ODL XLIV "Evaluation" Topic

some dubious speculations in there nate - particularly the reasoning behind Russell's rating - the fact is that Russell's rating is the product of a flawed equation that fails to take into account that certain awards and honors were not available in his era = awards that we all know he would have won routinely

that base 70 rating is about as good as it gets without application of DPoY or team D honors - remember when initially released the 12-13 Lebron season had a 70 rating but adjusted to 90 after he made 1st team all D - clearly being named 1st team is worth a 20 point swing - since there was no such thing back in Russell's era and since admin doesn't care to make the most obvious exception in the history of history Bill Russell is a 70


9/23/2013 6:41 PM

how does he rate that one 90 def season?

9/23/2013 7:09 PM
first year that they had all nba D team was his last year so at age 34 he was 20pts better on D than at any point for the rest of his career

and btw the real screw job is that the D rating is figured at least in part on steals, blocks and defensive boards which are all categories that WIS either fudged or made up whole cloth for players of Russell's era
9/23/2013 7:21 PM
That 90D for Russell is for the same .441 FG% season of Lucas. He only gets a 73D for 56-57 but the league FG% that year was .380. The rest of his career is similar in terms of D rating and league-wide FG%. I hear what you are saying about all D team felonious, another point of inconsistency for WIS.

I've played a lot with the D rankings and how they impact opponent FG%. My conclusion is that WIS does at least some "normalization" of the stats for anything pre-ABA. IMO there is a ton on inconsistency in the rankings, hence my team leading the dh$52 in points allowed last season. I spent like 20 hours running D correlation formulas and thought I had figured out the logic behind the WIS D. Clearly my assumptions turned out to be about 180 degrees different from the actual outcome.
9/23/2013 7:27 PM
I feel your pain felonious, I also hate how WIS "assumes" all of the 3point data for anything pre-arc in the league. They did a ****-poor job for a number of guys with those assumptions. 
9/23/2013 7:30 PM
I'm pretty sure a 50 is a 50 in terms of impact
9/23/2013 7:55 PM (edited)
Posted by felonius on 9/23/2013 7:55:00 PM (view original):
I'm pretty sure a 50 is a 50 in terms of impact
Not if you believe the WIS FAQ.
9/23/2013 7:56 PM
I think you've got it backwards; they do use season specific information to place the player on the spectrum from 1 to 100

but once that rating is arrived at it's application as a factor in the great equation is the same regardless of the season
9/23/2013 8:08 PM
#726 Gameplay
Q. What is normalization and how does SimLeague Basketball incorporate it?

In the context of SimLeague Basketball, normalization refers to adjusting player shooting percentages (2 point and 3 point) by comparing the player's percentage to the league average for that season.

In the SimEngine, the actual process is a bit more complicated. When a player takes a shot, we compare the shooter's percentage to the league average for his season. We also take the shooting percentage allowed (related to defensive rating) for each defender and compare that to the league average for that defender's season, with more weight placed on the defender guarding the shooter.

In effect, a player who shot poorly in a season where the league average was also low may shoot better when playing against players from seasons with higher shooting percentages, and vice versa. It is important to note that the player's actual shooting percentage, as opposed to his normalized shooting percentage, still accounts for more than half of the end result. This is reflected in the # numbers.

9/23/2013 8:26 PM (edited)
To me the above tells us that Lucas and his 50 D rating for 68-69 has a .441% value and any player from 12-13 with a 50 D rating would have a .453% value, and for 87-88 it would be all the way up at .480, hence 50 D players from the late 80's are not nearly as good as players from say 98-99 where the league shot a crazy low .437 thanks to the lockout.
9/23/2013 8:29 PM
i think my head just exploded.  Will weigh in when I'm done cleaning up the pieces.
9/23/2013 9:26 PM
There is a partial season 98-99 guy that I've used a few times (with no crazy ast% bump) and his efg is usually 8-10% higher than IRL, I may take him again this draft if he falls.
9/23/2013 9:32 PM
Still waiting to hear your thoughts on this ash. Is the FAQ a bunch of BS?
9/24/2013 9:24 PM
The league average for that season helps in determining a defensive rating.  The normalization has already occurred prior to the rating being posted.  Once the rating is in, it's in.  A 50 is a 50; a 90 is a 90, just as a 56 2pt%# is 56.  I understand how that language can be misleading... that's what caused my head to explode.  I also came across information in the knowledge base while researching this that is outdated by many years.  Good job, WIS.
9/24/2013 9:48 PM
Thanks ash, the FAQ is extremely misleading, but given my results from the last league I'm inclined to believe you and felon over the FAQ.
9/26/2013 3:07 PM
◂ Prev 1...5|6|7|8 Next ▸
ODL XLIV "Evaluation" Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2018 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.