Sunday 28 February 2016

Oscar Math

It appears to have been seven years since I last wrote about betting on the Academy Awards, which at the time of writing are just a few hours away.

The 10 years of betting on this event has resulted in a profit of £630.50, with a high of £1,381.68 in 2008 and a low of -£1,780.40 in 2007. 

As I mentioned in 2007, other awards leading up to the Academy Awards usually give a big hint as to where the statues might be headed. Unfortunately over the years, this 'edge' has vanished with many others letting the cat out of the bag!
You can predict the Oscars with math. We’ve got all of this data from previous years — Guild Awards, other award shows, critic scores, which categories you’re nominated in — and we can use all of these things to figure out how important each of them are to predict each Oscar category. Then plug in this year’s data, and we’ve got ourselves a formula. It’s not perfect. Math is probably never going to go perfect, but, frankly, neither are qualitative predictors. I think at least math can definitely add something to the conversation each year. - Ben Zauzmer
So it'll be a more cautious approach this year, as it has been for the last few. 

Of the 24 markets available on Betfair, all but three have an odds-on favourite, with eight (including Leading Actor and Leading Actress) at ~1.12 or less. 

What jumps out at me is that the Best Director favourite (at ~1.22) is Alejandro González Iñárritu whereas you can back his film (The Revenant) for Best Picture at ~1.44 and if this wins, it will be another profitable year whatever happens elsewhere.

I'll be staying away from Sound Mixing once again this year, and I have still never seen a Star Wars movie.  

Sunday 21 February 2016

Full Bloom

Tony Bloom
My old friend Scott sent me a link to an article in Business Insider about Tony Bloom and Starlizard. It's a good read, as are most of Scott's recommendations, although with rather too many references to Brighton and Hove Albion for my liking, and the "vying for promotion to the Premier League next season" line was particularly disturbing.

As I have previously mentioned, a relative works in the football betting industry, so much of the business model and the scale of the operation is not new to me.

However, I suspect that there are still many people who think that making a consistent profit from the top football leagues is realistic, despite their having far more limited resources than the enterprises detailed in the article, which does have some good quotes worth digesting:
  • "I wanted to gamble because I enjoyed it and, therefore, I needed to do it properly in order to win."
  • "Early on I was a hopeless gambler really, I liked to think that I understood the form and had a strategy but I was just guessing really."
  • "At university I made myself a promise that I would become fiercely disciplined. I wanted to gamble because I enjoyed it and therefore I needed to do it properly in order to win. I didn't want to lose my money."
  • "They don't beat the market all the time, just enough times."
  • "We're not trying to say this is going to happen, we're trying to say this will happen with a certain probability. If our probabilities are better than those of the bookmakers, then, in the long run, our clients will win money."
  • "...as the years have worn on it has got harder and harder to beat the bookies, with razor-thin margins squeezed even more"
I was also a little surprised to see Keith Sobey referenced in the article, albeit with a quote from back in 2010. As readers of this blog will know, Mr Sobey was the man behind, the not so magnifico, Galileo, "a previous sports-gambling fund managed by Centaur Global Ltd., which folded in 2012 owing creditors more than 2 million pounds ($3 million)".

"Centaur overstretched itself while still working on such algorithms when it opened its sports fund in 2010, according to Poots. Former Centaur CEO Keith Sobey couldn’t immediately be reached for comment". 

Perhaps not a surprise. What was a bit of a surprise was that while researching this post, I discovered that Peter Webb appears to have changed his name:
Priomha employs three statisticians to work with traders, and bases as little as 20 percent of decision-making on “qualitative” judgment, such as the effect on a team of the injury of a key player, Poots said. It hedges betting to offset potential losses.
Even if the fund’s model is effective, its expansion will be restricted by the volume of regulated betting in sports, according to Peter Blake, founder of Hook, England-based Bet Angel Ltd., which sells software to help individuals to wager on sports trading exchanges such as Betfair Plc.
What a tangled Blake we weave...

Interesting to note that while Priomha bases "as little as 20% of decision-making on 'qualitative' judgment", Starlizard makes "cold, calculated decisions about where to stake cash, separating the decision-making from the money and making it as mathematical as possible — no gut feelings". 

My gut feeling is to be very wary of investing with Priomha. 20% is a lot! Goodness knows how high this subjective percentage can go.  

Here is the "full Bloom" post:

Inside Starlizard: The story of Britain's most successful gambler and the secretive company that helps him win

Camden, North London. It's 12:45 p.m. on a Saturday, and the punks, tattoo artists, and tourists are gearing up for another day of drinking and shopping in the area's famous tunnels, warehouses, and bridges.

But the offices at the Iceworks building, overlooking the canal, are filled with smart young professionals. It might be the weekend for the rest of us, but for these workers it is the equivalent of Monday morning, the busiest day of the week, and the opening bell on their market is about to ring.

They have a huge sum of money on the line because they are in charge of the biggest gambling syndicate in Britain, believed to make up to £100 million, or $145 million, in a good year.

A mixture of men and women — mainly men — ages 25 to 45 gather around the TV screens dotted around the Camden office, some displays mounted on walls and others at desks. All are tuned to the weekend's football.

The first Premier League match of the weekend is about to begin. And with it, a weekly multimillion-pound gambling bonanza kicks off too. Their company can have £1 million riding on the outcome of a single match and more on the nine others that will follow in the next 24 hours.

But this isn't a bookmaker. It is Starlizard, a company that treats gambling the way hedge funds treat stocks. Officially, it describes itself as a betting consultancy that uses complex statistical models to generate football odds that are sharper than those offered by professional bookmakers. These are then sold to clients to help them beat the market. The company thus acts more like a betting adviser than a bookmaker — it doesn't actually take bets.

But the highly secretive company also masterminds one of the most successful professional gambling syndicates in the world, placing hundreds of millions of pounds worth of bets each year on behalf of high-roller clients.

Football is its biggest business, and if the goals don't go the way Starlizard's models predict, then people will lose a lot of money.

The bulk of the money Starlizard bets comes from Tony "The Lizard" Bloom, a maths whiz who earned his nickname for his cold-blooded decision-making at the poker table.

Bloom, a veteran gambler who owns Brighton and Hove Albion Football Club, made millions setting up an online bookmaker and poker websites in the 2000s, and his net worth — which is a mystery — is estimated by some to run into the billions.

Bloom set up Starlizard to run his sports activities, and the business allows him to bring the cool heads and statistical rigour of Mayfair's boutique quant investment world into the murky arena of Asian bookmakers. He told one interviewer, "I wanted to gamble because I enjoyed it and, therefore, I needed to do it properly in order to win."

Starlizard workers are invited to share in Bloom's winnings. They are offered a free stake in Bloom's syndicate, putting them in line for payouts of up to £500,000 every six months — assuming the match results go Bloom's way, of course. If they don't, employees and other syndicate members must top up Bloom's gambling pot from their own pockets.

But the syndicate is more successful than not, and Starlizard's record in steering Bloom to victory has won the softly spoken Brighton-born 45-year-old a reputation as one of the most successful professional gamblers in the world.

Despite the huge sums involved and the wild success enjoyed, both Starlizard and Bloom’s syndicate have gone largely unnoticed outside the world of professional gambling.

All Starlizard employees are made to sign strict nondisclosure agreements when joining the company, and Starlizard does not engage with the press. Bloom will give interviews only about Brighton, and even then he generally speaks only to local media.

Bloom and Starlizard each declined to speak with Business Insider for this article. Both also declined to give any comment.

But Business Insider has spent the past few months investigating the company to understand just how it works. We persuaded several former employees to speak on the condition of anonymity, talked to industry insiders, and combed through old press cuttings to piece together a definitive history of the company and its founder.


Tony Bloom, the most successful sports bettor of his generation, first tasted the thrill of gambling young.

"From the age of 8 or 9, I used to go down to the arcades in West Street with some friends and play with our pocket money on the fruit machines," Starlizard's architect told Brighton's local paper The Argus. He gave the rare interview shortly after taking over his boyhood football club, Brighton and Hove Albion FC, in 2009.

