 2, but our technology is not currently able to do that. For me, we should therefore match the bets at the prices asked for.

If you think about there being three options here: that we don’t match those bets- that we match them at the prices asked- or that we match them at the ‘best execution’ you suggest (by which you mean giving the price improvement implied), then clearly the philosophy of the company is to do the third of those. This is what we are working towards. In my judgment, it was better in the interim to do the second, than to leave it at the first. However, clearly many of our users disagree, and this is why we have rolled back to where we were. Personally, I think the second point is a better place to be than the first. I accept that bot users who previously benefitted from that arbitrage will disagree. But, looking back to the early days, many (and I think you were one, frog) objected to bots coming in and taking that arbitrage. You could argue – I would – that it is fairer that we should take that arbitrage, and pay tax and levy on it, than that someone should have a free lunch. However, it is clear that customers disagree and would prefer to leave that arbitrage to the bots. We have therefore decided that we will not match those bets until we can do so with the price improvement.

2. If (1) is the case, is this permanent or do Betfair guarantee they be implementing a best execution algorithm for this in the future?

I think I have answered that above.

3. If (1) is the case do Betfair guarantee that they will never change the current policy of best execution for bet requests on the same selection? For instance if I put in a request to back a horse at 2.0 and there is someone offering 2.5 on Betfair on that horse will Betfair never back it at 2.5 with that person and lay it back to me at 2.0?

I think that goes very much against the philosophy of the company as we set it up, and I would resist it very strongly myself.

Betfair Customer Services 19 Mar 18:41
Chatname &#8211- flapjack

– When will you introduce best price execution on the new matching system for all bets?

We are working on it but it will take a number of weeks.

– On another forum recently, someone made the comparison between Betfair and the bankers in the current Natwest adverts, i.e. you are not remotely interested in what your customers want or what is in your customers’ best interests? Are you aware that this is how a lot of people see you?

We have read all the criticism. Some of it is unfair, in my view (this is Mark here)- some of it pointed out things that we had not considered. Unfortunately you’ll never please everyone. I hope that your assessment that ‘a lot’ of people see us like that is an exaggeration. I think we do spend a lot of time listening to customers, and working to produce a product that they like, and if the perception is that we are not remotely interested, I think it is a misconception. I think we spend a great deal of time listening to customers’ feedback.

Betfair Customer Services 19 Mar 18:43
Chatname – The Magician

Where some users or third party developers informed about this proactively from betfair. Is it betfairs future intention that when changes are made ALL users received the same information regarding the operations of the betting markets

No, customers were informed preferentially. We made the decision that we would answer inquiries individually, and the first question was submitted by a third party developer which was then disseminated more widely. We’ve recognised that we should have communicated this issue to the customer base more broadly, a misjudgement for which we can only apologise, and we’re committed to doing a better job of communicating similar issues in future.

Betfair Customer Services 19 Mar 18:44
Another question to Betfair: To ensure people do not make figures up and blow things out of proportion &#8211- you say you made ?882.91 on the biggest market, but can you confirm that was the biggest takeout you have had from a single event? If it was not can you state what the biggest take out has been and on what event. I think this is important to stop people thinking you are exploiting this situation any more than they already think you are.

TIA. andyfuller

I (Mathias) can assure you that the example given was not unrepresentative or misleading (and cross matching was switched on for the entire market in question). I hope you understand that we have never given out commission data in relation to a market so are not particularly keen to go any further.

Betfair Customer Services 19 Mar 18:47
Chatname : mugsgame

Hi,

I would like to know your intentions regarding using the new bet matching engine on horse racing markets, particularly in play.

There is no immediate plan to operate the new matching process on horse racing markets. There are a number of additional issues with horse racing, particularly withdrawals. If we match bets across runners in a horse race and one runner is subsequently withdrawn we would have to void the bet on the withdrawn horse while honouring bets placed on those that come under orders. Having the ability to offer price improvements where possible is a higher priority, and we’ll revisit horse racing once that issue has been resolved.

Would it be possible for you to put some text into the market rules window stating if the matching engine is being used?

