You cannot be considered a serious gambler if you...

If you want to know where the market is headed I just read a piece that pretty much sums it up. Pissy little matchbook is nothing

Betfair is a Gambling Success Story...By Hartley Henderson

---

I recall fairly vividly my first bet ever. I was in second grade and a friend and I had a disagreement about a math problem our teacher gave us before recess. I told him the answer was 8 (or something like that) and he replied it was another answer (I believe 6). He asked me if I was willing to put up the 2 chocolate chip cookies I had in my lunch pail against his 2 Oreo cookies on which of us was correct. I agreed, and to my delight after the teacher gave the answer I later had his 2 cookies at lunch. Being the nice kid I was I only took one of the cookies, but it nonetheless was my first bet and a winner at that. Those types of bets happen all the time. When 2 teams reach a championship final game, invariably the mayors of the 2 cities will place a bet with the winner getting some token gift from the losing city's mayor. What makes all those bets different from betting at a sportsbook, of course, is that the bet is player vs. player rather than the bettor playing against the bookmaker, as most are accustomed to.

By 2000, online sports betting was well established, but it was all player vs. bookie. There were some unique variations of the traditional bookmaking model (such as what World Sports Exchange developed), but for the most part it was still the player betting against the bookie, with the bookmaker taking the edge via the juice built into the odds. With the internet becoming more sophisticated, and the ease of allowing players from different parts of the world to access the same information at the same time, two entrepreneurs, Andrew Black and Edward Wray took advantage of the new internet age by allowing players to bet against each other and making it seem as if the company simply acted as an intermediary. Under this model the players set the lines and bet against each other. The intermediary (Betfair) makes its profit by way of a commission charged on the net winnings of any given market. The company has no vested interest in the outcome, as it assumes no risk. One bettor would put up lines on a game or set a price they would like to have and another bettor would either take the odds being offered (backing the bet) or give the price to the bettor who requested the odds (laying the bet). What makes the model so attractive for all is that Betfair, because it takes no risk, doesn't have to build in an additional margin for matched bets. The odds offered on both the back and lay are, by definition, identical. In a typical NFL football game it would not be uncommon to see both teams having back odds of 2.0, or in American betting terms, each team is even money to win by the point spread. It makes perfect sense. After all, when I bet with my friend on the math problem there was no vigorish built into the bet - i.e. "I want 2 cookies and a bite of your sandwich to my 2 cookies in order for me to take that bet". At Betfair a commission of 5% is taken from winning wagers, although that percentage can drop to as low as 2% in certain conditions. So if both teams are available at even money, then a $100 winner would yield a profit of $95 and the losing bet would give up $100. This is in comparison to a typical book offering a 20 cent line where each player must wager $110 to try and win $100. In all fairness to sportsbooks, however, the vigorish incorporates both profit and risk, whereas in an exchange there's no risk for the operator to have to account for. While the ideal book will have equal action on both sides of a line, that rarely, if ever, happens. A sportsbook will always have a position at the onset of a game.

Several other places tried to emulate Betfair's model, but none have come close. Trading Sports, Tradesports, Mansion Exchange, among others, all closed shop. And numerous sportsbooks that offered peer to peer (p2p) betting gave it up when they couldn't realize the volume or interest that Betfair has. There are still a few existing p2p operations, like Sporting Index, Betdaq and Matchbook, but none can offer the amount of markets that Betfair does, nor generate the volume. Betfair has over 2 million clients and generates close to US$100 million a week in bets. Needless to say, with that type of volume the markets are always tempting and vibrant.