Bloom, then 39, was at that point rich enough to pump £80 million worth of unsecured and interest-free loans into the club. His net worth is unknown, but there is speculation it could run into the billions. The Daily Mail reported that Bloom has to date sunk about £200 million into Brighton since taking control.

The money Bloom has pumped into the club has helped Brighton win promotion to the Championship, where they now sit in fourth place, vying for promotion to the Premier League next season. His devotion to the club has made him beloved among fans, who can occasionally spot him on the train to away matches.

When Bloom took over the club, The Telegraph dubbed him a multimillionaire property developer, and The Argus attributes much of the wealth to a portfolio of private equity and property investments. Neither mentioned Starlizard.

The word that comes up most when you ask former Starlizard employees about Bloom is "nice." The 45-year-old is well liked in the office but is seen as something of an unknown, suggesting that, as at the poker table, he keeps his cards close to his chest.

Despite being a millionaire many times over, Bloom, as well as the associates who are believed to profit from his gambling, is not flashy with his wealth, according to former Starlizard employees. "They don't all drive around in Ferraris," one says.

He is described in the press as an intensely private family man, and he has a 7-year-old son with his Australian-born wife, Linda, a psychologist. The family divide their time between a home in North London and Australia.

Bloom and his wife are trustees of the Bloom Foundation, and his wife runs the charity Overcoming Multiple Sclerosis. Linda was found to have MS 15 years ago.

Born in 1970, Bloom grew up in Brighton, the seaside town an hour south of London. He was educated at Lancing College, a £23,000-a-year private school founded in 1848 whose alumni include novelist Evelyn Waugh, playwright Sir David Hare, and Sinclair Beecham, the cofounder of Pret a Manger.

The boarding and day school is focused around its 50-meter-high chapel, which dominates the leafy grounds. Chapel attendance is compulsory for all pupils.

Lancing itself is rural, surrounded by fields. But the school is a half-hour drive from Brighton and,according to former England cricket captain Mike Atherton's 2006 book "Gambling," Bloom used a fake ID during visits to town to get into betting shops at just 15 — three years below the UK’s legal age for sports betting.

Bloom went on to study mathematics at Manchester University, where he continued his sports betting. "Early on I was a hopeless gambler really," Bloom told Atherton. "I liked to think that I understood the form and had a strategy but I was just guessing really."

Bloom worked at the accountancy firm Ernst & Young after graduating. He was still betting on sports, earning a bankroll of £20,000 from bets by the time he left Ernst & Young in 1993 to enter the City as a trader.

After just six months as an options trader, Bloom decided to become a professional gambler. He bet on football and cricket, at one time losing £5,000 on a single game of cricket: the England v West Indies test match of 1994.

"I believe in betting aggressively," he told Atherton. "And, occasionally, to win big, you have to risk losing."

Despite the test-match setback, Bloom won more often than not. His success caught the eye of the bookmaker Victor Chandler, which approached Bloom in the late 1990s to set up its international betting operation.

This job introduced Bloom to the market where he would make millions, Asia.

Sports betting and gambling are huge in Asia, but gambling on football operates very differently there, using a system called the Asian handicap.

Originating in Indonesia, the Asian handicap system is meant to even the playing field for both teams by giving the underdog a theoretical goal advantage.

To give an example: Manchester United are playing Norwich. Manchester United, the favourites, have a 2+ handicap, meaning they must win by at least two goals for a bet on them to succeed. Odds are given in decimal format rather than fractions. A two-goal handicap would be Manchester United at 2.0, for example.

In most cases the handicap will be much more fine-tuned than the above example, reflecting more precisely things like form and injuries — Manchester United 2.8, say. Payouts become more complicated for handicaps such as these, but the essential thing to grasp is that the handicap system rests on the number of goals scored by each team.

If a team you've backed can't overcome the theoretical goal deficit — if Manchester United win by only 1-0 or 2-1 — you lose.

Bloom enjoyed much success setting up Victor Chandler's Asian betting operation, using his maths background to crunch stats on teams and come up with handicaps.

Bloom told The Argus in 2009: "I was one of the first people outside of Asia to take a keen interest and an understanding of it. I worked in Thailand for seven months, then Gibraltar for three years."

The 1998 World Cup in France was a pivotal moment in Bloom's life at Victor Chandler. Convinced the market was underestimating the odds of a French victory, Bloom persuaded Victor Chandler's management to bet everything it had won so far from the tournament on France to beat Brazil in the final. The host's 3-0 victory over Brazil netted a huge prize.

A former Starlizard employee described the story as the founding myth of Bloom's career. But no one Business Insider talked to knew the exact amount staked or how much Victor Chandler won. The company later became BetVictor, one of the UK's better-known online gambling brands.

Emboldened by his success at Victor Chandler, Bloom went on to set up Premier Bet in 2002. The company was an early online bookmaker that took bets under the Asian handicap system.

A BBC article from the time describes how Bloom operated the company:

It is heartening to find that in this age of robot dogs, online everything and space stations, Mr Bloom works [the odds] out himself.

He has all the statistics to hand to help him work out what the handicap should be, but it basically boils down to him watching a huge amount of football and making a judgment based on what he sees.

In 2005, the business was sold to Interactive Gaming for £1.2 million. More success came riding the online poker wave of the 2000s. Two online poker sites he helped set up, Tribeca Tables and St Minver, sold in the mid-2000s. The deals were performance-linked and worth up to $204 million combined. It is not clear how much Bloom made from the sales.

As well as gambling at work, Bloom did it for pleasure. Bloom made a name for himself as a formidable, high-stakes poker player in the early 2000s. Bloom is listed by PokerNews.com as the 15th most successful live — as opposed to online — poker player, having won $3.3 million at tournaments to date. His form earned him the nickname "The Lizard" at the poker table — he must be cold-blooded to make such ice-cool decisions, people said.

Poker player and TV presenter Victoria Coren dubbed Bloom a "poker phenomenon," writing in The Guardian in 2010: "If tournament winnings (the flawed yet standard measure of poker success) were divided by number of tournaments played, the low-key Lizard would probably turn out to be the biggest winner in the world."

Bloom told The Times in 2011: "Poker gives you a good grounding in lots of things, including reading situations and reading people and making tough decisions. Those skills can be used in business and certainly in running a football club."

Bloom is incredibly unusual in that he plays at the highest level merely for fun, rather than professionally. He told Atherton that he inherited his love of gambling from his grandfather Harry Bloom, who owned greyhounds. He told Atherton: "He was a small-time gambler and probably a loser, as 99% of people are, but he loved it, and the losing never became out of control."

Bloom, who also runs marathons in his spare time, challenged US professional player Daniel Negreanu to a head-to-head game with a pot of $500,000 in 2005. After five hours, Bloom lost and "with his characteristic calm walked away," according to VegasInsider.com. He could afford to lose $500,000.

Bloom clearly took a shine to his poker nickname, as he used it for his next venture, Starlizard, set up in 2006.

Starlizard represented a shift away from taking bets into advising on them. The company is a consultancy that offers proprietary odds analysis to rich clients looking to make smart, high-stakes bets. These high-rollers then use Starlizard's internally generated odds to identify "value" bets — instances in which the retail bookmaking market has underestimated or overestimated a team. In these cases, the risk-reward ratio is swayed in the bettor's favour.

An employee of Starlizard's rival Smartodds told The Guardian in 2011: "We're not trying to say this is going to happen, we're trying to say this will happen with a certain probability. If our probabilities are better than those of the bookmakers, then, in the long run, our clients will win money."

Starlizard specialises in estimates tailored for the Asian handicap market — Bloom's favourite — by calculating what it sees as the most likely scoreline for any given football match. This is crucial, as the handicap rests on the favourite scoring a certain number of goals. This advice generates revenue of about £13.8 million annually for Starlizard in fees from clients.