It’s a fair point to want to know if the new matching process is applicable in a particular market. It’s unlikely we’ll make changes to the market rules tab to communicate that, but we will announce future changes to the matching logic and the markets affected in the Service section of the forum.

Betfair Customer Services 19 Mar 18:47
Robin Ewe
19 Mar 18:34

Can you assure us that Betfair will never at any point in the future begin actively trading in the sports betting markets other than to hedge the risks from your multiples product?

No – but Betfair is not in the business of risk-taking on sports markets. That said, we’re always looking for better ways to meet the needs of our customers.

Betfair Customer Services 19 Mar 18:48
Chatname : drifterwins

How much tax will you be paying on money &#8220-skimmed&#8221- from the new system?

The Betfair bookmaking company (which is separate to the exchange) which operates cross matching pays tax at 15% on its profits (like any other UK bookmaker).

Betfair Customer Services 19 Mar 18:55
Lori
19 Mar 18:46

to paraphrase:

&#8220-We think it fairer that we take the arbitrage and pay tax on it, than those who already pay us commission on it, which we then pay tax on&#8221-

I don&#8217-t use a bot, and wish that bots were not an inevitable competitor to me when I trade, but they&#8217-re still owned by paying customers, and I think the customers in a &#8220-customer vs customer&#8221- site deserve the money more than the site. At least they don&#8217-t have to make decisions regarding voiding markets etc.

Also, not all of arb money goes to bots. Also, why not just have one runner in a tennis match

&#8220-Federer win, yes = 100&#8243-

Questions to Betfair

How can you convince us that you can remain neutral as an arbitrator when you have skimmed money from the market?

Why do you believe you&#8217-re going to be popular taking more money from a market than advertised commission rates?

If cross matching at best execution is so difficult, why not just have one-runner in two horse races and settle at 0 or 100 like many other places.

To the first point, we’re not taking a position – merely matching bets that otherwise would not have been matched. Most of our customers simply want to get bets matched, so we believe improvements in our matching process are in our customers’ interests. Having just one runner in 2-runner races would make cross-matching unnecessary but it’s confusing as most people expect to see both names.

Betfair Customer Services 19 Mar 18:58
The Magician (1)
19 Mar 18:54

another quick Q.

why dont you add this new matching algo to the horse racing SP calculation (with best execution), it would certainly make it more robust than it currently is

We are considering adding the cross matching with best execution to our SP markets in the future. However, this is a complex calculation and therefore will only be done at a later stage.

Betfair Customer Services 19 Mar 19:01
Chatname (WH)

Why is giving best execution so difficult?

If 2 people are backing at 2.0 and 1.8 in the same market, it is surely simple to ensure that the person whose request came in first gets the odds they requested and the second person gets better odds than they thought they would get

eg if 2.0 is waiting to back player A and someone submits 1.8 to back player B, then both get matched at 2.0.

if 1.8 is waiting to back player A and someone submits 2.0 to back player B, then A is matched at 1.8, B at 2.2

i cannot see why this is so difficult

It isn’t difficult to calculate for a single instance. Unfortunately Betfair’s bet matching process has to be able to calculate this across multiple markets for thousands of bets each second, and it would be an unacceptable customer experience if doing so caused any delay to the bet matching process. The existing bet matching process has been refined over years, and making fundamental changes to that means coming up with a whole new set of refinements. We’re working on this now and we will have it in place as soon as we’re satisfied the solution gives the performance customers expect.

Betfair Customer Services 19 Mar 19:02
Thanks very much taking part in this discussion. We’ll continue to answer the questions we receive on this subject, via e-mail, and post the relevant Q&amp-As here for you to read at a later date.

Previously:

BetFair Malta’s combo market maker (trading algorithm + human market makers) operating on the multiples

BetFair withdraws / improves its brand-new matching-bet logic, which was (kind of) endorsed by the Chairman of the Midas Oracle Advisory Board.

One un-hired job candidate and one HammerSmith employee tell all about BetFair Malta’s combo market maker (trading algorithm + human market makers) operating on the multiples.

