All Forums > SimLeague Basketball > Theme Leagues > ODL - Team Rosters & Commentary
12/5/2013 9:25 PM
great discussion.  I'm pretty sure you all know where I stand.  I really do wish I could build more "feel" teams, but at the end of the day, my love of mathematics mixed with my competitiveness (along with a lot of down time at work sitting in front of a computer) wins out.
12/5/2013 9:52 PM
Posted by cmcafeeky on 12/5/2013 2:32:00 PM (view original):
good stuff scud-

I am definitely a 'new guy.'
and I am in the feel category.... but I want to win, so I listen and listen and listen some more-

about the % of x and % Y, and advanced statistics,
the coefficient of the maritime configuration of Gibraltar blah blah snooze,
and 'there is only one way to skin this cat', etc etc.
and 'this guy is an idiot !!!', and 'that guy doesn't know this sim !!!'.,
and' I have a championship to prove it!!!' and "SABLE SABLE SABLE !!! etc. etc.

all great stuff and highly entertaining -

but this is much more of an exercise in mathematics and statistical manipulation than in knowing basketball.

Its much more an exercise in number crunching, and an owner with experience
in working Sudoku would benefit more so than a guy brilliant in his knowledge of the game of basketball
or being able to evaluate basketball talent.

I grew up a basketball and sports geek hooked on
Strat-omatic & Avalon Hill & APBA games,

so I love this sim. and WIF -
I could have only dreamed about his during my childhood - great stuff in my opinion


you actually played a complete APBA Basketball game ???? !!!!!   
12/5/2013 9:56 PM
Good stuff scud and great read on the discussion. I was clearly more in the "feel" school of thought when I first started playing, but things like the D ratings and how they are applied and not being what they "should" (example Bill Russell) IMO started my path to the dark side. After my one losing record team in this league and literally 100's of hours of data anaylsis I am about as far away from "feel" now as one could get.

I've been watching the NBA since I was about 10 years old in the early 80's. In the 90's there was one season where I was in something like 15 fantasy leagues and I was the comish of over half of them, plus I went about a decade where I would have gambling action on the entire NBA board of games almost EVERY night of the season, I can tell you all about things like the playoff zig-zag theory, critcal numbers for totals and sides, the impact of playing a 4th road game in 5 nights etc. etc. etc. (Btw have not made a bet since about 2007).

The crazy thing is before about 3 months ago I had never even heard of some guys like Art Williams or had a clue how good Tom Boewrinkle was (actually really had not heard of him either before WIS). Now I find myself holding my breath that guys I used to not even know existed fall to me in a draft, and I'm a guy that really KNOWS the NBA.

IMO playing in this league each season becomes a huge advantage for every subsequent season you play. A new guy coming in is at a serious disadvantage because there are cerain guys he is unlikely to consider and others he doesn't realize are posion pills. The more you play the better your odds are of eventually winning, but the competition level in this league is absolutely fierce and we brought back in some of the most elite players on WIS for ODL 45, I'm looking forward to this one.
12/5/2013 10:44 PM (edited)
Btw I too am one of those guys who HATES it when one of my players shoots worse than IRL. IMO aside of the individual and team penalty the key is a comparison between your own cumalitive ast% and the D rating of your opponent plus a slight bump if at home. After looking at a lot, and I mean a lot of data, IMO you are pretty safe against most teams if you can keep 70% ast% on the floor at ALL times. Also IMO the max negative impact another team's D can have over the course of a season is in the 5% range, and if your own average D is at least 70D then then teams will be hard pressed to exceed RL efg.

There are of course a numer of additional factors in play each game and things like WIS stat corrections and over-corrections are going to happen at some point in 82 games.

But IMO if all else is close to equal a team that has an efg of 55-57 with a D of 70 will more times than not be better than a team with a D of 80-90 and efg of 52-53 and an ast% in the mid 60's. There was a serious trend towards D in this draft, so we will all get to see how well my theory plays out this time.