Bloom is not listed as a director of Starlizard, but several of his key lieutenants are: Steven Edery, described by former employees as Bloom's right-hand man; Marc Sugarman, a former Citigroup equity analyst who is also on Brighton's board; and Adam Franks, a chartered accountant who knows Bloom from his Manchester University days and who is finance director for all of Bloom's businesses. Franks is also on the Brighton board.

Everyone Business Insider talked to within the industry and who had worked at the company said that despite not appearing on the paperwork, Bloom was in charge. It's unclear why Bloom is not a director. He even keeps an apartment in the same building as Starlizard's Camden offices, according to one former employee.

As well as being Starlizard's architect, Bloom is also the company's biggest client.

About the same time that Starlizard was created, Bloom established a gambling syndicate, a group of close associates who would pool their money together to make high-stakes sports bets.

By increasing the available pot of betting money, the group maximises potential winnings. Bloom runs the syndicate and is believed to provide the vast majority of the bankroll.

A 2014 article on the sports website Bleacher Report suggested that Starlizard accepted outside money if an investor could stump up at least £2 million. But former employees and industry insiders spoke of Starlizard and Bloom's syndicate as one and the same, and they say Starlizard spends most of its time dealing with Bloom's syndicate.

Starlizard's latest accounts, made up to June 2014, say the business is "diversifying revenue streams in order to reduce the risk of over-reliance on a particular client." In essence, the company is Bloom's gambling money manager. Starlizard and Bloom declined to comment on these specific claims.

While it's unclear why Bloom is not a director of Starlizard, it is clear why he set up the company.

Bloom acknowledged in Michael Atherton's 2006 book that he had "an addictive personality," saying: "At university I made myself a promise that I would become fiercely disciplined. I wanted to gamble because I enjoyed it and therefore I needed to do it properly in order to win. I didn't want to lose my money."

Starlizard allowed Bloom's syndicate to make cold, calculated decisions about where to stake cash, separating the decision-making from the money and making it as mathematical as possible — no gut feelings.

Starlizard helped pioneer a new, corporate approach to professional gambling, more closely resembling an investment bank or hedge fund than a bookmaker.

About 160 workers spend their days crunching statistics, building computer models, and doing huge deals on the other side of the world.

Some are fresh out of university, but ages range right up to the mid-40s. The mix of genders and races is diverse — all that matters is a razor-sharp understanding of the betting market and a head for statistics.

"It was a lot more professional and less laddish than working at a William Hill or a Bet365," a former employee says.

The company is split into four teams, each performing a specific role in the generation of odds. One generates data, another crunches that data into odds, a third decides which bets to take based on those odds, and a fourth places those bets on behalf of clients with bookmakers in Asia.

In the Camden office there are about 30 football researchers who generate internal data. They do this by watching matches and recording things like goal-scoring opportunities or shots on target.

A former employee says: "If a game was 0-0 but the home team had missed a penalty, the best scoreline to go back into a predictive model would be something like 0.8. If a team missed a penalty and had, say, two shots where they hit the woodwork, they probably deserved to win."

These researchers also make it their business to get as close as possible to the action, speaking to a network of contacts that includes journalists and league experts. Their aim is to get as much information on things like morale, form, team sheets, and training as possible.

A former employee told Business Insider: "Every aspect of football that you could think of was taken into consideration. I guess that's why they're so good at what they do. The weather, morale, anyone related to the club, [they] would be analysed under the microscope. It was pretty impressive."

The data generated by Starlizard's researchers is plugged into a highly complex statistical computer model, built by another team, the "quants." These are the computer whizzes you would usually find in investment banks.

These quants are based in a separate office, out in Exeter, and spend their days building and maintaining an algorithm that not only pulls together all of the data points, but also decides the right weighting for each.

Speaking about his gambling philosophy in general, Bloom told The Times in 2011: "A lot of otherwise good gamblers may read too much into injuries. Sometimes the odds can get too skewed because one or two players are out. When I analyse situations, I don't want to go overboard on one side."

The odds generated in Exeter are passed back to the Camden office, where a team of "selectors" reviews them. This smaller team — about 20 people — operates like the traders in a bank. It tries to identify mispriced bets in the retail market, based on the team's internal odds, and decide just how much to stake on behalf of Bloom and other clients. The computer model is tweaked nearly constantly, according to former employees, and it uses statistical models to predict the likelihood of every possible scoreline. It then churns out what it sees as the most accurate handicap for the match — Leicester at 1.18 against Aston Villa, for example.

These decisions are relayed to bet placers, the fourth team. As well as paying for access to Starlizard's proprietary odds, clients are paying for access to its black book of contacts in markets like China, Thailand, and Indonesia. Starlizard's odds are tailored to these markets, but it can be difficult to access Asian bookmakers unless you know the right people.

Bet placers operate like brokers, placing bets on behalf of clients. But the bet placers themselves work through a series of brokers in Asia, contacting them over the phone or using online messaging tools.

Working this way helps Starlizard obscure its presence in the market — the company is now well known in Asian betting, and knowledge of its ordering a bet would move the odds.

Jesper Søgaard, CEO of BettingExpert.com, a tip-sharing platform for amateur gamblers, says: "If they take a position, they will definitely move the entire market. They do as much as possible to not let others know about their position."

"I can't tell you any 'last weekend they had this position' — I don't know that. But what I can see is the market moved and that signals one of the big syndicates made a move. Especially on match day, you know it's the big boys playing."

Industry experts estimate there are up to 12 sizeable professional gambling syndicates worldwide, though exact figures are unclear. Some are even more clandestine than Starlizard, as they are involved in match-fixing.

This is an anathema to Starlizard's approach, which depends on as clean a game as possible to let the statistics come good. There is no suggestion of any legal wrongdoing in Starlizard's operations.

Starlizard will typically place bets as close to match day as possible to guard against any new information that could move against them — an injury to a key player in training, for example.

A Premier League match will create the most liquid pool of bets. Starlizard will try to bet at least £1 million on behalf of Bloom and any other clients. Placing bets of this size without skewing the odds would be near impossible in Europe, but in Asia — one of the most liquid gambling markets in the world — it can go undetected.

Starlizard must stake bets this big because its margins are razor thin — it's a volume game.
Starlizard doesn’t just follow major leagues like the Premier League. A former employee says, "Wherever there's football, they're betting on it."

(Bloom's syndicate and Starlizard don't place any bets on Brighton, given Bloom and other directors' roles at the club.)

Leagues as esoteric as Japan, Turkey, and Australia are closely studied to find value bets. Stringers in these markets will feed back information on things like form and likely team sheets.

When placing bets on smaller leagues, smaller sums must be wagered so as not to spook the market — £10,000 here, £20,000 there. Cricket is also bet on, though to a lesser extent.

The upshot of betting in less popular leagues is that the relative information void means it can often be easier to find an edge — bookies spend less time on the Romanian form tables than they would for La Liga.

The downside to this extensive approach, a former employee says, is that "someone has to be in the office betting on it." Starlizard's Camden headquarters are open 24/7, and it's not unusual for people to come in at 3 a.m. or 4 a.m. to watch a match going on halfway around the world.

That's because if things appear to be going the syndicate's way, Starlizard will make additional bets during the match on behalf of its client, doubling down to increase the potential winnings.

Weekends are a write-off for staff too, as that's when most football is played. It is frowned upon to be out of the office on Saturday and Sunday.

But there are significant upsides that make Starlizard's unusual hours worth it.

Bloom and other directors are not afraid to spend money to keep their staff happy and the offices are kitted out with the type of luxuries you'd find at Goldman Sachs or Google. Behind the Icework's smoked-glass windows there's a free gym with changing rooms, a steam room, and showers; a full kitchen offering free food; and a games room with pool tables and darts.

The company maintains a box at Chelsea's Stamford Bridge stadium, according to former employees, as well as boxes at other top Premier League clubs. Employees get to visit.

One former employee remembers the whole company being packed up on coaches and taken down to the Amex Stadium, the 30,000-capacity home ground of Bloom's Brighton.

After a stadium tour, Starlizard's first nine went up against a team of ex-pros drafted in for the day. Commentators were on hand, and Sky Sports' cameras, usually stationed for league games, were drafted into use. Everyone went home with a DVD of the day's action.

