Back To Reality

Back from the Mayan Riviera and I bring great news. 

The world will NOT be ending on December 21st 2012 as supposedly prophesised by the Mayans. As is often the case with these sensational stories, the truth is a little more mundane in that the date simply marks the end of one of the Mayan 52 year cycles, and has no more to do with the end of the world than any other artificial date such as the end of a century in the Gregorian calendar (which incidentally was 31.12.2000 and not 31.12.1999 as most seemed to think). Why any fool (and there are many apparently) would seriously attribute such irrational predictive powers to an ancient people is a little disturbing, but anyway, relieved that I may have more than 30 months left of life, and relaxed and rested, and hopefully without missing too many value opportunities, my thoughts turn once again to the upcoming football season. Not that it is really upcoming to be accurate, since the Scottish season is already underway with the Challenge Cup, and my spreadsheet updating begins again. By way of introduction, here was a comment I read on another blog:

"In football the best trading strategy is to have the current score in your favour. You'll never get a better tip than that"

Is this really true? It is certainly the philosophy for football betting that I have mostly followed myself over the past four years but it has not proven to be a profitable one. In fact, as I will explain later, I think that because so many people think this way, the opposite is true and the best strategy is in fact to oppose the current score.

When trading football in-play, there are two main approaches - you either bet hoping for goals, or you bet hoping for no goals. My natural psychological preference is to be on the current winning position and hope it stays that way, but I am not convinced this is the best approach. It certainly is directly opposed to my approach to other sports where I more often find value by opposing the current leader, but in the end it all comes down to which approach is most likely to offer value. 

I have learned (the hard way) that in football, having the current score in your favour is less likely to offer value than betting that the score will change. Why should this be? Most likely it is because more people are like me and want to be in a winning position (everybody loves a winner at any price) and thus the price for this is driven lower than it should be, and as a result, any value is therefore found on the side of change. I have to admit that I find it most unnatural to make a bet knowing that the price will in all likelihood drift away from me, but this is the position that I believe is most likely to offer value.

The Under / Over 2.5 goals market is a perfect example. When a game started, for years I, and my fellow 'feelgood' gamblers, would back the Under, looking to lay off 'at some point', simply because it was comforting to me to see the price immediately move in my favour. Supporting the price on the other side were the 'contrarian' gamblers. But where is the 'at some point'? The only point where it is logical to lay the bet off is if the price at that time reflects value, and the football markets are so efficient that the only time when value might exist is immediately after a goal has been scored, before the prices have settled down again, but of course for this bet, waiting for a goal is rather too late.

One approach is to lay the bet off after a certain number of minutes, or a certain number of ticks, but as anyone who has tried this over a number of games will tell you, it simply doesn't work. Indeed, there is no reason why it should because for it to work, either (or both) the back or the lay price needs to be value, and the market is too efficient to offer this. The typical scenario is that you win a few bets in a row, then you have two losers close together which wipe out your profits, erase any confidence and you give up. My personal experience is that if I layed off at 10 minutes, the game would usually end as an Under, or I would jump back in again after x minutes convinced that the game was certain to end 0-0, only to see three goals in quick succession! Then if I didn't lay the bet off, there would always be a goal within a minute of my starting to think about laying the bet off. Or so it seemed.

So having said all that, I am not at all convinced that in-play trading in football offers much value at all. The market is simply too efficient, and the only time when value might exist is immediately after a goal before the prices have settled down again.

Or there's always the old fashioned pre-game punt, and let it run method. This will be my primary focus this football season, and more on that later.

Summer Holiday

There will be a lack of incisive, sharp, brilliant and witty posts for a few days as I am off on my holidays to the Mayan Riviera. A thought I was pondering earlier concerns the Mayan gods, so I added a poll for while I am away. I expect at least 10 votes before I return. Happy punting.

Get Smart

The next time someone says they can't master exchange betting, you can tell them that a new study has shown that the brain is like a muscle, and will get more powerful with the right kind of daily exercises. This rather lays to rest the traditional view that intelligence is stable: either you have it or you don't, and nothing you do can change that.

Scientists measured the IQs of 70 volunteers, then put them through a brain "boot camp," consisting of memory tasks that increased in difficulty as the subjects became more proficient. After 19 days of exercising their memories and solving puzzles, the volunteers' IQs were retested, and the results were striking. Every single participant made significant gains in "fluid intelligence," the ability to solve problems, use abstract reasoning, and be quick on their feet. The longer people trained, the higher their scores rose. "Intelligence has always been considered principally an immutable inherited trait," study co-author Susanne Jaeggi says. This study "definitely challenges old opinions."