One un-hired job candidate and one HammerSmith employee tell all about BetFair Maltas combo market maker (trading algorithm + human market makers) operating on the multiples.

No Gravatar

Graham “Sharp” Minds, in a comment on Midas Oracle:

I am one of the people who Ed Murray has referred to in this blog. I was interviewed at Hammersmith by Betfair during the summer. For the time being I am going to concentrate on what Betfair has finally admitted from the past, present and future plans that betfair outlined to me.

&#8220-Loki Lab Rat 18 Mar 08:20 I can answer this one quickly. There is a trading team based in Malta which manages the risk around the multiples product. They have software which tells them what the risk is associated with a potential result is and suggests what hedge bets can be placed to mitigate that risk at current exchange prices. They then place the hedge bets to manage the risk. They place the bets using the same software as everyone else using the site and respecting any in-play delays.&#8220-

There is 3 objectives for the &#8220-Maltese&#8221- trading team, &#8220-Maltese&#8221- is in quotation marks because betfair were unclear to me about who and what parts of the trading team were still going to be based in London and those who were going to be in Malta.


i) Hedging
ii) Trading
iii) Arbing

All 3 amount to the same thing, Betfair playing their own markets.

On each of the 3 points

i) Hedging, this is what Betfair has admitted to and has been doing to from day 1 of the multiples. Betfair admit to taking the other side of the bet for the mutiples, so I guess like any other bookmaker it is only natural they &#8220-hedge&#8221- their bets to reduce potential liabilites.

However unlike traditional bookmakers, who will lay off a multiple when one runs up, the bets taken by Betfair multiples are placed directly back into the exchanges before they &#8220-run-up&#8221-.

This is one of the reasons behind the mutliple groups, they look at the back/lays placed on the mutiples for the first match and then the bets are placed into the exchange to reduce the liabilites and to cream off the difference between the price they have offered and that offered on the exchanges.

ii) Trading, this is what betfair are attempting at the moment, albeit unsuccessfully from what I last heard. Betfair have admitted they change their hedging positions, this by definition is trading. however the most serious aspect of their trading, is the in-running trading.

If betfair have accepted liabilities X &amp- Y for events A and B being the outcome of a match, then why would liability X for A or Y for B be no longer acceptable during a match unless there was an advantage to be gained by trading betfair’s position in-running?

Afterall, on betfair’s exchange with a 100% book, whether that 100% book was an efficient or inefficient, the net expectation of betfairs position would remain the same if they were trading hedging positions blindly to &#8220-balance&#8221- the liabilities.

In the instance of an inefficient market, if betfair were blindly trading (i.e. not using odds compilers, traders or other means to form an opinion on the market) in-running then there would be no net gain as betfair would be equally likely to be on the efficient and inefficient prices and the differences would average out.

iii) Arbing, this is betfairs ultimate goal for the multiples. If this new cross-matching bot is being called a superbot, then this will be the hyperbot. A bot which will perfectly arb the mutiple bets into the exchange so betfair can make money from it’s mutiples operation risk free.

But if betfair are denying that they never form an opinion to a market and place bets accordingly and they have no plans to do this in the future either. Then why are they covertly recruiting academics from universities, odds compilers from the bookmakers and recognised traders from their own customer base?

If betfair are playing their own markets, then what is to stop them abusing the position they have by owning the exchange. Having worked for several bookmakers in the past, I have yet to know one who would voluntarily wait 5 secs to place their &#8220-hedges&#8221-, wait in the queue like everyone else, make trading decisions according to the flow of money or not use their &#8220-warm sources&#8221- wisely.

Via Dave, Loki Lab Rat:

I can answer this one quickly. There is a trading team based in Malta which manages the risk around the multiples product. They have software which tells them what the risk is associated with a potential result is and suggests what hedge bets can be placed to mitigate that risk at current exchange prices. They then place the hedge bets to manage the risk. They place the bets using the same software as everyone else using the site and respecting any in-play delays. The line about &#8216-opportunities in-play&#8217- refers to the match situation and is nothing to do with beating the delay.