On quieter days in the office, the atmosphere is relaxed, with one former employee saying workers could leave to play sports in the middle of the day with no bother.

And management ensures there are regular treats to keep everyone sweet. One former employee described spending on staff parties as "obscene." 

Starlizard hires out exclusive bars and clubs in London in full — "They don't want any old riff raff turning up," one former employee joked. On these occasions, free drinks flow all night.

Camden, the birthplace of punk and home of Amy Winehouse, feels as if it has a pub on every corner, and former Starlizard workers say they would often socialise in the area after-hours.

One former Starlizard worker says: "There was a lot of drinking, socialising, and parties."

But these sorts of benefits pale in comparison to the biggest sweetener of all — the money.

Former Starlizard workers say the base pay was unspectacular, with those on the betting side earning something similar to what they would at a High Street bookmaker — £25,000 to £40,000, depending on the role. The quants who maintained the odds algorithm also made what they would in a similar role at a bank.

But after new arrivals pass their probation, former employees told Business Insider, they are called into finance director Adam Franks' office and offered what amounts to a golden ticket: a stake in Bloom's syndicate.

This is a once-in-a-lifetime opportunity to get a share of the winnings of one of the most, if not the most, successful sports bettor in the world.

Gambling winnings from Bloom's multimillion-pound pot are paid out to stakeholders generally twice a year, with payouts ranging from below £100 to upward of £500,000 every six months, according to former employees. The payouts depend on how big an employee's stake — dubbed "stars" — in the syndicate is.

Best of all, when Starlizard workers are invited to join the syndicate, they don't even have to put any money in; they just get a stake of the winnings as if it were a bonus. (And because it's gambling winnings, the money is tax-exempt under UK law.)

Most of the staff is in on the syndicate. A former employee says: "You could quite easily be getting £10,000 every six months — who would turn that down?"

If the free payouts sound too good to be true, however, it is because they are. As well as being in line for any winnings, those who opt in to the syndicate are also on the hook for any losses. If there is a losing run, the staff has to help top up the gambling pot.

But losses are rare. A former employee who was with the company for most of its 10-year existence recalled just one significant period of losses.

Yes, the syndicate may have a run of losing bets. But across the year, Starlizard almost always came out on top, and the irregularity of repayments meant people were shielded.

An employee at one of the High Street bookmakers who declined to be named told Business Insider, "They don't beat the market all the time, just enough times."

It's worth dwelling on that for a second — just one period of losses across almost 10 years. That's a stunning record. Gambling is a losing game in which the bookmaker always comes out on top. For ordinary bettors who put a few quid on a match, it's essentially a tax on people who don't understand the laws of probability.

But Bloom and his team have managed to build a statistical model that has allowed them to consistently beat the market for the best part of a decade.

As a result of these regular payouts, many employees stay at Starlizard for a long time. Stints of five years or more are not unusual.

A former employee adds: "Once you're in there, it's very hard to work anywhere else. The skill set for some of the jobs there is so specific to that industry. You have to start again almost if you want to work somewhere else."

Recruitment is relatively rare, and when new workers are brought in, they know little about what the company actually does. Interviewees are pulled through a network of recruiters who know little about the business — one ex-staffer remembers being quizzed about what Starlizard did by the recruiter who had referred him.

Vetting is extensive too, with interviewees forced to explain any résumé gaps and in-depth background checks.

Once in, employees are bound by a strict code of secrecy about what it is they do. They can't have Twitter profiles and are made to sign strict nondisclosure agreements. They can tell people whom they work for, though some are even reluctant to do that.

Starlizard enforces this level of secrecy because its advantage comes from keeping its value bets a secret — if word got out that it were recommending clients back, say, Swansea this weekend, it would move odds and the edge would be lost. The potential returns would diminish, making the risk not as attractive.

A former employee adds: "If Tony Bloom is making something a value bet this weekend, the likelihood is it'll be a value bet next weekend too." Leaks have the potential to scupper not just this week's work, but next week's too.

Bloom has good reason to fear leaks. In the early days of the syndicate, insiders were filling their pockets at the expense of the company.

Several former employees confirmed to Business Insider that early employees would front-run the syndicate, placing personal bets on teams when they knew Starlizard was making a value bet that week. That eroded the company's advantage by skewing the odds.

The problem was rife, with one former employee saying, "People treated it like MPs treated expenses." Starlizard declined to comment on this specific allegation.

To combat this problem, Starlizard barred employees from placing personal bets and broke up the business' operations to limit information sharing.

In the Camden offices, security is tight. Building passes will allow you to get into only certain floors, similar to an investment bank. That keeps different departments from sharing too much information.

And having the quants who maintain the algorithm based in a separate office in Exeter also silos knowledge. A former employee suggested only a handful of people knew how the whole thing worked.

As a result of the secrecy, it's hard to know exactly how much Starlizard bets on behalf of Bloom's syndicate across a year or how much is won. Starlizard employees in the syndicate know how much they get paid out personally and have an idea of bet sizes each week, but across the year it is harder to tell.

Former employees and industry figures I spoke with estimate that Bloom's syndicate makes profits of £20 million to £100 million depending on how the year went. Bloom and Starlizard declined to comment on these figures.

To win the amount of money that former workers say it does, the syndicate must wager huge sums each year. Former employees say the syndicate is looking for a return on money invested of just 1% to 3%. A bookmaker, by comparison, typically has a margin of 10% to 15%. BettingExpert.com's Søgaard says: "We'r
e talking turnover of more than several hundred million pounds a year."

It seems more likely that the syndicate's profits are closer to £20 million than £100 million. To make £100 million on a 3% margin the syndicate would have to be wagering £3.3 billion.

Still, the rumoured profitability has earned Bloom a reputation as the godfather of football gambling. Keith Sobey, who runs a London sports-betting academy, told The Wall Street Journal in 2010, "He's probably the most successful soccer bettor in the world." Everyone Business Insider talked to within the industry echoed this sentiment.

Bloom's net worth is unknown, but there is speculation he could be a billionaire. The Jewish Chronicle estimated his wealth at £50 million in 2009, but in the same year he loaned Brighton £80 million and was quoted by the same paper as saying: "This is all my money. It is not loans from banks, and it is not somebody else's money."

Whatever the exact figure, his wealth almost certainly runs into the hundreds of millions — at least.

Bloom is not the only professional sports bettor in Britain. And Starlizard is not the only gambling consultancy in the UK — not even the only one in North London.

Bloom's great rival is Matthew Benham, the owner of Brentford FC who founded Smartodds, another stats-based gambling consultancy.

Similarities between Bloom and Benham abound. Like Bloom, Benham is a former City trader. Like Bloom, Benham has adopted a hedge-fund-like model, paying computer whizzes to build algorithms that help him beat the market. And like Bloom, Benham owns his boyhood football club.

Despite the similarities between the two, Benham and Bloom are archrivals. The pair first crossed paths at Premier Bet, the online bookmaker Bloom founded. Benham worked for Bloom, but the pair had a falling out that left them bitter rivals.

Former Starlizard employees say the feud is discussed often in the office but little understood beyond rumour and gossip.

SmartOdds was founded in 2004, two years before Starlizard, and it is based in Highgate, North London — just down the road from Bloom's offices.

Benham took over Brentford in 2012 and is noted for following a "Moneyball-style" statistical approach to running the club, an approach that helped him win the Danish super league with his other football club, Midtjylland.

Brighton took on Brentford in the Championship last Friday. Bloom's team triumphed 3-0.

As Benham has risen alongside Bloom, the Asian market has opened up. New bookmakers like Pinnacle and SBO Bet, a former shirt sponsor of West Ham, have made it easier to access the market, while online literature has promoted a greater understanding of both how the market works and how to play in it.

"There's a lot more people who are professional sports bettors in European sports than there used to be," Bloom told Bloomberg last year.

The edge Bloom had from being the first to really understand the Asian handicap market has slowly been eroded.

