Jump to content

Community Blog

  • entries
  • comments
  • views

Contributors to this blog

The Locker Room - Odds Compiler




'The Locker Room' is an article series which'll give you a behind the scenes look into who we're and how we're working at Unibet.

Going through the sportsbook section looking for the next great odds on which to place a bet, you’ve probably wondered what it’s like to sit on the other side of the table, you know, being the one who decides the odds to be put on the Unibet site.

The first instalment here in “The Locker Room”, our series of a behind the scenes look at the inner workings of Unibet, is an interview with the compiler Alistair who’s one of our great Unibet Group odds compilers.

Alistair has been working as an odds compiler for more than 2 decades, and in this time he’s seen and worked through massive changes in the betting industry, including the change that’s made it possible for us all to be here - the internet implementation which began back in the late 90s.


What does a normal day look like for a compiler?

Depending on the sport you are responsible for, each and every day can be different to the last - though many tend to have a routine of sorts that runs over the course of a week, culminating at the weekend.

Golf is a good example of this.

On a Monday they would be using their database to assess the abilities of each player on the type of course they are due to play and price up the outright winner, plus an assortment of prop markets including Tournament Match Bets, Tournament Group betting, Make The Cut etc.   On Tuesday the 1st round 3-balls are released and these are priced accordingly.  Wednesday they tend to have a day off.  Thursday through Sunday they will be trading the tournament in play.

How do you decide the odds? Can you describe the process?

Most sports are similar to each other.  Generally speaking, in Football you have a goal expectancy and supremacy which determine most of the core markets.  The equivalent for Tennis would be 1st Serve / 2nd Serve success rate.

There are exceptions though – depending on the format, Cricket match prices can often be influenced by the weather & the pitch – and depending on where the match is played, this information might not be available until such point as the match goes in play.

The final part of the process, especially when you are new to compilation, is to check off your prices against the industry.  Sometimes you might find that you might not have to be 2/1 (3.00) about X to beat Y, and could be top price at 7/4 (2.75).

The process for deciding an odds is basically the same today as ten years ago, though the necessity of it is reduced thanks to the many trading tools and subsequently feeds that now power many sports.

Is there any difference when you decide an odds for football, tennis and general sports compared to decide a special odds for TV-shows and entertainment (Dancing with the stars, X-factor etc.)?

The world of Novelty & Political betting is a different beast again.  There are general trends in the extended series such as X-Factor and Britain’s Got Talent (historically participants who appear in the first televised show have a good record).  There are also some nuances that can catch people out with the side markets – a good example would be for Elimination markets as some ask for you to vote for who you want to leave, some for who you want to save.  It might sound similar, but can lead to vastly different outcomes.

Which stats do you consider the most when you decide an odds?

As there are simply hundreds of leagues now, we have a feed to provide our football prices.  However the theory behind pricing football is as described earlier.  There are of course some trends that, if you were pricing these matches yourself, could impact some of the expectancies.

A good example was highlighted to me by a former colleague who concentrates on German football a few weeks ago, when Bayern Munich host lowly Eintracht Frankfurt.

- The match was the first played after an international break, where Bayern had every fit player in their squad on duty.

-  Frankfurt’s league position of 17th defied their defensive shortcomings, but highlighted their inability to score.

- In the first 6 matches after the German winter break Bayern been involved in only 2 matches that registered 3 or more goals.

- Bayern were scheduled to play a Champions League ¼ final three days after this league match.

The market internationally made Bayern 2.5 Goal favourites, with a match goal line of 3.75.  Bayern won 1-0 !

What is the craziest game you have traded, where the betting did not go as expected due to unforeseen circumstances?

This happens more often in cricket than any other sport I have been involved with.  1-day and, even more so, Twenty20 matches can swing very fast and very quickly.  However the big swings in a Test Match can take days to evolve, but every so often things can change really quickly.

One match from each format came to mind quite easily though:

T20.  March 21st, 2014.  In the World T20 qualification group in 2014, Ireland played Netherlands in the final group match.  Only one of them or Zimbabwe (who played UAE) at the same time could qualify.  If Netherlands beat Ireland, this would come down to Net Run Rate - cricket's equivalent of Goal Difference.

Ireland batted first and scored an impressive 189/4.  At the innings break Ireland were 1/50 (1.02) to win the match. Netherlands knew they needed to chase this within 14.2 overs (86 balls) and started their reply with accordingly.

Usually you would expect this attitude to lead to the loss of a few wickets along the way, but they had 91 runs on the board by the time Peter Borren went at the end of the 6th over and in the space of 36 balls had gone from 12/1 (13.00) into 1/3 (1.33) to win the match.  Now needing 99 runs in 50 balls, they lost their next two batsmen in 8 balls for 9 runs.  The match was now 10/11 (1,91) each but chance of qualification was surely gone.

The Dutch ticked over for the next 2 overs, but Tom Cooper and Wes Barresi picked on the two spinners, Stirling and Dockrell in the 10th and 11th overs to plunder 41 runs.

Going into the 14th over, the Dutch only required 13 runs to seal a win within 3 balls to spare, breaking Irish and Zimbabwean hearts in the process.  They only won one match in the competition proper - against England!


One Day Internationals:  March 12th, 2006.  After 4 games, the 5 match series between Australia and South Africa, the two best teams in this format at the time, was in the balance at 2-2.

