[Cc]

Hittade följade hos en hyfsat renommerad Brittisk community! Jag håller inte med om vissa parametrars värden i sammanhanget men ganska intressant summering nevertheless:


SPORT PREDICTIONS (Human vs Computer)

As you may already be aware, many Bookmakers, sports-betting funds and professional gamblers use predictive math modules (Algorithms) to calculate what they consider to be the correct odds ,a look at salaries will show that Quantitative analysts (quants) have taken over the top paid positions within betting companies,the question we are aiming to answer:


Can a maths module beat Human ingenuity?


Their are big differences between brands of cars,we can see that from the asking price, the same is true of Algorithmic betting systems (ABS),some have invested millions whilst others are the home made do-it-yourself kits, humans are no different ,everyone as a different level of ability!


The first thing we need to point out, algorithms are only as good as the instructions given,behind every good algorithm is a good human, without the human input, the algorithm wouldn´t exist!


With regards to predictive sports-betting modules, we should take into consideration the type of sport, for example, tennis is an head to head match whereas football is about two teams, cricket on the other hand is about a team taking on a singular person (you could argue two), if a player as a bad day in a sport such as tennis, he/she will probably lose, in football, the person who is playing poorly could be replaced (in most cases), when they can´t, their are another 10 players on the team.


Lets consider some of the variables in various sports such has Tennis, Football and Horseracing (European) .


TENNIS(Pre game)


The first note would be the difference between ATP (Men) and WTA (Women), the form of a mans game has more chance of holding up, whilst the variation might be only 2-3% it would need to be built into any calculation, other considerations would include:


-Surface:some prefer clay, others might be better on hard-court ,grass or carpet.

-Form: How is the person and opponent playing.

-Head to head : consider past performance against same opponent (consider conditions too)

-Ranking : Figure out the true ranking for conditions.

-Time of year : Some players prepare to peak at a certain time.

-Fitness : Look for any injury issues such as pulling out of recent tournament.

-Tiredness : Some players need games to be fit,others can´t play too many,you might need to consider recent tournaments or the stage of such an event.

-Time of day : Some people are born to get up in the morning, others are night owls

-Weather : Everyone has their optimum conditions,what if their are breaks because of rain,it will effect some more than others.

-Location : If i was playing in London i would do better than in Pakistan, it´s home and the other is a journey + change in available food, remember horses for courses!

-Circumstances: Might have been a recent change in coach or family illness, find out everything!


FOOTBALL (Pre game)


Football leagues vary greatly in quality and some are so bad you couldn´t build an effective calculation,, for the sake of this article, lets consider the league to be honest, such as the English premier league.


-Missing players : How will they effect the balance of a team,what about people playing with injuries?

-Returning player : Not fully fit, how will it effect performance.

-League positions: The further into the season, the more serious the table can be taken,however….

-Current form : Look at 5/10 game form, home and away!

-Head to head : Teams change year to year but this stat tends to hold-up!

-Home/Away : Always consider a teams differences at home and away.

-Crowd effect: Size of crowd anticipated,will be dependent on day,time,weather,opposition.

-Traveling : One team has to travel,how far/long, expected number of traveling fans,local derbies less of advantage.

-Referee: who´s been given the whistle,look at average bookings, advantage for certain teams,favoring home side,prior with players!

-Weather: effect the number of goals,if it´s very cold a team full of Scandinavians might be better than Africans and vice versa!

-Circumstances: Coach ban, new incoming back room,club takeover etc


HORSERACING (European)


I always throw in horse-racing as i consider it to be the best example of humans superiority over prediction software,it´s important to watch earlier races if you are playing on the exchanges, you will learn a lot about the following contingencies!


-Handicap mark : You have heard the saying, weight stops trains, it certainly stops horses!

-Distance: At a stiff track it´s worth looking for horses that stay a bit further and the opposite is true, if a horse stays 1600 metres on an easy track 1400 might be enough if it´s a sitff finish.

-Weight : Some horses are better off giving weight to inferior animals, whilst others are better when they receive weight from better types

-Going : If a horse doesn´t like the ground it won´t win, unless the race is extremely biased or the others don´t like the ground.

-Starting stall : A good draw can make the difference of many lengths, you also need to consider how it will effect the horses style of racing, for example, if the animal is an hold-up type he might be happier with a bad draw, the opposite would be true of a front runner, he would have to use much energy!

-Size of field : You have heard the saying, the bigger the field, the bigger the certainty, that comes around has the pace is more likely honest so the race should be properly run, it clearly doesn´t suit every horse.

-Jockey : Top jockey could add to the horses chance but he will take away from the odds, look for jockeys that suit certain types

-Type of track : It´s no good a big horse running round a sharp track and vice versa, consider the symetrics and whether the track is flat,undulating, has a stiff finish, the distance from the last bend, assign whether it suits front runners or horses that come from behind.

-Pace in the race: No point in betting a front runner if their are another 5 in the field, it would be like playing cut-throat, same as an hold-up horse wouldn´t be suited by lack of pace.

-Trainer form : A quick look at the stables recent form can be enlightening.

-Watch: the horse in the parade ring and going to start, it can be very revealing.

-Gambles : You are not a sheep, if the boat has gone wave it good-bye!

-Jump racing: Type of fence, how many,how stiff,will my horse be suited!


With regards to horse racing, i put a line through half the races everyday, a quick trained eye tells me i don´t like this race, too many reasons to list!


The above are just some of the basics, their are many other issues such as the will to win, a tennis player might have an eye on the next tournament and could take things easy after getting in a bit of practice, a football game as so many contingencies such as more important games awaiting.


CONCLUSION:


For in-play and automated trading an algorithmic tool is a a distinct advantage, especially, if the person controlling the module is adept at tweaking ,afterall, teams and players don´t always perform to expectations,algorithms remove emotion from the game and predictive modules can work faster than the human mind.


Percentually, the human eye should perform better than the predictive module, however, mass balances out, some sports suit data based algorithmic modules (Tennis) and others suit the naked eye (Horseracing),


See it like a marriage, the advantage of the naked eye shouldn´t be undermined, like the wife, we shouldn´t undervalue the maths!