So too has the advantage he gained from bringing statistics and computer modelling to the market. Others have cottoned on, and many more sports analytics consultancies have sprung up. As punters have got smarter, so too have bookmakers.

Former employees say as the years have worn on it has got harder and harder to beat the bookies, with razor-thin margins squeezed even more.

But the syndicate continues undeterred. Starlizard continues to sharpen its high-tech computer model daily and find its edge. Tony "The Lizard" Bloom, a gambler at heart, is unlikely to give up just because the odds are against him.

Friday 12 February 2016

Sinking Putts And Stocks

Very occasionally, hidden among a pile of detritus, there's an interesting article on the Yahoo! Finance page. I'm sure I have written on this topic before, but apparently it wasn't on this blog, (or at least I can't find it), so it was probably published elsewhere, but it starts with the known phenomenon of pro golfers being more likely to save par than to make a birdie, when putting from the same spot. Being, at least nominally, a Finance related page, the author Cass R Sunstein extends the example to the stock market:
Can professional golf help explain what is now happening with the stock market? I think that it can, because it offers a clue about an important source of this month’s market volatility: human psychology.
The best golfers make par on most holes. They also have plenty of chances to make a welcome birdie (one under par) or to avoid a dreaded bogey (one over par). To do either, they have to sink a putt.
A stroke is a stroke, so you might think that whether a pro makes a putt can’t possibly depend on whether the result would be making a birdie or avoiding a bogey. But you’d be wrong.

A study of over 1.6 million putts shows that professional golfers are significantly more likely to succeed in sinking a par putt than a birdie putt of equal distance and difficulty. Remarkable but true: If the average top golfer putted as well for birdie as he puts for par, he would make an additional $1.2 million a year.
Why do golfers do so much better when they are putting for par? The best explanation, coming from behavioural science, is that most people are “loss averse,” meaning that they dislike losses a lot more than they like equivalent gains.
A loss from the status quo is very painful, and so people will do a lot to avoid it. A gain is good, but it isn’t nearly as good as a loss is bad. Like the rest of us, professional golfers are affected by what John Maynard Keynes called “animal spirits”: the feelings of the primitive creatures who lie within us. Hating the prospect of losses, golfers focus intensely on avoiding those bogeys, and often succeed.
Which brings us to the stock market. Of course it’s true that the recent volatility, and the sharp declines, have a lot to do with real-world events, including slower growth in China and rapidly falling oil prices. But the fundamentals remain pretty solid, and the ultimate effects of such factors are at least partly a product of psychology.
Investors know that stocks go up and down, but losses loom much larger than gains, and when the market gets especially volatile it’s tempting to sell. Even if your portfolio ends up the same on March 15 as it was on February 15, the interim losses tempt many people to get out. And if it’s a terrible month, a lot of people will want to avoid more bogeys -- and scale back their holdings.
A closely related phenomenon is called “probability neglect.” When an outcome stirs strong emotions, people tend to neglect the likelihood that it will occur. If the prospect of a bad result gets the heart racing -- a plane crash, a terrible disease, a loss of 30 percent of your portfolio -- most people will take strong steps to avoid it. They will pay too little attention to a comforting thought, which is that worst-case scenarios usually don’t come to fruition.
Loss aversion and probability neglect operate at the individual level, but much of our behaviour is a product of social interactions, which multiply their effects. Even when the fundamentals are strong, making significant market declines unlikely, investors are affected by the actions of other investors. Like a bank run, a decline in stock prices creates its own momentum.
In the most extreme cases, what happens, and what we are now witnessing, is an “informational cascade,” in which investors attend to the signals given by the behaviour of other investors, even if their own information suggests that the other investors are wrong.
Informational cascades help fuel sell-offs. If many investors are perceived to be selling, there is a snowball effect, as the “should sell” signal gets louder, not because people have reliable information that selling really makes sense but simply because of the behaviour of others.
The good news is that in ordinary circumstances, investor cascades are halted. The smart money is aware of everything I have said here, and if the fundamentals really are strong, savvy investors start buying. They aren’t loss averse, they don’t neglect probability, and they spot opportunities when they see them. If there are enough of them, they can stop and eventually reverse dramatic movements driven by animal spirits.
History tells us that in the long-run, equity markets will do just fine. In the short-run, however, the prospect of bogeys can create a lot of havoc, especially if a lot of people decide that they want to get out of the game.

Rugger Rigours

Apparently James has not given up alcohol for Lent. How else to explain the rambling comment below which can only have been written after several beers, a bottle of wine, a few mixers and possibly a glass or two of port. I've had a few daft comments on this blog before, but this one has to be read to be believed. He writes, and I use the term 'writes' very loosely:

That post was awesome dude! Can you tell me how to get...
Sorry, I don't know what came over me.
I must point out that I am not in the habit of eating 8 plates of Singapore noodles but if it's offered to me then I will, with all due reluctance, eat it. I decided to leave the 20p that I was short-changed by with my host in lieu of a tip. A miser I am not!
Ash Wednesday - I knew it was. After all I laid on pancakes for the padres (the relations not baseballers) the day before. However, last night I could take it no longer and had a cider, a bar of chocolate and a "treat". So that's me going to hell and accounts for my eye problems.
Proper football - Of course, we all know that to be Rugby, it being closer to the game from which Rugby Football and Association Football evolved from. Indeed both codes initially prohibited the forward pass and one code still does, to this day. Early soccerists had to dribble towards their opponents goal and either shoot or pass back to another player.

Although why they had to salivate towards their opponents goal I am not sure but heroin and cocaine was freely available at any local Boots apothecary back then and you know what these soccerists are like when it comes to excess. Pretty much like any 8 plate chomping maths and computing grad with time on his hands.
Brain damage from manly versions of football - There was a video on YouTube (I can no longer find it) about a programme along the lines of "Who Hits Harder, Rugby or Football (not soccer)". They did some scientific tests and showed that with all the padding footballers hit A LOT harder than rugby players. Does that mean that rugby players hold themselves back because they are not as heavily padded? I suppose with a helmet on you are more likely to go in head first. At school I was taught to go in with the shoulder but I usually just grabbed people from another school around the neck (an oik is still an oik). Until someone did it to me and I have had a twisted neck for the past 30+ years. Still, I have all my errr... you know... departments... schools... FACULTIES! That's the word.
I should point out that James and myself had an earlier side conversation about the relative lack of comments on our respective blogs. James suggested that our blogs may be viewed as too 'highbrow', a claim that may be more applicable to James' blog than to this humble effort, for which the reason is more likely that I have a habit of reproducing the comment in a later post, and then, not always favourably, dissecting it.

Overall the ratio of comments to posts here (1.69) isn't too shabby, but I suspect this number has declined in recent years. Overall, 2,102 posts have generated 3,559 comments (excluding the spam nonsense which is discarded).

Before moving on, regarding James comment on rugby versus American Football, in my opinion the use of the thick padding and helmet in the latter results in more reckless tackling and hitting. Helmet to helmet collisions are particularly dangerous:
Former Carolina Panthers running back Eric Shelton sued the NFL in 2010, alleging that a helmet-to-helmet collision caused him a spinal cord injury that left him paralysed, and he was not appropriately compensated for his injuries.
In 2011, a Frostburg State player died from a helmet-to-helmet collision.
On October 17, 2011, a 16-year-old high school football player in Homer, New York died from bleeding in the brain suffered from a helmet-to-helmet collision
The relative lack of padding in rugby results in an emphasis on technique. I wouldn't say that rugby players 'hold themselves back' so much as play the game respecting the reality of physical limitations. Given the choice of tackling correctly and not getting hurt or tackling with the wrong technique and risking injury, there's quite an incentive to learning and playing the game 'correctly'.
Anyway, after eight plates of Singapore noodles, (a typical lunch for James is shown above), James is no doubt naturally well padded, so next time he steps out for the annual Posh School Old Boys v Plebby Oiks match, he'll be in good shape. Well, not good shape, as in good fitness shape, but at least prepared for the rigours of rugger.  

