Thursday, 13 October 2016

Messy Approach

I was looking through my Blog Roll today, and hidden under the Horse Racing section, a category I seldom visit, I happened upon the RaceTrader blog, which it turns out isn't that much about horse race trading. Maybe it started out that way, but recent posts are all football related.

The writer, Crazy Horse, doesn't post too often, three or four times a month, and the latest post is on the subject of statistics, and their usefulness for trading.

He posits: 
How many times have you researched a bet, only to find that the result is the opposite of what you predicted?
I'd hazard a guess that the answer depends on the price of the bet, and question whether when betting you are predicting a result at all. What you should be doing when researching a market, is identifying which of the possible outcomes (if any) is undervalued.

If I back a draw in football, I'm by no means predicting that the game will end as a draw. I'm backing it because the market estimates the probability of a draw at a certain percentage, whereas I estimate that the percentage is higher.

The writer continues, and after correctly stating that "all market prices are largely correct and most people wouldn't know value if they saw it" asks:

So why bother betting or trading at all if stats offer only limited advantage and there is scant value around? Critically, most of us have the ability to watch football matches. If you can read a game - you can get your edge.
The problem with this assertion, that you can get your edge, is that it makes the assumption that you have the ability to 'read' a football match better than everyone else watching the same game. As much as many of us might like to think we have this ability, the truth is that statistically, it's rather unlikely.

The writer goes on to describe 'trading' (his term) the Under 2.5 goals market in a recent La Liga game:
The other night I was trading in the 'Under 2.5 goals' market in Celta Vigo v Barcelona (why you might wonder! Ok, Messi wasn't playing and I know Celta can be tight in defence - but even without these stats I still would have looked at the game!) Celta's previous three La Liga matches at home finished: 2-1, 0-4, 0-1. Barcelona's previous three La Liga matches away finished: 0-5, 1-5, 0-1. Last season the corresponding fixture finished 4-1 in favour of Celta Vigo. Did I think the game would go over 2.5 goals? Almost certainly. So why trade the Unders? Simply because the odds 'decay' from kick off and I could watch the game, deciding when to enter and when to exit.

From memory the odds on offer at the start of the game were around 2.72. I watched for a few minutes and then got in at 2.6. It was all Barcelona but Celta were holding there own in defence. Then Celta started launching their own attacks - the game began opening up and I started coming out of my position. This was around the 20 min. mark and I was laying at around 1.98. A couple of minutes later Celta scored. I would have been on for about a 30% profit. The goal just game a bit soon / I didn't exit quickly enough - so it was a scratch trade 0.00. No profit, but I was able to avoid a loss by exiting when I felt things were about to go against me.
So a few things about this, and yes, I did wonder why anyone might think they had some kind of an advantage trading this market. 

That Messi wasn't playing was well known and already priced in.

That "Celta can be tight in defence"... what does that CAN there even mean? Every team is capable of keeping a clean sheet. The probability of it happening depends on the quality of the opposition. A clean sheet against Barcelona is far more unlikely than a clean sheet against well, pretty much anyone else! 

Then we have the 'Norsted Nonsense' approach of looking at both teams' previous La Liga games, without any context whatsoever. Who were the opponents? Barcelona, Real Madrid and Atletico? Or Granada, Osasuna and Valencia? There's a difference! 2-1, 0-4, 0-1 takes up space in a write-up but tells us nothing useful. 

And what does last season's fixture have anything to do with today's game? Nothing. Teams change. Managers change. Tactics change. Motivation changes. 

"Did I think the game would go over 2.5 goals? Almost certainly."  The market opinion was that there was a 63.2% probability of more than 2.5 goals in this game. What does 'almost certainly' mean in probabilistic terms? 80%? - in which case just back the Overs and go out.

"So why trade the Unders? Simply because the odds 'decay' from kick off and I could watch the game, deciding when to enter and when to exit."

Ah yes, that's a great reason, except that this is gambling, not trading. The thinking that time-decay offers an advantage is wrong. As mentioned before, this is akin to picking up pennies in front of a steam-roller. It's a steam-roller that can, and will, hit you at any time, however clever you think you are. 

That Crazy Horse escaped from this venture with a wash was fortunate, but if he continues with this approach, he won't always be so lucky. If betting was this easy, we'd all be doing it!  

1 comment:

James said...

Oh well, another one for Green All-Over's betting blog cemetery.

Your blog lists are piling up at the bottom of each section with blogs that have bitten the dust.

Tales of initial exuberance brought crashing down to Earth by the Gods of Mean Reversion.

Some people haven't been reading their Buchdahl.

Not a Buch by Mr Dahl but Squares & Sharps, Suckers & Sharks by Mr J. Buchdahl.