Mime-Version: |
1.0 |
Sender: |
|
Subject: |
|
From: |
|
Date: |
Thu, 18 Mar 1999 12:33:09 -0600 |
Content-Type: |
text/plain; charset="us-ascii" |
Reply-To: |
|
Parts/Attachments: |
|
|
At the risk of thumping a deceased equine....
I recently remembered that I had written a program several months ago to
compare rankings of different ranking systems with actual game results.
The program simply looks at all games that did not end in a tie and did not
involve UNO (as not all systems include them this year) and tracks how
often the higher rated of the two teams in each system won or lost the
games.
I reran it with the current data and the results are below, listed from
highest prediction percentage to lowest. As many expected, KRACH came out
on top.
KRACH Right 565
KRACH Wrong 193
KRACH Pct 74.538
Massey Rignt 561
Massey Wrong 197
MASSEY Pct 74.011
RPI Right 558
RPI Wrong 200
RPI Pct 73.615
CHODR Right 557
CHODR Wrong 201
CHODR Pct 73.483
CCHP Right 551
CCHP Wrong 207
CCHP Pct 72.691
Heal Right 535
Heal Wrong 223
Heal Pct 70.580
It should be noted that I did not use any of the systems' home ice
adjustments.
Also, while KRACH is apparently always the best at matching with past
results (those from which the rankings are derived), I have found in the
last year that the CCHP has always been better at predicting future results.
Does anyone know why that would be true?
HOCKEY-L is for discussion of college ice hockey; send information to
[log in to unmask], The College Hockey Information List.
|
|
|