League vs League standings?

Quahogs · Post by **Quahogs** » Fri Aug 06, 2004 5:39 am

Actually I think the stats do speak for themselves, it says NY League 4 has more cumulative pts than any other league. The cumulative pts are based on stat comparisons btween all 195 teams. Weaker teams in NY league 4 add no benefit to the stronger teams in regard to the OVERALL league comps. This tells me NY League 4 has done an overall better job at managing the player statistics that are available to each and every team regardless of league talent disbursal.

Team Herron · Post by **Team Herron** » Fri Aug 06, 2004 6:58 am

Wouldnt a fair analysis be to treat each league like it was its own team and score the league. Ie the league with the most points gets 12, second gets 11 and so on. If we did that for each stat, wouldnt that show the best league regardless of the strong to weak variation?

Gordon Gekko · Post by **Gordon Gekko** » Fri Aug 06, 2004 7:39 am

You guys are WASTING your time on something that will never be changed. Then again, maybe your time isn't valuable.

Brian Walton · Post by **Brian Walton** » Fri Aug 06, 2004 5:00 pm

Quahogs, you made the key point far more succinctly than I, and hopefully with less controversy. Pitt, perhaps you should get out more and work off some of that aggression. Gekko, thank you for the sage advice. 418 posts here may qualify you as a time-management expert and an expert in many other areas, too, I suspect. Seriously all, my intent was to raise a concern in hopes of some civil discussion.

[ August 06, 2004, 11:17 PM: Message edited by: brian ]

KJ Duke · Post by **KJ Duke** » Sat Aug 07, 2004 3:44 pm

Year-to date --- here are the best and worst by league and city.

Best league Las Vegas #1 (without a doubt)

Worst league, New York #5

Best by city, Las Vegas (also, without a doubt)

Since SportsBuff.com is not only leading overall, but he is also competing in what is clearly the best and most balanced lge, he is right where he belongs as the best team in the NFBC, so far.

Methodology

1) I totalled all team stats by league and ranked each league as if all of its teams were a single team. I then ranked each league 13 to 1 in each of the ten scoring categories.

2) I then measured league quality top-to bottom by calculating standard deviation within each league for each scoring category. I then ranked each league 13 to 1 by lowest standard deviation for each of the ten scoring categories.

Results below show league rankings by:

(1) total scoring

(2) overall league balance

(3) the combined score of lge scoring and balance

Here are the raw results:

TOTAL SCORING

1 LV-1 99.5

2 NY-3 93.0

3 NY-4 91.5

4 LV-2 83.5

5 CHI-2 82.5

6 LV-4 78.0

7 CHI-1 63.5

8 NY-1 62.5

9 LV-3 61.0

10 LV-5 61.0

11 NY-2 52.5

12 CHI-3 46.5

13 NY-5 35.0

LEAGUE BALANCE

1 LV-1 103.0

2 LV-3 83.0

3 Chi-2 81.0

4 LV-4 77.0

5 NY-1 76.0

6 LV-5 69.0

7 NY-3 69.0

8 LV-2 64.0

9 NY-5 63.0

10 NY-2 63.0

11 Chi-1 61.0

12 Chi-3 57.0

13 NY-4 44.0

OVERALL

1 LV-1 202.5

2 CHI-2 163.5

3 NY-3 162.0

4 LV-4 155.0

5 LV-2 147.5

6 LV-3 144.0

7 NY-1 138.5

8 NY-4 135.5

9 LV-5 130.0

10 CHI-1 124.5

11 NY-2 115.5

12 CHI-3 103.5

13 NY-5 98.0

OVERALL BY CITY

1 Las Vegas 155.8 76.6 79.2

2 Chicago 130.5 64.2 66.3

3 New York 129.9 66.9 63.0

At least for now, this should put this debate to rest since Las Vegas #1 is the best league on both measures.

And Las Vegas (the city) is also best on both measures. Sorry NY, not only are you clearly inferior to Vegas, you are currently worse than Chicago.

(I am in LV-5).

[ August 07, 2004, 09:49 PM: Message edited by: KJ Duke ]

Gordon Gekko · Post by **Gordon Gekko** » Sat Aug 07, 2004 4:37 pm

It's fun to play with numbers. Meaningless, mind you...but well done.

JT the Buffmeister · Post by **JT the Buffmeister** » Sun Aug 08, 2004 1:34 am

DUKE - very impressive post. Thanks for the time and effort. The league strength debate is interesting and it really needs to be based on overall statistics, not ranks. I've always thought that you adapt to whatever scoring system there is. From a business perspective, you create a scoring system that will attract contestants. I'm not sure that adding strength of schedule would attract more business.