For example, when just about every favourite won their international Euro 2008 qualifiers in early June last year (known as &#8216-Black Wednesday&#8217- in some quarters), Moldova scored late-ish against Greece to square the match. The multiples team heavily laid the draw as the match drew to a close as it was a much better potential result than a Greek win. That meant that when Greece did score with the last kick of the game, our losses were mitigated to some extent by the hedge bets placed in-running.

The multiples product is run under Betfair&#8217-s Maltese bookmaking license and is regulated by the LGA there. Therefore the team has to be based in Malta. The operation is an arms-length operation &#8211- there is no special access to any functionality or data from the exchange.

Loki Lab Rat:

frog2, It is best I leave questions on cross matching for the Q &amp- A. I am not really that close to the project. I just thought I would answer the one on the multiples team to prevent any speculation getting out of hand.

Rab Bibater in a comment on Midas Oracle:

It is also pertinent that this team operates under Maltese jurisdiction- which of course puts it outside the scrutiny of the UK’s Gambling Commission!

However, I think there is also another issue that needs to be explored, and that is the general efficacy of a policy that allows Betfair’s own employees to trade on their exchange. Many people within Betfair have access to historical trading data unavailable to the broader betting public- such as what stables are backed and when- the significance of early money for particular horses etc- some are also able to access the betting records of individual persons. I specifically remember a comment made on Betfair radio concerning the boys in the office and how they had mopped up all the value regarding a particular horse from a gambling yard. The problem for Betfair, and one that is not likely to go away, is that nobody is sure any longer as to who it is they are actually trading against.

Business Risks & Prediction Markets

No GravatarTake a look.

Previous blog posts by Chris F. Masse:

  • Red Herring’s list of the top 100 North-American high-tech startups includes Inkling Markets —but not NewsFutures, Consensus Point, or Xpree.
  • Professor Koleman Strumpf explains the prediction markets to the countryland people.
  • Professor Koleman Strumpf tells CNN that a prediction market, by essence, can’t predict an upset.
  • Time magazine interview the 2 BetFair-Tradefair co-founders, and not a single time do they pronounce the magic words, “prediction markets”.
  • One Deep Throat told me that this VC firm might have been connected with the Irish prediction exchange, at inception.
  • BetFair Rapid = BetFair’s standalone, local, PC-based, order-entry software for prediction markets
  • Michael Moore tells the Democratic people to go Barack Obama in Pennsylvania (a two-tier state), but the polls and the prediction markets tell us that that won’t do the trick.

Polls Vs. Prediction Markets

No Gravatar

Asia Times:

[&#8230-] Outperforming Taiwan&#8217-s polls shouldn&#8217-t be hard. They&#8217-re notoriously bad as a forecast of election outcomes. In late 2006, for example, many media polls underrated the pro-independence party&#8217-s support – a recurring problem. Taiwan&#8217-s prediction markets did a much better job of estimating vote shares (the island&#8217-s two markets both called the Kaohsiung mayoral election wrong, but that contest was a statistical dead heat). &#8220-Most opinion polls usually have 20 to 30% &#8216-no answer&#8217-,&#8221- said Lin Jih-wen, director of the Center for Prediction Markets. &#8220-We don&#8217-t have missing data or a sampling bias, that&#8217-s our strength.&#8221-

The market has now picked the strong likelihood of victory by the China-friendly candidate Ma Ying-jeou. Time will tell if it&#8217-s got the right guy. But even if it doesn&#8217-t, the markets&#8217- enthusiastic reception shows how Asia – like the US and Europe – has embraced such markets as a powerful fortune-telling tool.

A &#8220-powerful fortune-telling tool&#8221-? Jesus. :-D

&#8220-Outperforming&#8221- the Taiwanese advanced indicators which they are feeding on? Humm&#8230- :-D

[Mike Giberson will write a comment, below, reminding me of Prof Koleman Strumpf&#8217-s work (PDF file) showing that the historical prediction markets were accurate enough, even though the scientific polls were not invented yet. Yes, I know of that, Mike, but I still don&#8217-t get whether it&#8217-s a puzzle or a mystery. :-D &#8230- Do you?]

