Wednesday, February 6, 2013

The Forecast and Methodology

There was comment posted on my last post and I wanted to respond to it for a couple reasons.  Firstly I would like to thank and congratulate Joe as being the first commenter.  He mentioned how he didn't know what kind of system I was running  because it has some flaws.  I have added a page showing the model's methodology and I encourage everyone to check it out if they haven't read the earlier blogs of which the page is mostly comprised.  I would also like to point out that I have on numerous occasions raised questions about certain flaws in the model, such as the lack of a large database.  In the girls' model, there is only one full season of information and it is less reliable because of that lack of data.  I would also like to point out that I don't always agree with how the model rates the teams, another point I have made numerous times.

The model depends on as much information as possible and is meant to bring objectivity to the table, not 100% accuracy.  I do this so people have more information and can make informed expressions about their favorite teams.  I read a lot of the comments made on other sites and there are several that make arguments on one piece of information neglecting to mention other important facts that may offset some of their argument.  I would also like to mention that every model or system or person making an educated guess as to the ordering of teams has flaws.  I try to be as up front about mine as I can.

Finally, I will respond to Joe's expression regarding the girls' C bracket.  Joe, you say you’re ok with FP being a 5, but Galway can’t be a 6 because they play a weak D twice.  FP plays in the same division as Galway and plays the same weak D twice.  The fact is that every team that has a realistic chance at winning sectionals plays bad teams.  The four teams I have ranked ahead of FP all play 2 games against teams that are winless combining for a 0-50 mark.  Every ranked team after FP also plays at least one game against a winless team and most of those play two as well. 
For the record, Galway and Lake George (who has played 11 games against D schools to four for Galway) are .0004 apart in their rating and 6 through 10 are all within striking distance of each other.  As for Waterford , I have pointed out their weak schedule (and 43 point average MOV, the largest in the section) and losses out of league and also that I would probably have them #4 without the model’s assistance.

Joe, I appreciate you reading and your comment.  The best way to stay informed about this or any topic is to debate it with other people who have the interest.  I also want to say that set up the prediction model for the boys' and it actually looks more like my personal opinions that the forecast model does.  I'll be revealing those soon.


  1. I was referring to the two games Galway played against Heatly. My opinion is if you say they play the same schedule and FP beat them twice there is no way a 6 fit.

  2. Thanks for clarifying. The difference in the rating between FP and Galway is .087. The difference between Galway at #6 and Middleburgh at #10 is .030. Just because they are back to back in seed doesn't necessarily mean they are close. Heatly is also 12-5, granted playing in the CHVL, and are #4 in the D bracket, that's not really weak.

    I can see valid arguments for Hoosick Falls and Mekeel Academy to be rated higher than Galway but ultimately it won't make much of a difference. HF was crushed by Hoosic Valley both times they played. The model is designed to determine who has the best chance at winning and none of the teams we're talking about are close and only two are rated to even win one game.