It reminded me of my first exposure to Betfair, confusing at first, but stick with it and specialise and I think anyone can get the hang of it. It took me a lot more than 19 days to understand it though.

Humans 0 Rats 1

So why did do many people have trouble working out such a seemingly simple problem? (See previous post). It appears that "our brains are just not wired to do probability very well."

Researchers did an experiment called 'probability guessing'. Subjects were shown a series of cards or lights which can be one of two colours, say red and green. Things are arranged so that the colours appeared with different probabilities, but otherwise with no pattern. For example, red might occur twice as often as green.(Think here perhaps of the over/under 2.5 goals market in football). The task of the subject, after watching for a while, is to predict whether each new member of a sequence will be red or green.

There are two basic approaches. One is to always guess the colour that you notice occurs more frequently. The other is to match your proportion of red and green guesses to the proportion of red and green you have observed in the past. Rats and other non-human animals favour the first approach whereas humans usually adapt the second strategy, trying to deduce a pattern.

Thus humans are outperformed by rats!

To humans, random events often look like non-random events. Sound familiar?

Monty Hall

No doubt most readers are familiar with the Monty Hall problem, so bear with me while I repeat it. The reason is that I was reading an article the other day about how such a simple exercise in probability was so misunderstood by so many people. The problem is this:

Suppose you are on a game show, and you are given the choice of three doors: Behind one door is a car, behind the others, goats. You pick a door, say No. 1 and the host, who knows what's behind the other doors, opens another door, say No. 3, to reveal a goat. He then asks you if you want to switch your choice from door No. 1 to door No. 2. Is it to your advantage to switch?

Apparently when this question was posed in an American magazine in 1990, and the columnist gave the correct answer, she received 10,000 letters, and her readers "acted as if she'd advocated ceding California back to Mexico".

Surveys show that 13% of Americans don't believe that plants create oxygen, 24% don't believe that light travels faster than sound, and 35% believe that you can make milk radioactive by boiling it, but on the Monty Hall question some 92% believed the published answer was wrong.

Apparently close to 1,000 Ph.D.s wrote in, many of them mathematics professors, who seemed especially irate. One professor wrote "How many irate mathematicians are needed to change your mind?" Someone from the US Army Research Institute remarked "If all those Ph.D.s are wrong the country would be in serious trouble."

But the answer was correct.

One leading mathematician of the 20th century, Paul Erdos, said "That's impossible" and then when presented with a formal mathematical proof of the correct answer, he still didn't believe it "and grew angry." Only after a colleague arranged for a computer simulation with hundreds of trials did Erdos concede that he was wrong.

Part two on probability decisions later.

ASHE Wednesday

I do like my stats, and today there is a report out from the Office for National Statistics' Annual Survey of Hours and Earnings (ASHE).

According to ASHE, the "mean" gross annual earnings across all employee jobs in 2008 came to £26,020. The figure for full-time employees is £31,323. The "median" gross annual earnings for all employee jobs was the more modest figure of £20,801, and for just full-time employees, the median rises to £25,123.

A gross annual salary of £31,759 - measured across all jobs - puts you into the top 25%.

£44,881 puts you into the top 10%, and a gross annual salary of £58,917 gets you into the top 5%.

For the top 1% you need to be making a gross salary of £118,027.

Of course, in reality it's not as straightforward as this because the ASHE is a sample of 1% of people who pay tax via PAYE. It doesn't include the self-employed - businessmen, contractors etc - who make up the ranks of the really wealthy, but for someone like me, it does provide some general goals.

Of course, these are gross figures, so when comparing them to net profits from gambling, the zero tax on the latter needs to be taken into account as does the fact that I regularly spend more than 35 hours a week trading! Not next week though.

So, very approximately, taking 33% off for taxes, NI etc, you would need to make the following per day from betting to be at the following levels:

Median: £38
Mean: £48
Top 25%: £58
Top 10%: £82
Top 5%: £108
Top 1%: £216

Unlucky Sunday 12th

Well, all in all quite an eventful weekend. Saturday saw my third best day of the year with a nice basketball trade proving very profitable, but then Sunday saw my second worst day of the year with a not so profitable baseball trade.