Another feature which Betfair offers, that most sportsbooks don't, is betting while the game/tournament is in play. Some sportsbooks do offer inrunning wagering, but the limits are usually quite low, because the risk to the operator is too great, and the number of events is relatively small. As well, they are always offered as updated odds. Betfair simply allows the bettors to update the markets to what they feel are fair odds. No company, other than Betfair, offers inrunning wagering for horse racing. World Sports Exchange actually has a similar idea to Betfair. The company offers markets on certain events with a buy/sell share price on each. The share price is worth $100 if it wins. The concept is quite good, but the number of markets offered is limited and the bettor is still playing against the house rather than against another player directly. Consequently WSEX sets the market price, not the player. As well markets are always paused when an event is off the air due to a commercial or such, whereas Betfair markets stay open until the close of the event. Inrunning horse racing is something to behold. Right now Betfair only offers inrunning horse betting for European racing and selected major races, like the Kentucky Derby or Breeder's Cup. But for a short time they did offer betting inrunning on North American racing. What makes betting on horse racing while it is being run unique is the strange odds people are willing to offer. Every year there will be at least a handful of races where a huge favourite will appear to be faltering and the odds will climb to 999/1 (the maximum odds allowed). Sure enough the horse will find a second wind and the person(s) that offered 999/1 on a 2/5 shot is in the news provided he didn't jump off a bridge. Horse racing has always been Betfair's biggest volume sport.

I was curious as to why Betfair succeeded while so many others failed, so I contacted to Mark Davies, one of Betfair's founders, who was willing to provide answers.

Two things made the difference: we got it right; and we were first to do so.

Today, everyone who tries to set up a site like ours does exactly that: makes a site that looks like ours. But people forget that the way to do it wasn't so obvious when we started: six sites set up with a similar idea to [get rid of] risk inside six weeks of each other - and five of the models failed. Where ours differed is that it was the only order-driven trading system, which means it collated supply and demand and matched it up on a best-execution basis. To put that into English, if you went into your grocer and asked for 5 apples, for which you would pay 20p each, and he replied that he didn't have 5, he had 50, and anyway they weren't 20p, they were 10p, you would tell him not to be such an idiot: you'd take five of his 50 and you'd pay him 10p for them. Betfair's system allowed this to happen, whereas all the others needed the bid and offer to be perfectly matched: they were basically just bulletin boards of one sort or another where you could post what you had for sale or tell people what you wanted to buy. That meant that the more business they had, the more cumbersome their site became, whereas with us, whether we had 10 available or 10 million available at a given price, the customer saw the same thing. It seems obvious now, but it wasn't at the time: like all the best ideas, ours was very simple, and very well executed.

By the time anyone else had re-created the software, we had significant liquidity, and ultimately this industry is a liquidity play: there's no point in having much better prices if you can't fill the demand. And exchanges naturally lend themselves to size: the biggest will always be multiples bigger than the next biggest, which will be multiples bigger than the next biggest, in turn.

I was also curious why so many sports leagues were willing to partner with Betfair, while in North America the leagues tend to see sportsbooks as the enemy. Right now Betfair has agreements with several European leagues, the ATF tennis federation and recently announced a provisional agreement with the National Hockey League. Why Gary Bettman has had a change of attitude is uncertain, but the NHL clearly doesn't see Betfair as a combatant anymore. Mr. Davies commented as follows:

Because Betfair takes no risk, [it only] matches up supply and demand perfectly through its technology, it has no exposure to the outcome of any event. This means that its interests are entirely in line with the regulator: a fair result, no matter what the result is. In addition, it has a full audit trail of all transactions on its site, which it is happy to share. Transparency is the key tool in fighting corruption, and many sports recognise that and appreciate the access to the information, which Betfair offers free of charge. It is a totally different proposition from an opaque offering where the person holding all the cards may benefit from a specific result.