To everyone. We are having a blast this year. We've always thought that our LV1 league was extremely strong and well balanced (with MeatLoaf in the bottom third, it has to be storng!). Every league in the NFBC is extremely strong. Think about it. There are more than 15 million people in this country that have played fantasy sports. I have thousands and thousands of people playing fantasy sports at our site... and we are all competing against 195 of the best in the NFBC. If you throw out a few inexperienced players, you have the makings of very strong leagues from LV to Chi-town to NY.

Good luck to everyone. We are all making the turn down the home stretch and it's getting very interesting. I know myself and my partners are pumped and focused... and ready to enjoy the strategy of the last couple of months. Now, who will be this year's Reggie Jackson? We need him (although it would help if he qualified at MI instead of OF).

Cheers! JT

PittIsIt95 · Post by **PittIsIt95** » Mon Aug 09, 2004 2:33 am

No aggression in my post. Just trying to drum up some responses to hear some other angles on what makes a league competitive. I just happened to use a different methodology. Numbers can be manipulated in many different ways. And as you can see from other posts, most of the arguments make sense.

JT the Buffmeister · Post by **JT the Buffmeister** » Mon Aug 09, 2004 5:35 am

By the way PITT. Your post and analysis was also top notch. It was an interesting perspective. Thanks. Overall, using cumulative league stats instead of cumulative rankings eliminates the benefit of poor teams in a league... from a league v. league perspective.

Dyv · Post by **Dyv** » Mon Aug 09, 2004 7:42 am

Even using overall statistics produced by a league isn't necessarily representative of anything in particular. I'll use my team, as an example...

We have 10 solid pitchers that we rotate on matchups and have had as many as 12 P's on our roster at one point... again, based on matchups and how many starts per week. This week Jose Contreras rides our pine... last week we missed Zach Greinke's great performance due to his turn on the bench.

So, our league didn't get those stats. In other leagues if you had people playing different strategies and SP were harder to draft or pick up via FA then you may not leave as many good performances on the bench. Some people went with reliever strategies, others punted closers entirely. I've seen leagues where someone would punt the power numbers and go with pitchers and speed. Some teams in every league are picking up 2-start SP each week to try to maximize strikeouts and wins... others are benching SP to pick up WHIP/ERA. All of these 'management' items can affect the 'raw' numbers being produced.

I'm a member of the '2nd weakest' league NY5... and with the exception of 1 dead team we've had tough competition amongst the rest. 13 of the 15 owners have logged into their teams in the last 24 hours. One dead, and the other hasn't been on for about 10 days.

How about some stats wizard working out the stats of offense vs. pitching.. I believe most would agree that few strategies seem to involve 'skipping' hitters, but that with SP vs. MR vs. closers there are a number of strategic plays teams could make. How is the variance between hitting and pitching if broken down? The same, or tighter?

The numbers seem to suggest some things, but I'm curious if there's more.

Dave

KJ Duke · Post by **KJ Duke** » Mon Aug 09, 2004 11:42 am

Originally posted by Dyv:

Even using overall statistics produced by a league isn't necessarily representative of anything in particular. I'll use my team, as an example...

We have 10 solid pitchers that we rotate on matchups and have had as many as 12 P's on our roster at one point... again, based on matchups and how many starts per week. This week Jose Contreras rides our pine... last week we missed Zach Greinke's great performance due to his turn on the bench.

So, our league didn't get those stats. In other leagues if you had people playing different strategies and SP were harder to draft or pick up via FA then you may not leave as many good performances on the bench. Some people went with reliever strategies, others punted closers entirely. I've seen leagues where someone would punt the power numbers and go with pitchers and speed. Some teams in every league are picking up 2-start SP each week to try to maximize strikeouts and wins... others are benching SP to pick up WHIP/ERA. All of these 'management' items can affect the 'raw' numbers being produced.

I'm a member of the '2nd weakest' league NY5... and with the exception of 1 dead team we've had tough competition amongst the rest. 13 of the 15 owners have logged into their teams in the last 24 hours. One dead, and the other hasn't been on for about 10 days.

How about some stats wizard working out the stats of offense vs. pitching.. I believe most would agree that few strategies seem to involve 'skipping' hitters, but that with SP vs. MR vs. closers there are a number of strategic plays teams could make. How is the variance between hitting and pitching if broken down? The same, or tighter?

The numbers seem to suggest some things, but I'm curious if there's more.

Dave I have the stats by each category so I could post when I have a chance - not sure what you're arguing though - that offense alone may be more relevant to lge quality than overall stats ?

Secondly, no statistical analysis is perfect. We all know that luck plays a significant role in the process, so even if the stats show one league to be better, it doesnt neccessarily mean its owners are better, just a greater probability that they are.