What is it about Sunday the 12ths? The previous Sunday the 12th (April) was my number one worst day of the year, and now this. In April it was Easter Sunday, so I put the bad luck down to the fact that Jesus really is the son of god, and he was upset at me for not going to church to worship him. Fortunately there are no more this year. Anyone else have lucky / unlucky dates? I have to go back to November 2006 to find a profitable Sunday 12th.

A big win followed by a big loss will help with the Premium Charge though, but what a difference it would make to my morale had the bad day come first. Value doesn't always equate to winners, so I'm not too concerned, and the month is still nicely in profit. Just not quite as nicely as it was for a few hours with some of the averages hitting a high.

Mind you, I wasn't expecting too much from July and with two quiet weeks ahead, that expectation is unlikely to change. It's the baseball all-star game next week, and the week after that I am off to the Mayan Riviera for a well deserved (if I do say so myself) break. Perhaps some of the Mayan gods can help with my investing.

Betting Quote

The race is not always to the swift, nor the battle to the strong - but that's the way to bet - Ring Lardner

Laying Alan Keyte's Tips

One of the longer running threads on the Betfair forum appears to be coming to an end. The original poster, unionjack, started the thread on October 26th 2007 with this:

I am starting a new thread.

I wish to make £2 on every tip of Alan Keyte's by laying them instead of backing and using a Recovery Martingale Staking system to keep on target.

You can see his tips every day on the naps section of the Racing Post.

Most of his tips are very short prices, so this should give interesting reading.

I am paper trading for the time being, using a bank of £1000, but if it works, I shall be going for the Real McCoy, especially with the season changing so soon and turning into good laying weather

Today
Wolverhampton 9.20
BOZ
Lay £2 @1.86 (was 1.77 this morning, so a wee drift looks good)

Well, he was right in that it did make interesting reading. Alan Keyte (Western Mail) was selected because although he picks a fair number of winners, most of them are at short prices so for a laying system this is ideal.

Almost two years later, he’s calling it quits after a tough few days where Alan Keyte just keeps picking winners. Short prices mind you, but with a recovery staking system, stakes can get uncomfortably high rather quickly with a losing run, and the current sequence dates back over two weeks. Starting with a bank of £1,000 he made a profit of £2,967.38 which is a profit not to be sniffed at. I hope someone continues updating the thread with a ‘what-if’ total, because but we’ll see.

The forum is littered with a host of rather pathetic and ill-considered ideas, threads that typically last no longer than a mayfly, but this one was great – a simple idea that attracted almost 4,000 posts. Not many threads can boast that kind of total. Simple, because if a tipster regularly loses money over a season (and most do) then surely the way to a golden future is to lay those selections. Perhaps it’s not quite as easy as it seems but well done unionjack for the idea, for sticking with it, and for knowing when to quit. He could be quite amusing too, which is a pleasant change. It would be interesting to hear Alan Keyte's take on it all too.

Recklessness, Weetabix and eBay

I am noticing a certain gung-ho approach creeping in to my investing of late. For the last two Sundays, I have lumped a few hundred on the apparent winner of the weekend's golf tournament at short prices and both times the winner delivered. I have done these using my BETDAQ account, for three reasons.

One is that since I am not actively trading the event, then there is no point getting a 20% reduction on any winnings.

The second reason is that I've noticed the BETDAQ prices tend to lag slightly behind those on Betfair, so it is often possible to find value after a price change on Betfair.

The third reason is that whereas I treat my Betfair account as an investment / trading account, and take it rather seriously, I have the dangerous attitude that my BETDAQ account is more of a 'play account' where riskier (i.e. not such good value) investments can be made without fear of messing up my 'real' numbers. This could be a good thing because I tend to be over cautious sometimes, but it could be a bad thing if my assessment of value is wrong. (See previous post).

Of course losing a few hundred from any account is not something I want to happen though, and if I do, then I should really want to lose it from my Betfair account and help out with the Premium Charge. Speaking of which, another hit today, though not as bad as recent weeks, which must mean the summer is here.

Speaking of BETDAQ again, I do find their interface somewhat clunky to use. I have another account with Weetabix (WBX), and their screens and navigation are just as good as Betfair's, but once again, the problem here is the lack of liquidity. Handy for some of my outright punting though.

