Questions for usacanary re prediction tables (warning long post...)

I?m afraid still have a couple of questions about your prediction tables. Now, I know you tore a strip of me last time I dared to question the wisdom of your predictive efforts, but could you please answer these simple questions for me?

1) If you were only interested in the points needed for 2nd place and 6th place why did you spend most of the time talking about the prospects of different teams? Surely what you should have been talking about is the updated predicted threshold for those aforementioned positions, irrespective of the team that actually occupied said position? The performance of specific teams should be irrelevant but that is what your posts kept talking about.

2) Does it disappoint you that taking a simple average of the last five seasons actually produces a more accurate prediction of the final outcome? To illustrate lets look at the minimum points needed for finishing 2nd or 6th. I.e. 1 point ahead of the predicted totals for the 3rd place team and 7th place team (ignoring goal difference). Note this isn?t the predicted total for the 2nd /6th team, just what it takes to finish ahead of 3rd / 7th which is what actually matters.

Your Feb prediction: points needed for 2nd= 80 / points needed for 6th = 74
Your April prediction: 2nd = 79 / 6th = 72

5 season average: 2nd = 81 / 6th = 72

Actual: 2nd = 81 / 6th = 73

You might say that actually you did quite well ? that both your predictions were ? in total ? only two points out. However, the 5 season average ? a much simpler process was only 1 point out in total. It got the 2nd place spot on (this season) and was only 1 out for the points require for 6th.

This season it was better than your system, and it required much less f**king about.

Of course the fundamental problem with your system is that towards the end of the season current form becomes more and more significant than ?season? form. Even in the last round of games there is still the possibility of a variation of 3 points for each team.

Plus the sequence of games matters. E.g. Cardiff v Boro then us v Pompey.

Thus your system, I think, is more effort than it is worth. It basically 'predicts' that lots of teams at the top of the table will end up with lots of points.

Posted By: Cart Gap Disco on May 27th 2011 at 15:57:06

Message Thread

Questions for usacanary re prediction tables (warning long post...) (NCFC) - Cart Gap Disco, May 27, 15:57:06
- Wow..... this tilt took a long time to mature......... (NCFC) - usacanary, May 27, 16:09:55
  - Ad hominem as usual. Just answer the questions please (n/m) (NCFC) - Cart Gap Disco, May 27, 16:12:25
    - This........ (NCFC) - usacanary, May 27, 16:17:56
      - Really? That's the best that you can do? (NCFC) - Cart Gap Disco, May 27, 16:26:04
        
        easiest thing to do, is remember when he next posts 7 paragraphs about (NCFC) - pants, May 27, 16:37:22
        
        If I make people smile every now and again.........my job is done. (n/m) (NCFC) - usacanary, May 27, 16:40:36
        
        Oh I see... (NCFC) - Cart Gap Disco, May 27, 21:04:00
  - ROLF (n/m) (NCFC) - SCC 28, May 27, 16:10:25
- Ben. Please delete this post. (NCFC) - Tricky Hawes, May 27, 16:06:46
- Oh my goodness, what manner of madness is this? (n/m) (NCFC) - jafski, May 27, 16:05:15
  - Its Friday! (NCFC) - usacanary, May 27, 16:28:34
- Oh no. (n/m) (NCFC) - Arizona Bay, May 27, 16:00:31
  - Sorry (n/m) (NCFC) - Cart Gap Disco, May 27, 16:02:25

Reply to Message

In order to add a post to the WotB Message Board you must be a registered WotB user.

If you are not yet registered then please visit the registration page. You should ensure that their browser is setup to accept cookies.