But a proper statsistical analysis is certainly better than a suspect analysis (such as the lge totals on the home page) or anecdotal evidence (i.e., my lge has 4 top teams therefore my lge is better).

Greg - how about we dump those meaningless league totals on the home page - their significance is somewhere between completely meaningless and misleading. What is the name of our service provider ? oh yeah "STATS" - kinda ironic that they didn't come up an appropriate statistical measure, isnt it.

Dyv · Post by **Dyv** » Mon Aug 09, 2004 3:01 pm

Hey Zef - basically... what I'm curious about is if there is a variance across the board in all statistical categories, which might suggest simply 'better owners' or if there might be skewed towards variance in pitching which might suggest 'more strategic play' and offer an explanation for the variance.

Dave

KJ Duke · Post by **KJ Duke** » Mon Aug 09, 2004 7:40 pm

Originally posted by Dyv:

Hey Zef - basically... what I'm curious about is if there is a variance across the board in all statistical categories, which might suggest simply 'better owners' or if there might be skewed towards variance in pitching which might suggest 'more strategic play' and offer an explanation for the variance.

Dave I see your point. The stats actually prove your suspicion correct, but still validate that the better leagues are in fact better across the board.

The stronger hitting leagues also show stronger pitching stats.

Lges 1-5 in batting: avg pitching pts = 41.4

Lges 5-9 in batting: avg pitching pts = 35.8

Lges 9-13 in batting: avg pitching pts = 30.0

(The actual "statistical correlation" is significantly positive as well).

However, when it comes to league balance, or standard deviation by category, this same ranking broke down for the pitching stats ... I suspect for the reason you suggested - that the mix of pitching strategies within a league can cause greater variances - and thus have less significance to actual league balance. Statistically, the pitching variances showed virtually no correlation to batting variances. Furthermore, whereas batting variances showed strong correlation to batting totals and even to overall scoring, pitching variance showed much lower correlation to pitching totals and virtually no correlation to overall scoring.

... good observation on your part ...

The next step of course, given that pitching strategies may render pitching variances of little significance with respect to league balance, is to re-rank with batting variances only.

The results were striking similar to my previous rankings, as you can see below, which suggests that (1) you are correct in suggesting that pitching variances within leagues may have minimal significance ; but (2) because this information is in fact nearly random, the previous outcomes and conclusion are hardly even affected by their exclusion.

For the record, here are the revised outcomes for variances only and then for combined score (with rankings under the previous method in parentheses):

League Balance

1 LV-1 59.0 (1)

2 Chi-2 53.0 (3)

3 LV-3 50.0 (2)

4 NY-1 43.0 (5)

5 LV-2 37.0 (8)

6 NY-3 36.0 (6)

7 LV-4 33.0 (4)

8 NY-2 30.0 (9)

9 LV-5 27.0 (7)

10 NY-5 27.0 (10)

11 Chi-1 26.0 (11)

12 Chi-3 25.0 (12)

13 NY-4 9.0 (13)

League Pts and Balance

1 LV-1 158.5 (1)

2 CHI-2 135.5 (2)

3 NY-3 129.0 (3)

4 LV-2 120.5 (5)

5 LV-4 111.0 (4)

6 LV-3 111.0 (6)

7 NY-1 105.5 (7)

8 NY-4 100.5 (8)

9 CHI-1 89.5 (10)

10 LV-5 88.0 (9)

11 NY-2 82.5 (11)

12 CHI-3 71.5 (12)

13 NY-5 62.0 (13)

oh, and Gekko,

fun? yes,

imperfect? yes

meaningless? no

PittIsIt95 · Post by **PittIsIt95** » Tue Aug 10, 2004 2:04 am

Gentlemen, you can argue this till you are blue in the face. But the true measure of a COMPETIVE LEAGUE is the variance between first and last in a category. The tighter this is the MORE BALANCE there is throughout a league.

Believe me, I would ALWAYS want to be in a league where guys are throwin categories because then I could get higher totals in those categories to make a run for the overall championship. Teams in well-balanced leagues really have no shot at the big money. You need weak owners at the bottom to juice up your numbers for the overall.

Throw all your stats, but think logically. 2 teams throw saves in a league. Well that means other teams can carry 4 closers and get in the top 5 of a category. Whereas a league that has balance gets stuck in the middle of the pack because they only can get 2 good closers (maybe 3 max). THAT IS THE TRUE EXPLANATION of a balanced league. You can argue who's league is better, I am arguing what leagues are more balanced (and the only measuring stick is VARIANCE - my major was in Applied Mathematics, so at least I have some background with statistical analysis to back my point, how bout others!! :rolleyes: ).