One day a viable challenge to Betfair will appear. I was reading at the weekend about how once invincible eBay are now facing a serious challenge from Amazon's auction site, and are seeing profits down for the first time. I can see some parallels here - both eBay and Betfair had first-mover advantage, and while I don't know the details, it appears that eBay have managed to upset customers with their pricing strategy and that customers are now 'flocking to Amazon in droves' where word is spreading that the 'auction experience is preferable'. Once tipping point is reached, it will be very hard for eBay to win those customers back. And so far as I know, eBay haven't even started making their most successful x% customers pay a far higher commission rate for their services.

All things being equal, customers stay with the brand they are familiar with, but once they become disenchanted, they're gone. A lesson here for Betfair I think. Abolishing the Premium Charge and refunding (with interest) all Premium Charges taken would go a long way to making this customer very loyal. Actually, I would even forgo the interest right now, since it is pitiful.

Value

There is some complete nonsense written on the Betfair forum. The subject of value comes up regularly with people who really should know better arguing that value in sports betting doesn't exist and that there are other ways to make money on Betfair than by finding value. This reminds me of the Victor Meldrew quote "What language are you talking in now? It appears to be bollocks".

The simple fact of the matter is that the only way to consistently win is by making bets that are value. If you are a long-term winner, then you are finding value (even if you are blissfully unaware of it). If you are a long-term loser, then your bets are not value. It is that simple.

It seems that one of the problems is that whereas the value on a coin toss can be easily determined (2.1 is value, 1.9 is not), the value on a sporting contest is not, but just because the true probability is hard to determine, doesn't mean that a correct price doesn't exist. Possibly one of the easier sports to calculate probability on is snooker or darts. Conditions are always the same, and if Player A has beaten Player B in 80% (of a large number) of frames ever played, it's reasonable to assume that Player A will win a best-of-three contest 89.6% of the time, and is thus value at 1.12 (excluding commission). Yes, it is quite possible, probable even, that there are other factors at play - recent form, emotional state, physical factors, but this is a reasonable starting point. The true test of whether or not you are finding value is whether or not in the long-term your bets on Player A at 1.12 are profitable. If you are getting value, then they will be; if you are not getting value, then they won't be.

Other sports are that much harder to evaluate. Of individual sports, tennis has its different surfaces to consider, golf has over 100 possible winners, motor sports are subject to accidents and mechanical malfunctions, but if you favour these for betting, it holds true that if your bets are value, then you will make money and if they are not value, you will lose. There is no getting away from this.

With team sports, personnel or tactical changes mean that one never has the luxury of comparing say Chelsea with Liverpool over 100 games with the same line-ups and be able to compute value. There is always a subjective element when determining the correct price, (what weighting do you put on current form, how 'current' IS form - 6 games, 20 games?, injuries, suspensions, recent away games in Armenia etc.) but it is possible to give yourself a starting point by rating teams. Once again, if you are finding value, then you will make money, if you are not, then you'll be losing money.

One argument against the need to find value is that so long as you can back at price x, and lay at price y (x-0.01), then you will make money. This is true, but the problem is that I think you will soon find that if your back at price x is not value, then you will have trouble selling the bet at price y, which is even poorer value.

I haven't mentioned horse-racing. As regular readers will know, I think this is undoubtedly the hardest sport to consistently make money from unless you have access to inside information. An outsider like me has almost no chance of finding value consistently, and even if I could be bothered with rating hundreds of horses, there is no guarantee that the horse would be run cleanly anyway. Unless you're an insider, there will always be others that know more than you do about any particular race.

Betfair Forever

A poster on the Betfair forum wrote that "this [Betfair] is the ONLY betting exchange and forever will be so"

Nonsense. There's no doubt that Betfair are the dominant company in the Exchange betting business right now, but history is full of examples of companies who got 'too big for their boots' and ended up pale shadows of their former selves. It's not long ago that eBay seemed invincible in the online auction market but profits are now down as buyers and sellers are moving to Amazon in droves, and word spreads that the Amazon experience is better. The advent of Craigslist has also reduced eBay's business at the lower end of the market.

Betfair's position as market leader could similarly disappear very rapidly if, or more likely when, a viable competitor appears on the scene and once tipping point is reached, they will find themselves left high and dry. People have no choice right now but to live with the Premium Charge, the underhand introduction of cross-matching, and the Arcade but they have memories, and the goodwill in many cases is gone.

