Thursday 19 January 2012

Where Are Your Eggs?

Don’t worry, this is not a personal, anatomical question. I'm not going to ask you to participate in some kind of bizarre "cough-and-wobble" test; and the question is not in the same parlance as “Do you dress to the left or the right?”

Cassini from http://green-all-over.blogspot.com/ kindly referenced my last post again today, but I did spot in his comments section this from Rob the Builder (I hope both Cassini and Rob don't mind me quoting it here):

I spent a while reading Eddie's piece. Heavy going but nice to see some real detail and thought on a blog. But doesn't his example show one of the limitations of calculations. However good the maths, I didn't see the formula include whether De Jong and Barry were holding down in midfield, or whether Mancini was 'going fo it' with Johnson, Balotelli, Silva and Aguero.

Correct, there was no formula for this (hang-on while I knock one up) but Rob does raise an interesting point that I feel is worth pursuing. And it's all to do with our individual approach to betting/trading.

If I use some of the well-known "ready-made" systems such as Poisson, Rateform, Goal Supremacy and like, or whether I spend some time to generate my own (hopefully) unique stats based on a self-created system, the question remains as to how best should I apply them? I mentioned a short while back that (for me at least) it would be heaven to be able to fully automate all my betting activities. Just press a little button in the morning, then check again in the evening to see how much profit I had made. Ooh, lovely. I'm sure we'd all like this idea - and doubtless there are some clever individuals out there who have managed to create just this situation.

One of the keys to being able to achieve this wonderful scenario is to have an algorithm that you can solidly rely on - and more importantly solely on. No account can be taken of which players have been selected, what their motivation levels are, whether one team has just appointed a new manager, or any other non-mathematical criteria. The algorithm has to stand alone and must succeed alone with no other factors involved.

Most people however (and I include myself here) are not in the enviable position of having a fully-automated solution, and so need to work a bit harder and use manual intervention. Of course we can still use any of the ready-made systems or any of our own bespoke systems to help us make our decisions, but we then have to complete the process manually.

And it's at this point that we have a distinct advantage over our fully-automated counterparts. Here, before placing our bet, we can introduce other factors that a push-button approach is unable to do. Our systems may tell us that team A will easily beat team B and that the offered price appears to be good value. Okay, all well and good. But then we can also look at other factors such as team selection, motivation and the like to "flesh-out" any selections that we may decide to make. Some of these additional criteria will ultimately override the mathematical ones and cause us to drop the bet.

Anyone who does not use a fully-automated system but who is relying solely on an algorithm or mathematical equation to make their selections is, in my opinion, a fool. They are missing a trick and their P&L will undoubtedly suffer from it. Why wouldn't you use everything that is available? It doesn't make any sense.

So to answer Rob's point about there being no formula for who's playing, or how the team is set out... well yes there is a formula: it's just in your head and created on-the-fly when you assimilate relevant match information. And this "formula" should be used each and every time AFTER having applied a mathematical approach.

Eggs all in one basket is not good, as the cliche states.

______________________________________________________________________


Traded the Italian Cup match between Inter and Genoa this evening, and have a loss to report. I think I can happily blame my wife for this though (well, you can't honestly believe it's my fault, can you?).

She's going away for a couple of days and was buzzing around like a blue-arsed fly, trying to get all her stuff together.

"Where did you put the holdall? You know the one with the black along the side? I need it now."

Yes, I think you know what I'm going to say next. I decided to "just slip away quickly" from the match to sort out her problem. I placed an exit order which I assumed would get filled in the next ten seconds or so and then walked out of the room. The rest, as they say, is history.

Okay, don't try and tell me what I should have done. I know full well. I was in Baghdad when you were still in your dad's bag, so I don't need telling...

Hmm, then again, it looks like I do.

Soccer: -£143.56 | Tote: |  Total P&L:  -£143.56 

 

Soccer Showing 1 - 1 of 1 markets

Market Start time Settled date Profit/loss (£)
Soccer / Inter v Genoa : Over/Under 2.5 goals 19-Jan-12 20:00  19-Jan-12 21:48  -143.56

5 comments:

  1. Blog's a great read, added it to my list and bookmarked your previous post on Poisson.

    When you take into account players, morale and all the rest aren't you just "muddying" the water so to speak.

    Isn't it better to have a purely mathematical system where there's no tweaking involved?

    ReplyDelete
  2. Hi Brian

    Thanks for adding me to your list.

    I would disagree with you though on your "muddying the waters" point. For example, if your purely mathematical system says that Man Utd should be 1.50 away to Bolton but the given odds are 1.60, you might say they were value.

    But what if they have a Champions League match a couple of days later and have filled their team with second-team players? Still value?

    ReplyDelete
    Replies
    1. True I suppose in that sense then yes your correct. Though surely just one player not playing shouldn't make that much of a difference, and if it does then doesn't that mean that the team isn't a good team if one person makes such a bug difference?

      Delete
  3. I said "filled with second-team players", not one player missing.

    But even then, look at Arsenal at the moment. If Van Persie's not playing, I would expect the odds to move accordingly. Pure stats would/could not take account of that.

    ReplyDelete
    Replies
    1. Stats - nothing but ....
      just want to invite you to my new blog
      undergroundtrader

      Delete

Note: only a member of this blog may post a comment.