Australia won the toss and elected to bat first. They had a reputation of going strong early with keeper-batsman Adam Gilchrist opening the batting and they didn't let anyone down with the 96/0 they scored from their 15 over powerplay.  When he fell for 55 a couple of balls later, captain Ricky Ponting came to the crease and compiled one of the most impressive innings the game has seen - 164 from 105 balls.

In terms of trading, it was the perfect storm.  Australia had managed to put together 434 runs.  To put this into perspective, no team had scored even 400 in a ODI before.  There was a fair bit of blood on the trading floor......

We had lost plenty already on the innings runs market and didn't have the capability at the time to offer prices shorter than 1/100 (1.01) in the match book.  We therefore just had a price available on South Africa and No Offers on Australia at the half way mark.  The genuine price was probably closer to 1/250 (1.004) and, with 10 years of reflecting on the match, wish I'd allowed our big short price backers get just a little bit of value.

However when South Africa lost their opening batsman for 1 off 6 balls, the game looked as dead as a dodo.  Then captain Graeme Smith and the dynamic Herschelle Gibbs put on 187 runs in 20 overs to put the hosts back in the game.  We had laid plenty of South Africa to momentum customers, but had at least managed to make Australia a winner in the book!

With 10 overs (60 balls) left, SA needed 90 runs but, crucially had lost 5 wickets.   We were certainly cheering the Aussies and patting ourselves on the back for a trading recovery job well done.  Disaster arrived though when Australia called on Mick Lewis (who had so far recorded 7-0-72-0)  to finish his overs.  His last 3 overs went for 16, 8 & 17 to leave both the bowler and the compiler as broken men.  SA won, 9 wickets down with a ball to spare.

The match left a big enough impression on the world to have it's own WIkipedia page.... https://goo.gl/Wbku3B

Test Match:  November 2011.  There have been many classic games but the one which stood out for swings was another between South Africa and Australia.  This one stood out mainly as, on Day 2, I was training a new member of the team on our cricket trading tool.  Opening prices for the game were SA 6/4 (2.50), Draw 7/4 (2.75), Australia 9/4 (3.25).

SA had chosen to field against Australia on a damp morning and it looked like they had made the right choice.  Australia closed Day 1 on on 214/8 and were eventually all out an hour before lunch on Day 2 for 284.  Overnight prices:  SA 4/5 (1.80), Draw 3/1 (4.00), Aus 11/4 (3.75).

By lunch on, SA had made their way to 49/1 from 13 overs.  However by the end of the 2nd session, Australia were batting again - and were already 3 wickets down for just 13 runs.  The session stats read:  18.4 Overs - 60 runs - 12 wickets. South Africa had gone from 4/5 (1.80) overnight to 6/1 (7.00) by the time they were bowled out (very little chance of a draw by then) and back to an even money shot by the time they went to tea.

Australia were bowled out for just 47 in their 2nd innings and South Africa, having conceded a 1st innings deficit of nearly 200 runs now only needed 236 to win the match and were rated at 1/2 (1.50) to get home.

By the close of play, SA had reached 81/1 from 17 overs and were making it look easy again.  They went on to get 236 for the loss of just 2 wickets and two batsmen getting a century.

How do you see the development for the industry and your job specifically?

When I started in the industry back in 1992, you could bet in a shop, and if you had a credit account, on the phone.

If you wanted to bet 90m prices on football, you could only do so up until kick off and, if the match wasn’t live, your bet had to be a minimum of a treble.  If you wanted to bet on other sports, good luck!

The internet changed the face of betting for the better, as did TV coverage of sport in general.

 The process, as detailed earlier, is the same, though the necessity of it is reduced thanks to the many trading tools and subsequently feeds that now power many sports.

 How does live betting work, compared to pre-match?

Most of my experience on live or in play betting has come in Cricket.  The game has evolved greatly since I first traded a one day international between India and Sri Lanka many years ago.

For a start, back then, quite often we were the only firm betting on the match so you had no point of reference if you hadn’t done any pre-match research on ground stats etc.  Thankfully the live schedule wasn’t as heavy as it is now and allowed time to collect all relevant information to go in to battle well-armed.

Going in to trade a match now is easier for a number of reasons:

- Cricinfo has an excellent stats database to obtain a quick look up of the number of 1st Innings runs scored at each ground.  From this you arrive at the average for the ground & add a team factor to determine your run line and also the number which you believe will be enough to win the match

- We have a tool that, whilst still humanly operated, drives both derived markets and a scoreboard!

- There are many other firms now betting on cricket.  If you have to pick up a match from a few overs in, and have a basic understanding of how a match flows, you can do so quite easily without sticking up any arbs on run lines or match prices.

 In case some of our members are interested in a career as a compiler, do you have any tips for them? What kind of background/education do you need to become a compiler?

A mathematical background is pretty useful if you want to cut it as a compiler.  But you also have to be able to interpret your stats.  A common sense factor, if you like.

My background has been more about on-the-job numeracy.  I have had three good employers who have either placed me on courses, exposed me to what it takes to be a (semi)pro-gambler and finally one who was brave enough to give me the opportunity to put these skills into practice.

For anyone wanting to enter into the world of compilation, the best advice I could give them is to ask advice whenever you are not sure, and to not cut any corners.



Recommended Comments

this was a great read wish i could learn to compile odds , i have lots of free time at home and would love to try

Link to comment

Well that was insightful, cool idea that, to let us "meet" the team.

Thanks for the peek behind the scenes on a super smart guy, with an incredibly complicated job.

Link to comment
  • Create New...