Dyv · Post by **Dyv** » Tue Aug 10, 2004 2:32 am

Originally posted by PittIsIt95:

Gentlemen, you can argue this till you are blue in the face. But the true measure of a COMPETIVE LEAGUE is the variance between first and last in a category. The tighter this is the MORE BALANCE there is throughout a league.

Believe me, I would ALWAYS want to be in a league where guys are throwin categories because then I could get higher totals in those categories to make a run for the overall championship. Teams in well-balanced leagues really have no shot at the big money. You need weak owners at the bottom to juice up your numbers for the overall.

Throw all your stats, but think logically. 2 teams throw saves in a league. Well that means other teams can carry 4 closers and get in the top 5 of a category. Whereas a league that has balance gets stuck in the middle of the pack because they only can get 2 good closers (maybe 3 max). THAT IS THE TRUE EXPLANATION of a balanced league. You can argue who's league is better, I am arguing what leagues are more balanced (and the only measuring stick is VARIANCE - my major was in Applied Mathematics, so at least I have some background with statistical analysis to back my point, how bout others!! :rolleyes: ). Finance.

The teams in your proposed example can carry 4 closers - and therefore less SP. Not a significant factor as the league still has 32 closers in their starting rosters, just like all other leagues, etc... I don't think pitching variance has anything much to do with it.

I'm still willing to accept that there is a #1 league and a #13 league

Thanks for the calc work Zef... I actually thought it would be more randomized outside of pitching. Still - there's a reason I'm in first place in NY5 and still only 12th overall... lol

Thanks for your time and work there.

Dave

KJ Duke · Post by **KJ Duke** » Tue Aug 10, 2004 5:36 am

Originally posted by PittIsIt95:

Gentlemen, you can argue this till you are blue in the face. But the true measure of a COMPETIVE LEAGUE is the variance between first and last in a category. The tighter this is the MORE BALANCE there is throughout a league.

Believe me, I would ALWAYS want to be in a league where guys are throwin categories because then I could get higher totals in those categories to make a run for the overall championship. Teams in well-balanced leagues really have no shot at the big money. You need weak owners at the bottom to juice up your numbers for the overall.

Throw all your stats, but think logically. 2 teams throw saves in a league. Well that means other teams can carry 4 closers and get in the top 5 of a category. Whereas a league that has balance gets stuck in the middle of the pack because they only can get 2 good closers (maybe 3 max). THAT IS THE TRUE EXPLANATION of a balanced league. You can argue who's league is better, I am arguing what leagues are more balanced (and the only measuring stick is VARIANCE - my major was in Applied Mathematics, so at least I have some background with statistical analysis to back my point, how bout others!! :rolleyes: ). I think the stats I've presented provide more information than just 1st/last variance - I am not only trying to show balance, but overall lge quality as measured by both balance and total stats accumulated by the lge. This is more along the lines of applied statistics and logic, rather than coincidental "stathead" analysis. And - I didn't think I was really arguing with anyone, just taking a look at the evidence.

As to your point about closers, Dyv has already answered - you run 4 closers you are killing your potential for W's and perhaps K's, not to mention the quality players that you would have to give up to acquire that many closers in the draft. While I agree it may be difficult to win overall by throwing a category, I'm not sure others in that lge benefit if someone does throw a category.

And finally, yes, I do know how to use and interpret stats in the real world. Given your "degree", I am certain that you can grasp and test the validity of my findings yourself.

KJ Duke · Post by **KJ Duke** » Tue Aug 10, 2004 5:53 am

Originally posted by Dyv:

The teams in your proposed example can carry 4 closers - and therefore less SP. Not a significant factor as the league still has 32 closers in their starting rosters, just like all other leagues, etc... I don't think pitching variance has anything much to do with it.

I'm still willing to accept that there is a #1 league and a #13 league

Thanks for the calc work Zef... I actually thought it would be more randomized outside of pitching. Still - there's a reason I'm in first place in NY5 and still only 12th overall... lol

Thanks for your time and work there.

Dave [/QB]Dyv, well my lge is in the lower half as well and I did think coming out of the draft there were a few weaker teams in it, but there are also a lot of strong teams which I suspect is the case in every lge - IMO leading even the worst lge is very impressive given the level of competition.

And of course, its not over yet ... I'll probably run the numbers again late in the season to see how they come out. But if they come out as expected, particularly with Vegas leading by a strong margin, I am going to consider heading to Chicago next season.

(My early season thesis was that the commitment level of flying rather than driving to compete may lead to stronger top-to-bottom comp in Vegas -so far, that looks correct - not that there aren't NY owners that are just as good as any LV owner, but the odds of having more weak players in your lge may be better in NY-Chi).