Loading ...
Sorry, an error occurred while loading the content.

Re: [hockeydisk] upcoming disk

Expand Messages
  • David Gowdey
    Many thanks to all for all the work on this great project  ... per game basis, and some are independent. ... the others, such as chance of shot taken, are
    Message 1 of 10 , Oct 7, 2013
      Many thanks to all for all the work on this great project 

      From: Brian Senecal <brian@...>
      To: hockeydisk@yahoogroups.com
      Sent: Saturday, October 5, 2013 10:39 PM
      Subject: Re: [hockeydisk] upcoming disk

      OK Guys....the disks are uploaded to apbahockey.com

      Be sure you get the right one....the 48 game season is the main one and the other is marked as pro-rated 82 game season.  They can be found int he Downloads section

      Also, remember, you must be registered to download anything.

      On Sat, Oct 5, 2013 at 12:10 PM, D. Atkinson <boomhound@...> wrote:
      So, to get into some more technical detail so that you can understand where I think the issues will be.....
      Some of the ratings run on a "per minute" basis, some run on a "per game" basis, and some are independent.
      For example, all the non-minor penalties run "per game". Many of the others, such as chance of shot taken, are per minute. These *should* scale as the disk is extrapolated from 48 to 82, but I'm sure they are perfectly linear, from effects that I've seen in testing. Thus, there is a chance that in scaling the disk, one might end up with more major penalties, or less game misconducts, than expected. The 48 game disk was tested and adjusted to optimal performance on a 48 game season. So, there is a small chance of some strange numbers...my guess is that these will occur with penalties rather than goals/assts/etc.
      One more note....this disk, as have all past disks, has two main issues that are unfixable and I do the best to mitigate the problem with a happy balance:
      1) all high assists players that take many shots too will end up 20-30% low on assists. This has been a long term flaw in the game engine. To get their assists up, at some point shots on goal (and thus goals) have to be sacrificed by that player. So, I try to get them as many assists that I can, only giving up shots/goals to get them in the bracket at the lower end of the acceptable range. This year, Zetterberg and Crosby are the biggest culprits, expect their assists to be 20-30% low. Crosby is a problem with this every year since he gets many goals and assists....a few high puck possession guys are the usual problems.
      2) goons will get too many PIM. This also happens every year, and is typical of players where many of their penalties are fights and misconducts rather than minors. Parros, Blanchard, Volpatti fit the bill this year. The trouble with them is that with a high PHY/INT, they will still get minors even if their PEN RATE is 0. I try to get their fight totals per minute about on, and let the other numbers fall where they may. I often sacrifice MISC to get PIM down, because I don't have full control of minors in the ratings.

      The two disks are going to Brian for posting shortly. Enjoy, and I hope this year's disk works out well for you and your league. It was a challenge with the short season (smaller sample size). But, I got some new testing tools made, and the voting system that Ken Danila developed worked fabulously to get the ratings done in a timely manner. Next year, with a full NHL season, things should go more smoothly and quickly. I apologize for the late release (I was really hoping for early Sept), but the short season complicated things.

      Again, feedback on the disk is welcome.


      On 10/5/2013 5:18 AM, Bill Corfield wrote:

       Thanks once again for all the work you and the team put into these disks. The work of a few allow the many to have hours and hours of fun. We very much appreciate it. 

       I'm still unclear on how best to proceed for our upcoming season. We plan on playing an 82 game schedule. If I understood you correctly in prior emails, we can use the 48 game disk WITH NO PROBLEMS AT ALL, AS IS for our upcoming season. But in a recent email, you said if there was enough interest for an extrapolated disk, you'd put one together for those leagues that wanted a 82 game disk. 

      What's the difference/advantage/disadvantages of using the 48 game disk vs. the 82 game one? If you're running a full 82 game season/league, which would you want to use?

      Would you please answer in non scientific terms? Dumb it down for the average APBA guy, ok? 



      -----Original Message-----
      From: D. Atkinson <boomhound@...>
      To: hockeydisk <hockeydisk@yahoogroups.com>
      Sent: Mon, Sep 30, 2013 9:30 am
      Subject: [hockeydisk] upcoming disk

      I finalized the combined disk over the weekend, and it will be going
      into error check tonight. Hope to release by the end of the week, if
      there aren't too many issues with it.
      Question: there seems to be a lot of questions surrounding a disk
      projected out to 82 games, for leagues that want to play a full season.
      I believe I can make one of those fairly easily next weekend, if there
      is enough demand. If your league would like a projected disk, let me
      know (league name) and if there are enough requests, I'll make one after
      I release the standard one. None of the ratings (like SKT/SPD/PAS) will
      change, only the stats based numbers on the disk will be extrapolated to
      82 games.

    Your message has been successfully submitted and would be delivered to recipients shortly.