Thursday 11 February 2016

CTE - Countdown To Easter

Thanks to James for pointing out my rather embarrassing typo (now corrected) in the subject of yesterday's post. He even tried to give me an out by suggesting maybe I was employing a cunning play of words, but not on this occasion.

As for the post itself, the revelation that umpires have been involved in this way shouldn't be a huge surprise to anyone. It might sound harsh, but anyone betting blind on obscure events, in whatever sport, frankly deserves to lose their money. If I'm surprised at all, it's that there is actually enough money to be made from these events to justify the cost of bribing these officials. Where are all the losers coming from?

James - and I should point out here that I do have other readers (though perhaps not commenters) - commented on my Yobs post:

My trick vis-à-vis Liverpool was to go in June. Therefore, there was no danger of meeting any soccerist scally-wags.
However, I did get short-changed by the Chinese buffet across the road from Lime Street station. Mind you, I did have 8 plates from their "all you can eat buffet".

I suppose I shall now flop into bed and watch the Super Bowl. I am not as into it as I was during the 80s. It was the World of Sport's 15 minute highlights during the 70s that made me into a curious fan of the Steelers. Much to the annoyance of my school's gym master as we would often have a line of scrimmage, followed by a forward pass, during a game of rugger.
Good to know that James is doing everything he can to maintain his sylph-like figure.

As for the Super Bowl, I have to say that I am with James on this. When Channel 4 started covering the game in 1983, I would stay up for the whole thing for the next few years, but my enthusiasm for the sport has faded since then.

One reason is that the game just takes so long, and while it's not so bad when you're at home and have other distractions, watching a game live is frankly quite tedious (although the college version is much more fun).

Unlike its cousins, rugby or proper football, there is just no flow to the professional game. The frequent and lengthy stoppages are perfect for advertisers, and were also perfect for trading in the pre-court-siding era, but they don't do anything to make the game watch-able.

There's also the constant tweaking of a massive rule book, (at 86 pages, a simple game this is not), and there were no less than nine rule changes prior to the 2015 season.

The Onion reported on a few of the more unusual rules:
  • In order to meet with league requirements, the home team should have 36 balls for outdoor games and 24 for indoor games, all of which must be available for testing with a pressure gauge by the referee two hours prior to the start of the game. Jesus Christ
  • On a kickoff, the clock does not start until the ball has been legally touched by a player; if it is illegally touched, the player is sent to prison for life, although the clock still starts
  • There are 45 seconds between plays. Sounds simple enough, but as the clock ticks down, players have to shout out what each second is divided by three or face a 10-yard penalty
  • Players must catch the ball with the NFL logo right-side-up and facing the cameras in order for the reception to be ruled complete
  • Balls are to be spotted short for the Lions until a majority of officials on the field determine it's no longer funny, at which point the Lions automatically forfeit
  • The pylons and goalposts extend upward infinitely until they finally reach another universe where football is played sideways
  • Following a touchdown, players may spike, spin, or roll the football, though no rolls may contain a spinning motion and no spun balls may be rolled after the spin is completed; either results in a loss of touchdown.
  • By rule, the exact definition of pass interference shall forever remain as mysterious as the definition of love and elusive as the definition of beauty
It's also become clear in recent years just how much damage the game is doing to players physically. It's brutal, with Chronic traumatic encephalopathy (CTE) a major issue that could yet see the game's popularity decline significantly (to be replaced by rugby perhaps?) and with thousands of other players suffering life-changing physical injuries. The NFL reportedly reached a settlement in 2013 with around 4,500 former players (or their estates) with chronic traumatic encephalopathy.

A high profile, but by no means atypical, example of other non brain issues came just last week as the Super Bowl winning quarterback Peyton Manning was asked about injuries:
Denver Broncos quarterback Peyton Manning has had four neck surgeries, including spinal fusion surgery in 2011, he's had major knee surgery, and Wednesday he said doctors have already told him he will need hip replacement surgery "at some point.''
Manning revealed the health nugget Wednesday when he was asked a question about the revelation from Boston University researchers that Hall of Famer quarterback Ken Stabler suffered from chronic traumatic encephalopathy (CTE) and whether Manning had concerns about his health when he transitions into his post-NFL life.
If the Broncos move on from Peyton Manning after this season and the quarterback wants to continue his career, the Rams have discussed the possibility of bringing the future Hall of Famer aboard.
"Certainly when you have injuries, when you have surgeries, the doctors sometimes will mention to you, whether you ask him or not, you're probably headed for a, you know, a hip replacement at a certain time of your life,'' Manning said. "And I said 'Doc, I didn't ask you if I had to have a hip replacement, I didn't need to know that here at age 37, that's for sure.' And I look forward to that day when I'm 52 and have a hip replacement. You know, am I going to have some potential neck procedures down the road, I don't know the answer to that. The hip part was true, this doctor told me that. I've seen a lot of doctors and he was nice enough to share that information with me.''
It's also rather telling that the NFL are trying to fudge the numbers on how long the average NFL career is:
According to the NFL Players Association the average career length is about 3.3 years. The NFL claims that the average career is about 6 years (for players who make a club's opening day roster in their rookie season).
As this excellent Business Insider article concludes:
So yes, NFL. The average length of a career in your league six years... but only if you don't count anyone who is below average.
I'm not sure that's how averages work, but nice try, Number Fudging League.

Yes, other sports have their share of injuries, but in most they are incidental. American Football seems to share with boxing and other fighting sports that injuring opponents is part of the 'game'. Parents are less inclined to let their sons play the game than in years past - what used to be seen as a 'tough' game is now being seen in a more negative light:
That's exactly what Debra Pyka thought when she signed up her son, Joseph Chernach, for Pop Warner football in Wisconsin, then later in Michigan, when he was 11 years old, in 1997.
If only she knew then that her son would be dead at 25. Joseph hung himself in his mother's shed on June 7, 2012. His brain was later found to have severe CTE, a degenerative brain disease that has been linked to concussions in football. Joseph Chernach had played sports, including wrestling, pole vaulting and football most of his young life. But he spent almost four years playing Pop Warner football from ages 11 to 14.
"My son was the class comedian, loved school, always fun to be around," Pyka told me. "But we noticed after high school Joseph changed. He got depressed, angry, paranoid and withdrew from sports and his friends. We just didn't know why. After learning about CTE, I knew he had it even before we got the results. The symptoms were all there."
Finally, I am not a religious man, (my brain has yet to succumb to CTE, and is still capable of logical thought), I am told that yesterday was Ash Wednesday, the anniversary of Jesus's cremation presumably.

It marks the first day of Lent, a period when "many people, even non-churchgoers, will seek to give something up that they enjoy during Lent. Frequently this takes the shape of alcohol, chocolate or other ‘treats’."

Well that sounds like a whole lot of fun. After much deliberation, I've decided to give up watching any American Football. I have principles. If you choose to give up alcohol, chocolates or 'treats', you're a better man than me.

PS: I have it on good authority that James is planning on giving up his eighth buffet plate - not that seven visits is enough, as I also hear that James is busy working on a new food-stacking algorithm.   

Wednesday 10 February 2016

Official Tennis Fixes

Still betting on obscure tennis matches? From Sean Ingle of the Guardian comes this well written article explaining some of the court-siding and fixing problems in tennis in lay terms:


Two international tennis umpires have been secretly banned, while four others face being thrown out of the sport for life on charges of serious corruption, the Guardian can reveal.

Umpires from Kazakhstan, Turkey and Ukraine are among those alleged to have taken bribes from betting syndicates in exchange for manipulating live scores on the International Tennis Federation’s Futures Tour – which allowed crooked gamblers to place bets already knowing the outcome of the next point.

The Guardian has also learned that Kirill Parfenov, an umpire from Kazakhstan, was decertified for life in February 2015 for contacting another official on Facebook in an attempt to manipulate the scoring of matches. Yet the tennis authorities never publicly released details, alerting only a small number of tournament directors and national tennis federations.