We Are Not Amused

To the surprise of many, Andy Murray lost in the semi-final today. At 1.35 pre-game, against an experienced player, this was never value. I layed off my pre-tournament back of Murray one round too early, but I'm not complaining. As I have said before, in-play tennis betting just isn't for me, but I seem to be making more pre-game or pre-tournament bets these days. The Queen will be disappointed though.

I was reading the forum before the game, and an astonishing number of people felt that 1.35 was huge value, and that Roddick had no chance in reply to someone who suggested 1.35 was a value lay. Good for the original poster. Funny how all the 'experts' go very quiet when they are shown to be wrong.

Speaking of value, the New York Yankees were 1.34 pre-game yesterday, in a game where the best team wins approximately 60% of their games over a season. Value? Not for me, and it made me a nice profit when the Yankees lost the game to the Seattle Mariners. Today they play the Toronto Blue Jays and pre-game were 1.54. Again, not value in my opinion. Currently the game is tied after four innings, and could go either way. I'm on the value side of this one whatever happens.

Manchester United have signed Michael Owen! Now there's a surprise. Not that he left Newcastle, but that Manchester United wanted him. A great player in his uninjured prime, but nothing special these days.

Another surprise transfer is Ron Artest's move to the Lakers. Although he and Kobe Bryant had their issues during the play-offs, I suspect that off-court there is a mutual respect and he should help the Lakers improve with his all-round game. Trevor Ariza moves the other way in the deal. The early signs are that Lakers are again worth a pre-season back, although the prices aren't firmed up yet on Betfair, and BETDAQ (where thes punts have to be made these days) haven't even bothered to put the market up yet.

Fortune And Misfortune Are Two Buckets In The Same Well

Thanks to Tommy Haas, Andy Roddick and Serena Williams my trading month is off to a good start even if my account remains in negative territory for the time being. It could well have been very nicely in profit, but value does not always mean big profits.

In a baseball game yesterday, I had made a big lay of the Texas Rangers at 1.05 in their game at home to the Los Angeles Angels. This is one of my strategies, and risking £125 to possibly win £2,500 is a strategy that works well in many sports. 

Anyway, to cut a long story short, the Angels ended up managing to tie the game in the ninth inning. Usually this leads to a big pay-day. Unfortunately, the market refused to budge by too much. In this scenario, I would expect the price on Texas to move out to 1.6 or so. They will still be favourites, with the bottom of the ninth still to come, but as outs are recorded, the price should move closer to 2.0. I backed back enough at 1.5 to ensure I was green-all-over, but was looking at just £50 on Texas versus £2,000 on the Angels. With two outs recorded, it was looking good, but sadly Texas hit a home run to win the game. No regrets at not taking bad value though. My bet was value, and betting with value will continue to be profitable in the long run. The poor value backers got lucky this time, but luck runs out. It's just frustrating to be unable to fully cash in on a value bet. Winning £50 is ok, but one can't help but feel disappointed!

June Report

As of a month ago, June was my lowest performing month. Was. After a month where I could hardly do a thing wrong, I had my fourth best month ever. A very pleasant surprise, and June moves up to 8th spot as a result. The Lakers win in the NBA as predicted in October was a big part of this, but the Stanley Cup, NRL, Wimbledon, Volleyball, Baseball, Golf and women's basketball all contributed. The Premium Charge did not and it will get July off to a bad start in a few hours when they rob me again before I've made a penny.

I take back my comment about baseball liquidity being lower this year. It is still not half-way through the season and I have already made 83% of last season's entire total. Home 'big' teams lost 12 of the 26 matches since I started laying them, and if the percentages continue like that I will be very happy. Three of my teams have only played one game at home though - the Yankees, Red Sox and Cubs so smallish stakes for the moment. The Mets lost 5 of 8, and the Dodgers 3 of 5. Stand by for a News of the World exclusive - "I made £20 million laying the Mets, in July".

July is currently one of my worst months, and as I am planning to take advantage of the quiet time to take a few days of holiday soon, it'll probably continue to be just that. But all work and no play makes Cassini a dull boy - so he's been told. My choices are Kirkuk, the Gaza Strip, the Swat Valley or Costa Rica.

June was also a good month on the health and exercise front, with not a single drop of alcohol passing my lips, and a total of 20.2 miles was run as I shed a whopping 2.8lbs. I do love my stats.