This was actually best evidenced in a tennis match in 2007 at the Prokom Open in Poland between Martin Arguello and Nikolay Davydenko. Arguello was a relative unknown and Davydenko was a star. The few sportsbooks that offered the match had Davydenko as about a 1/10 favorite. Betfair, however, had money come in on the match for ridiculous amounts that would never be seen for such a minor tennis match. As well, the money was all on Arguello. Davydenko pulled out of the tournament with "an injury" at the start of the second set and Betfair immediately halted any payouts. They decided to launch an investigation into the unusual betting patterns and decided that the game was suspicious and ruled all bets invalid. This event demonstrated precisely what most sportsbooks and Las Vegas have been saying for years - that if leagues and sports betting establishments work together they can help weed out cheating and catch criminals. It is quite possible that if Betfair had a similar arrangement with the NBA, games involving Ted Donaghy would have been recognized as suspicious. Of course what Betfair has that no other traditional sportsbook does is transparency. Because all odds are posted on the site, along with amounts matched, it is easy to determine unusual betting patterns - not just by Betfair, but by the bettors themselves. Traditional sportsbooks would have to set up systems to identify the patterns they are looking for and then rely on people at the books to catch it. Furthermore, if the volume isn't high, or if limits on the event are low (as would have been the case with a nothing tennis match in Poland), it probably would never be caught.

Of course not everything at Betfair is perfect. Some markets do indeed get very little volume, making the spread between the back and lay quite high and virtually unbettable. Further, the company has had a falling out with other UK based sportsbooks like Ladbrokes and William Hill that are upset at what they see as preferential tax treatment. Those sportsbooks also object to the idea of being allowed to bet against horses to win, as they feel this may lead to race fixing and corruption. In all areas Betfair has been seen as operating fairly, and in fact have won awards for their format. As Mr. Davies commented:

Every independent observer or commentator who looked at the issue concluded that Betfair was right and Ladbrokes was wrong. Eventually Ladbrokes conceded defeat. Most now accept that it was a commercial battle dressed up with legal rhetoric: the bottom line was that we were a very competitive offering [compared] to the rest of the industry, and they did everything they could to slow our progress. They aren't the first to adopt that tactic, and they won't be the last. You can hardly blame them: you'd probably have done the same in their shoes!

Betfair does not allow betting from U.S. citizens, and never has, claiming they don't want to operate in a grey area. However, with a recent decision in Washington State that exonerated Betcha.com (which was also a p2p betting site), and with an administration that seems less confrontational, and many states now looking at legalizing sports wagering, it isn't inconceivable that Betfair could someday soon enter the U.S. market.* After all, a few reports have suggested that the volume of business that could be generated by legalized U.S. online sports betting exceeds $30 billion a year, and a tax on that amount is nothing to sneeze at. As well, with Barney Frank and others pushing strongly for the elimination, or at least a watering down, of the UIGEA it's obvious that it's just a matter of time before online gambling is legalized in some way, and at that point it will be hard for the U.S. to defend their protectionist efforts. As for the tax issue, Betfair already has a system whereby Australian bets are separated from UK bets so that Australia gets the tax from bets for events in that country. So there's no reason they couldn't do the same in the U.S. Furthermore, the NHL is willing to work with Betfair. So if they can demonstrate to leagues how the transparency of their system will help guarantee the legitimacy of games and catch cheaters, and if they can show that p2p betting is strictly betting between peers (which is not illegal technically anywhere in the U.S.), it isn't inconceivable that the U.S. leagues may soften their stance; possibly, but not likely. Betfair also has the advantage of never having operated stateside, so the DOJ has never expressed an interest in them. Finally, Betfair recently purchased TVG networks in the United States in an effort to get their foot in the door, even if it's not by way of offering betting to Americans on their sports betting website.

Betfair is indeed a huge success story, and if their system ever does open to the U.S. market be prepared for a treat that Canadians, Europeans, Asians and others in the world already have.

When you guys use Bayesian theorem, what is your time frame of your data  (for instance % home team to win, do you use whole season, half season, multi-season)?

Interesting stuff.

I think the Trade Exchanges will eventually push out and eliminate the "bookie" all together. Surprised they haven't caught on in North America too much yet other than Matchbook.

 non-serious gambling

Interesting to read old treads...

Don't know about this 1, a junkie needing action doesn't mean you're not a serious bettor....IMO

If you're in it just for the money then probably not the wisest idea, but I don't think all gamblers r just in it for the profit (the action is what a gambler feasts on)...Just my .02