The International Tennis Federation also kept quiet over the case of another umpire, Denis Pitner of Croatia, who was suspended for 12 months at the start of August 2015 for regularly logging on to a betting account from which bets were placed on tennis matches. The ITF has also never publicly acknowledged that four more officials are facing serious corruption charges, and only did so when prompted by this newspaper.

The Guardian’s investigation will raise fresh concerns about the extent of corruption in tennis and the lack of transparency at the ITF, the governing body of the sport. There are also questions over whether the ITF inadvertently created the conditions for corruption to thrive.

In 2012 it signed a lucrative five-year deal worth $70m with the data company Sportradar to distribute live scores from very small tournaments around the globe. That meant the bookmakers could provide odds on those matches, particularly on the lucrative in-play market, where odds shift as the games progress – and unscrupulous gamblers had a prime opportunity which they could ruthlessly exploit.

Under the terms of the Sportradar deal, umpires are asked to immediately update the scoreboard after each point using their official IBM tablets. This score is then transmitted around the world to live-score sites and bookmakers, allowing the latter to update their prices as the match proceeds.

However, the umpires are alleged to have deliberately delayed updating the scores for up to 60 seconds – allowing gamblers to place bets knowing what was going to happen next. In some cases, the Guardian has learned, umpires are alleged to have texted the gamblers directly before updating the score on their tablet computer.

In effect the umpires are accused of “courtsiding” – a practice among gamblers whereby observing events live can provide an edge before betting markets react to changing scores – and it meant that bets could be placed on the outcome of games and sets in the knowledge that the chances of them winning were much higher than the odds implied.

The ruse was carried out in ITF futures tournaments in eastern Europe, the lowest rung of professional tennis, where there was little or no television coverage or security, and the poorly paid or volunteer umpires were more susceptible to taking bribes.

The Guardian approached Richard Ings, a former professional umpire for seven years who was also a senior executive responsible for umpires at the Association of Tennis Professionals, who said the revelations were “deeply troubling”.

“Over a 15-year period I have been involved in professional tennis officiating both as a professional umpire and administrator of officiating for the ATP,” he said. “During that period I have seen tennis umpires breach the code for officials for relatively minor offences. But I have never before seen umpires breach it for tennis-integrity issues related to gambling on tennis and courtsiding.

“It is deeply troubling, but not at all surprising, that the risk to the integrity of tennis driven by gambling has expanded beyond players and their entourages to now include umpires and other tournament officials.”

In 2014 the French umpire Morgan Lamri, who worked on the Challenger and Futures tours, was banned for life after being found guilty of being in breach of four unspecified articles of the Tennis Integrity Unit’s rulebook. However this is the first time that so many umpires – those charged with protecting the integrity of the game – have either been banned or faced bans.

Senior figures inside the sport have told the Guardian they fear the allegations are more damaging than the recent more historical claims around match-fixing, for several reasons.

• It shows that corruption extends beyond players’ fixing matches and into those who are supposed to be the game’s arbiters.

• It also exposes the fault lines in tennis’s claims that is doing all it can to be transparent. In the past the names of players who have been banned for life have always been publicly released. Yet here the ITF stayed quiet until it was pressed by the Guardian.


• The revelations raise the question as to whether the ITF decided not to release that fact that Parfenov and Pitner had been suspended because it feared the embarrassment.

• It calls into question whether the ITF’s $14m-per-annum contract with Sportradar has acted as an inadvertent facilitator of corruption. By providing a live data stream from those events most vulnerable to corruption due to small prize pools, a lesser degree of oversight, and negligible media attention, did it help corruption thrive?
In a statement, the ITF said that it could not comment further on the four officials are who currently suspended pending the completion of ongoing investigations.

“In order to ensure no prejudice of any future hearing we cannot publicly disclose the nature or detail of those investigations,” it added. “Should any official be found guilty of an offence, it will be announced publicly. The ITF code of conduct for officials was amended in December 2015 to include public reporting of officiating sanctions from 2016 onwards.

It also insisted that the Sportradar deal had helped the game expose corruption, not fuel it.

It added: “Our deal with Sportradar, like those in place with ATP and WTA, by creating official, accurate and immediate data, acts as a deterrent to efforts by anyone trying to conduct illegal sports betting and/or unauthorised use of data for non-legal purposes.”

The revelations have also renewed the spotlight on a sport stung by claims at the Australian Open that players on the men’s main ATP tour have fixed matches.

Partly due to the explosion in the number of events that can be gambled on during play, the number of suspicious incidents flagged up by bookmakers has risen sharply in the past three years.

Figures from the European Sports Security Association, a trade body that represents 18 bookmakers including William Hill and Ladbrokes, show that 49 suspicious gambling alerts were raised about tennis in the first nine months of 2015. In contrast, only 16 alerts were raised about other sports over the same period.

The world No2 Andy Murray has already urged the game’s authorities to be more “proactive” – warning them that “as a player, you just want to be made aware of everything that’s going on. I think we deserve to know everything that’s out there”.

Senior sources within the sports integrity community believe that the introduction of the Tennis Integrity Unit (TIU) in 2008 has helped stem the flow of new cases at the very top of the game. But they fear the TIU, which is supposed to be the sport’s watchdog, does not have the resources or power to tackle widespread abuse at the lower rungs of the tennis ladder.

During the Australian Open a combined statement from the ATP, WTA, ITF and heads of all four grand slam events, announced an independent review into the TIU “aimed at further safeguarding the integrity of the game”.

That review, headed by Adam Lewis QC, will also address issues of transparency and resourcing at the TIU and how to extend the scope of tennis’s anti-corruption education programmes.

As the TIU board chairman, Philip Brook, admitted last month: “It is vital we repair the damage and do so quickly. We are determined to do anything we need to remove corruption from our sport.”

How the scam worked

• Tennis umpires at low-level professional events often input scores manually on to an IBM tablet. The scores are thereby transmitted to the ITF’s data partner, Sportradar. Bookmakers access this scoring data from Sportradar to service their in-play, or live betting markets, where bets are placed and often cashed out while the game is in progress.

• The four umpires who have been suspended are alleged to have deliberately delayed the inputting of these scores, thereby giving gamblers, some of whom may have been present at courtside, 30 seconds to a minute of advance warning before the betting odds moved in response to the updated score. In some cases, the umpires were texting the score directly to gamblers before it had been officially updated.

• Gamblers could therefore manipulate and take advantage of minor shifts in the in-play market through knowing the scores in advance of the odds shifting to reflect them.

Sunday 7 February 2016

Yobs, Double-Pushes and Tangled Webbs

James, author of the much praised Programming for Betfair, corrects me on his recent computer upgrade project which was actually:

Windows 10, old boy, not IE 10.
To back-up your argument that the bookies now use "wisdom of crowds" with which to set their initial odds...
I did some consultancy two years ago for a company who, shall we say, used to send out teams of agents to collect coupons with your columns of x's on them. They have since gone online and wanted to put out a new product, which has an innovative back-end (NDA signed), now available on their website.
Wanting me to do some risk analysis for them, I was tasked with a) predicting the outcome of football games, and b) setting initial odds so that they could seed their markets to make them look popular (as Betfair, initially did) and in such a way that they did not lose too much of that seed money.
I had the solution pretty much worked out on the train home and emailed it to them the next day.
a) Forget it. Use wisdom of the crowds (i.e. Betfair) to determine odds prior to going in-play. Their markets shut when a match goes in-play.
b) Create a Dutch book and spread your seed money accordingly. So long as you Dutch properly then you won't lose any seed money as you are winning back your own money.
I think I made them look a bit stupid with such an obvious solution and they were initially reluctant to pay but I got my fee in the end.
Any way, it was a nice day in Liverpool. Oops!
The story has some credibility, until the "nice day in Liverpool" reference, at which point the whole tale unravels as such an event is surely not plausible. My 1977 visit there was somewhat traumatic, and has left an indelible impression of the city, or at least the red part of it. After being attacked by a large group of 'fans' shortly after exiting the coach, the game was followed by a night in the car park at a Liverpool hospital while a 12 year old had his injuries checked out, before it was determined he had a fractured skull and would be detained. The coach eventually arrived back in London around 6am. Credit to the group of Everton fans (they were hosting Stoke City at the same time - FA Cup weekend) who 'rescued' him and brought him back to the waiting Palace coach.