Read the previous blog posts by Chris F. Masse:

  • Business Risks & Prediction Markets
  • Brand-new BetFair bet-matching logic proves to be very controversial with some event derivative traders.
  • Jimmy Wales accused of editing Wikipedia for donations.
  • What the prediction market experts said on Predictify
  • Are you a MSR addict like Mike Giberson? Have nothing to do this week-end? Wanna trade on a play-money prediction exchange instead of watching cable TV? Wanna win an i-Phone?
  • The secret Google document that Bo Cowgill doesn’t want you to see
  • BetFair’s brand-new matching-bet logic is endorsed by the Chairman of the Midas Oracle Advisory Board.

Towards a prediction market label?

No Gravatar

Our good doctor Emile Servan-Schreiber of NewsFutures (some time ago):

[…] create a “Prediction Market” label that we could use to differentiate our products and our approach from the growing number of &#8220-wisdom of crowds&#8221- barbarians (eg, blubet, guessnow, etc.) who are shamelessly exploiting the &#8220-prediction market&#8221- brand even while their offerings have little resemblance to such a thing. […]

Maybe not such a bad idea. But who would define the prediction markets, who would apply this prediction market label, and would the media care about all that??

Collective Prediction – Combining human and machine intelligence in prediction economies

No Gravatar

MIT&#8217-s CCI:

Think of a domain in which you would like accurate predictions of future events: sales volumes for a company&#8217-s products, outcomes of sporting events or military conflicts, crime rates by neighborhood, terrorist actions, new products introduced by a company&#8217-s competitors, or the clinical outcomes of different medical treatments someone you know might receive for cancer.

Now imagine a network of humans and computers that makes predictions in this domain&#8211-not perfectly, but better than was possible before. And imagine that these predictions get better and better over time as the network learns from its own experience. We propose to do some of the essential research needed to help create such networks.

A number of researchers have recently developed prediction markets in which participants buy and sell predictions about uncertain future events and are paid only if their predictions are correct. Such prediction markets have been found to be surprisingly accurate in a wide range of situations (including forecasting product sales and US Presidential elections).

We propose to build on this previous work to develop prediction economies &#8212- networks of people and computers paid (either in currency or points) for accurate predictions about future events. A prediction economy can include (a) one or more prediction markets (b) markets for various other kinds of information relevant to the events being predicted, and (c) markets for services by people (such as image analysis) or by machines (such as multiple regression, machine learning, and data mining).

Importantly, both people and their automated agents will be allowed to participate in any part of the economy. For instance, automated agents can do &#8220-program trading&#8221- in two related prediction markets whenever they see inconsistent prices in the two markets. In this way, prediction economies provide a flexible new approach to integrating human and machine expertise: People have an incentive to create new automated agents whenever they can codify useful expertise algorithmically, and they have an incentive to participate in markets directly when they can do a better job than the existing automated agents. But when people can&#8217-t improve on what the automated agents are already doing, then they have no incentive to intervene.

Drawing on theories in organization science, computer science, cognitive science, and economics, this work will develop new forecasting and collaboration tools that blend human and machine capabilities to more accurately forecast risks and opportunities, thus helping to build more agile systems in many domains.

Hummm&#8230- More meta than the prediction markets? Is that possible?&#8230- Will their proposal fly? Humm&#8230-

Read the previous blog posts by Chris F. Masse:

  • Ratted by his bank, sex-addict New York governor Eliot Spitzer (alias “Client 9”) resigns.
  • BBC’s coverage of politics is dull like taxes, death and German sausages.
  • Never talk when you can nod, and never nod when you can wink, and never write an e-mail because it’s death. You’re giving prosecutors all the evidence we need.
  • Is Justin Wolfers a libertarian? Probably not.
  • The information technology that caught Eliot Spitzer
  • Eric Zitzewitz’s 10 minutes of fame
  • Fun with conditional probabilities