2015 World Rankings Topic

That's actually a really good point. Looking at this I'm just trying to see where we can improve. The 100+ losses is what gets us and I'm partially to blame (gulp) for that.
3/19/2015 2:57 PM
Posted by mchales_army on 3/19/2015 1:34:00 PM (view original):
Posted by MikeT23 on 3/19/2015 1:12:00 PM (view original):
FWIW, I've always said that these rankings won't determine if you enjoy your world.   That's the most important thing. 

That said, I always enjoy these rankings because they confirm what I already know.    I commish, or am in, 3 of the top worlds in HBD.   And I recognize that the two I commish aren't for everyone.   Hell, we're not good in the turnover ranking because we send people who can't keep up on their way. 

I ended up using only a 10th of a point per deviance from the baseline when it came to Turnover and Rollover Time, because otherwise, it skewed the rankings too much.

Similarly I counted Playoff appearances and WS appearances double. This is because they are a big factor in my mind. The fact that the highest total you could have in WS was 8, and playoff appearances was 32 and a team that only had 4 and 18 wasn't penalized enough to accurately reflect what I value as indicators of parity.

The one thing I would change if I did it agian is on the Most Wins and Most Losses category.
 
I think I should have dropped the highest from each. So this way if a world had 10 instances of 100 win teams, and their highest was 125, but their next highest was only 108 it reflected on them a bit too harshly this time in my estimation. Dropping the outliers would have produced a better, more accurate picture...

I may revisit that category alone and see how long it would take to redo that portion.

I think adding up the top 4 or 5 wins/losses might be a better indicator of incidence of juggernauts and super tanking.
3/19/2015 8:22 PM
Average of top 3/bottom 3 might work best.   That said, if the top/bottom is 120, it's probably a ****** world even if 2-3 are 94 wins/losses that would bring the average near 100.    I'd much rather have 3x100 loss/wins teams than 1x120 team.   
3/19/2015 8:32 PM
Adding up and/or averaging still doesn't solve the problem of the one abandoned team dropping 130+ in a world that before that incident had no teams over 110.

Eliminating them altogether does that. 

If I dismiss the most extreme and yet you still have 120+ as your high/low, that 's pretty telling imo.
3/19/2015 10:16 PM
got some work to do 
3/19/2015 11:39 PM
If you're not replacing abandoned teams, you're earning your lower ranking.
3/20/2015 7:01 AM
Posted by frymaster99 on 3/19/2015 11:39:00 PM (view original):
got some work to do 

You're not helping yourself by letting a tanker return. 

3/20/2015 12:06 PM
I'm surprised at seeing Cal Ripken's wait number at 81, especially since last season we started less than 72 hours after rollover (including a 48 hours grace period we gave owners who had missed getting in before rollover), and this season started after LESS than 24 hours waiting time.  That's four days...not sure how in the two seasons prior we waited a total of 77 days, because I've been in the world for all of that time and do not remember a wait that long at all.  
3/20/2015 1:36 PM
Looks like each season counted in days took to complete and then added together and subtracted from 360 days to get total wait time??? 
Here is my world.

Date Transaction Days to next season start    
34 1/26/2015 5:44 Contract Renewal      
33 10/25/2014 21:48 Contract Renewal 92    
32 7/26/2014 5:54 Contract Renewal 92    
31 4/8/2014 21:58 Contract Renewal 108    
30 12/27/2013 6:03 Contract Renewal 103 Basline (?) Difference
Total     395 360 35
3/20/2015 2:03 PM
Posted by sgmedia on 3/20/2015 1:36:00 PM (view original):
I'm surprised at seeing Cal Ripken's wait number at 81, especially since last season we started less than 72 hours after rollover (including a 48 hours grace period we gave owners who had missed getting in before rollover), and this season started after LESS than 24 hours waiting time.  That's four days...not sure how in the two seasons prior we waited a total of 77 days, because I've been in the world for all of that time and do not remember a wait that long at all.  
I'm assuming the seasons counted for your world were 26-29, since 30 just started.

Looking at the last login dates of the players, it appears that there was a big wait before season 26. Season 25 ended 12/19, and 26 didn't end until 5/4. That's almost 50 days of waiting. There was also a wait of about 18 days between 27 and 28. That accounts for most of the 77 day figure.
3/20/2015 2:11 PM
I like the rankings and the effort. I had done my own rankings several years ago (looks like ~5 based on the files I have), so I know how much effort is involved.

If you are looking for measures to consider for future iterations, here is how I did mine:

I was most interested in parity myself, and I didn't want to have to scrape too much data. So pulled all my data from the team pitching stats page.

I measured standard deviation of team wins, team runs scored, and team unearned runs (an attempt to measure bad fielding based tanking from pitching stats)

For each of those categories, I compared the world's standard deviation to the average of all worlds, giving either a positive score (lower than average) or negative score (worse than average). And zero basically meant average.

I had do so some monkeying around with weightings for the 3 categories to balance things out a bit and make the score on the wins measure the most heavily impactful. I also only looked at the most recent two years (weighted 2/3 for the most recent year) because I didn't want to penalize a world too much that was trying to correct prior imbalance.

If I did the rankings again (it would probably take 8 hours or so of boring cut and paste work), my guess it is would have an order that looks a lot like your list. By which I'm not saying my way was any better or worse than what you are doing, just a different idea for how to do it.
3/20/2015 2:23 PM
Posted by hineiii on 3/20/2015 2:03:00 PM (view original):
Looks like each season counted in days took to complete and then added together and subtracted from 360 days to get total wait time??? 
Here is my world.

Date Transaction Days to next season start    
34 1/26/2015 5:44 Contract Renewal      
33 10/25/2014 21:48 Contract Renewal 92    
32 7/26/2014 5:54 Contract Renewal 92    
31 4/8/2014 21:58 Contract Renewal 108    
30 12/27/2013 6:03 Contract Renewal 103 Basline (?) Difference
Total     395 360 35
That's 100% correct.  I just added the days over 90 each season. So in your example it was 2+2+18+13 = 35.
3/20/2015 2:37 PM

Here's Cal Ripken's:

  Date Transaction   WAIT TIME
3/6/2015 9:41 Contract Renewal    
29 11/29/2014 5:41 Contract Renewal 97 7
28 8/11/2014 21:44 Contract Renewal 110 20
27 5/4/2014 17:45 Contract Renewal 99 9
26 12/20/2013 1:48 Contract Renewal 135 45
        81

3/20/2015 2:47 PM
Hmm...not sure how it calculated that 97. Last season, as I said, we gave owners that missed 48 hours then filled in under 24 after that (It may have even been 12...can't recall).
3/20/2015 3:11 PM
Posted by sgmedia on 3/20/2015 3:11:00 PM (view original):
Hmm...not sure how it calculated that 97. Last season, as I said, we gave owners that missed 48 hours then filled in under 24 after that (It may have even been 12...can't recall).

You must also remember that sometimes there is a logjam of worlds waiting to rollover. ADMIN limits it to two or three per day or something.

hopkinsheel's worlds are approved the very same day they rollover without exception and wait time even in his worlds vary between seasons. Sometimes it is right at 90, but sometimes it is 91 or 92.

Anyway it isn't the 7 that's hurting, it's the 45.

Lastly, if you rolled from S29 to S30 really fast, that will not be evident until S31.

3/20/2015 3:40 PM (edited)
◂ Prev 123456 Next ▸
2015 World Rankings Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2024 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.