Back to James' comment, and for a time in my life, I was one of the afore-mentioned 'agents' collecting coupons on a Wednesday night, earning a little extra pocket money.

Having mentioned that "the crowd" usually does a pretty good job at pricing up markets and by extension determining handicap lines, the BLUnders system hit a rare double-push the other day. The San Antonio Spurs v New Orleans Pelicans game lines were -13 on the handicap and 207 on the total points.
As you can see from the above screenshot, both lines were hit.

Somewhat related to the topic of current markets being the result of "an equilibrium price that is formed by matching the supply and demand" rather than the previous process of bookies setting the odds, an evolutionary adaptation necessary for the traditional bookmakers to survive, I keep meaning to link you to this video from Peter Webb about the failure of exchanges to take over the betting world. Peter may have missed the point about the 1% a few weeks ago, but I believe he has it right on the topic of exchanges. They are a far better way of betting, and the fact that they appear to have lost their direction, hopefully a temporary setback, and the possible reasons for this, is an interesting one.  

Saturday 6 February 2016

Sub-Optimal Staking Exposed

In reference to my earlier post in support of variable staking, James, currently recovering from the stress of installing IE10, had this to say:

I agree entirely. If the other side of the trade has negative expectation then you have the edge, regardless of stake. Edge and stake are not related in that respect.
Poor money management affects the variance in profits and level stakes is just as likely to destroy your bankroll by yielding less profit on a winning bet as too big a loss on a Kelly bet.
The optimal bet is to chase your winnings and not your losses. If you consistently bet less than you should then you are building a smaller cushion for when the inevitable losing streak comes.
Kelly betting conforms to "chase your winnings not your losses" by sizing the bet to your current win/loss streak. If you are in a losing streak then Kelly will make smaller bets relative to the decreasing bankroll. 
Webbo, who of course initiated this debate with his ill-considered Tweet responded with:
Largely agree with you of course and I'll just clarify a couple of the points.

By letting the bookie do the work, I mean let them decide on the exact price. It's much easier to know that a price is too big than to know by exactly how much each time.
I have to interrupt here, because I don’t understand this statement. For a start, let’s be clear that these days the bookie doesn’t ‘do the work’
...it is us as bettors collectively whose are those that set the odds, not the bookmakers.
Individual odds, or prices, are nothing more than an equilibrium price that is formed by matching the supply and demand [for bets] from punters. Betfair Exchange is the clearest example of a true market. On Betfair, by introducing the bets they are willing to place/lay and at corresponding odds (“supply and demand”), a market price is formed. Other bookies or betting houses are not markets as such. They determine their own odds. But deep down they cannot escape the fact that they are part of a global market and that they have to adjust their prices in response to evolving events. Bookmakers have different strategies to set their odds, but at no point can they go against the market.
Thus the market decides on the ‘exact’ price and the traditional bookies pretty much fall into line, except by offering worse odds than the exchanges and the new model Sportsbooks (e.g.Pinnacle Sports), and hoping to get some interest from non-price sensitive long-term losers and a few quid in their FOBTs. (They may not yet be long-term losers, but traditional bookies will soon limit you if it appears they won’t be).

As for the statement that “It's much easier to know that a price is too big than to know by exactly how much each time”, I think if you are serious about investing rather than looking for entertainment, you really need to have a price in mind before you can say an offer is “too big”. When you read a tipster saying that a selection “offers some value”, you can be pretty sure they really have no idea what their 'true' price is. It’s one of my gripes about tipsters that they will tip a selection at say 2.5, but not tell you at what price that selection ceases to be value. Not to mention that different punters have different thresholds for when it is worth their while getting involved. 0.1% value is one thing, 5% another, and 25% another altogether (in the case of the latter, you're probably missing some information).
The only way you can determine that you have value is by calculating your own prices, and once you have done this, it’s easy to know how large any edge is.

Webbo continues:
I'm sure bookies do hate all winners but most of those who use a variable method are less likely to be able to price up markets as accurately as the bookies on a regular basis. Of course you could be varying based on odds ranges only but this is then not much different to level stakes if that’s the case. They will also likely see more volatile swings of which they are less likely to be able to handle. I’m guessing a bit here as I’ve never been a bookie and I’m not saying that I’m correct, but after an early period of success I think they’d be more confident that the variable staker will come a cropper than the level stakes bettor.

Of course there will be some exceptions but very few people will be able to show a set of results that have been improved by variable staking.
Again, the bookies do not price up markets, we punters collectively do, and we generally do a pretty good job. 

I would argue that bookmakers are actually more concerned about variable stakers. If you consider the example of the casinos and card-counting in Blackjack, the success of the strategy hinges completely on varying stakes, and this is the behaviour that leads to one being shown the door. Level stakes bettors are free to play all night, and to come back next day. 

When someone bets to level stakes, it reveals that they don't know the size of their edge, and by extension if they even have an edge. The variable staker should thus be seen as more sophisticated and knowledgeable, i.e. just the type of client bookmakers don't want.  
  
What this sentence means - "They will also likely see more volatile swings of which they are less likely to be able to handle" – I have no idea! I thought Deepak Chopra was commenting for a moment!

Webbo followed up later with a further comment:
We are all singing from the same hymn sheet here by the way. We all know that Kelly is the optimal approach in theory but 90% (maybe more?) of punters won't be able to use it or any variable staking to their advantage. The twitter statement was merely intended as advice for most people.
It was poor or misguided advice at best. I’m not sure we do all know about Kelly, but regardless, the Tweet made no reference to Kelly anyway. It simply stated that "variable staking can leave you exposed with just one bad bet" which is meaningless. As I said previously, it’s idiotic staking that can hurt you, not variable.

Webbo changes the topic, continuing:
Anyway here's a more interesting question for you related to this. How low do you let your bank get using Kelly if you have an edge? How low would you expect it to go?

In your mathematically perfect world you would utilise as much of your edge and bank as possible but what if these sees you lose up to 40% of your funds along the way?
Our simulations over more than 10 year’s worth of data show that if we make the most of our edge in all markets we could see our bank more than 40% lower than the previous bank high and it can take up to a year to see a new high.

This does return by far the most profits in the long term but when the time comes for us to drop such a percentage, how do we know there's not some external forces that have come into play or that the edge has diminished? At the same time we want to maximize our edge as quickly as possible.

Appreciate your thoughts.
Again, I am not advocating anyone use full Kelly for sports investing - there are a number of problems, not least that to use Kelly it must be assumed that the investor is able to maintain his edge indefinitely - no easy task. As I wrote previously:
The curse of Kelly (or more conservatively fractional Kelly) is that accurately calculating your edge on a sporting event is very difficult. For that reason, sticking to a maximum bet size of 4% of your bank will serve most people well.
In the unlikely scenario that you are able to maintain an edge indefinitely, and accurately measure it, choose your bank size, apply full Kelly and go with it knowing that you are betting optimally.

The question of "how low do you let your bank go" is silly. Why choose a bank size if you are not committed to it? I like the inimitable Ian Erskine's recent take on this:
After the July seminars a guy asked me to sort his bank of 10k out which I did. He then lost £1856 which is less than 20% of his bank. He sent me a load of abuse and stated he was no longer continuing.
I replied and stated that in that case he did not have a 10k bank clearly he had a less than 2k bank and had he informed me of that I would have started him on much smaller stakes relative to a 2k bank. I then got another load of abuse. That is how so many of you operate. You do not commit, you fanny around trying to do everything to £2, £5 stakes and wonder why you're not rich. You give up as soon as a bad run comes, you stake money you don’t want to or can’t afford to lose, so have a meltdown as every bet leaves you teetering on the cliff. If you just stopped took a breath and set yourself up properly then you would see it really is not that difficult and once set you control your emotions and operate the same daily.