Btw, I had heard more than once here that the ast% floor was assumed to be about 50 on WIS. Again, IMO the floor will vary depending on how good the D you face is. A PERFECT example is what happen to icc this previous season:

Icc won 23 games with MJ and Gary Payton. It would "feel" like The Glove would be a good enough running mate for MJ to perform as well as he did in RL. Icc had a weighted team ast% in the upper 50's so by normal WIS assumptions again MJ should approach RL. Here is the end result:

MJ RL from the arc: 29/93 sim: 27/116 that 2 fewer makes in 23 more attempts, no big deal right, but look at him from the field:
MJ RL: 990/1837; sim: 857/1781, that means MJ made 133 FEWER baskets in 56 MORE attempts, he shot 53.9% IRL and 48.1% in the sim.

Yes there were other factors in play, but at its core IMO it was the lack of team ast% that doomed MJ when playing in a league that is now loaded with some of the best individual D seasons in NBA history. And I just double-checked icc did not run slowdown in any of the 8 games against me, so that was not the cause of this.

Feel can still work here IMO, but there are some statistical floors that winning teams need to be aware of and if they fall below them are no longer likely to be a winning team IMO.
12/5/2013 11:03 PM
natenoy, u might also want to consider that icc played half court and that MJ played more than 102% of his real life minutes.  Also, there are no "stat corrections".
12/5/2013 11:26 PM
Posted by ncmusician_7 on 12/5/2013 11:03:00 PM (view original):
natenoy, u might also want to consider that icc played half court and that MJ played more than 102% of his real life minutes.  Also, there are no "stat corrections".
Nc I'll concede that those are factors for MJ's historic underperformance. But my observations on ast% come more from my own experience. My team here in ODL 43 had Ray Allen, Shaq, Granger, and 48mpg of real PG's, but no one else had ast% and my weighted average was in the low 60's. It NEVER went below 50, but my entire squad missed their RL numbers by at least a couple of points. I ran two 6-point guys for 48mpg (used 95% Granger at C for 4-8mpg to protect my usage) which also minimized my exposure to the individual penalty, but I won 32 games and badja used the #1 overall pick on me for ODL 44. In ODL 44 I had 30mpg of NO real PG, but my starting lineup had a combined ast% of 68.5 and my weighted average was just over 70% because of Art Williams. More on Williams in a sec.
12/5/2013 11:45 PM
Not that Williams alone had a major impact but it was virtually impossible for him to be on the court with a combined team ast% below 70. Here is the result:

Arc RL 47/140; sim 55/136, not major but 8 more triples in 4 fewer attempts
Field RL 189/464; sim 210/466; so 21 more made shots in 2 more attempts.

Rl efg was 45.8% putting him the "likely to ignore" category, sim efg was 51%, which when added to his ast% and rebounding makes him a highly viable 6th-round back up PG for this league. The good news is that I took him again, so I'll see first hand if last season was a random fluke.

Btw you REALLY don't believe in WIS stat corrections? If it happens a few times then maybe its random observation of the extremes, but when you see a guy go 16/21 one game then the very next go 2/18 and you see it over and over and over again then IMO its a stat correction.
12/6/2013 12:04 AM
two words:  extreme outliers.

They happen too often & are the cause of many things that we falsely attribute with grand explanations - explanations that often make a ton of sense.  Corrections are one of these things.  This realization was a huge moment for me as I quit trying to do what "should" happen and started trying to build a squad that had the least chance of a) outliers occurring and b) those outliers determining the result of my game.

12/6/2013 12:51 AM
feel is bs and that is too bad - the site's main customer base is the baseball population and their approach to that engine has colored and influenced their approach to everything else - basketball is not money ball, there is too much flow and dynamism to the game but if you're trying to go by feel with this engine you're either being delusional or a fatalist - 55% is better than 50%.
12/6/2013 1:09 AM
Posted by ashamael on 12/6/2013 12:04:00 AM (view original):
two words:  extreme outliers.

They happen too often & are the cause of many things that we falsely attribute with grand explanations - explanations that often make a ton of sense.  Corrections are one of these things.  This realization was a huge moment for me as I quit trying to do what "should" happen and started trying to build a squad that had the least chance of a) outliers occurring and b) those outliers determining the result of my game.

word - do not underestimate the random- the completely unromantic approach to game winning in this engine is simple - line up your stat boxes to minimize negatives, maximize positives, drive efficiency and win the possession battle and above all DO NOT PAY ATTENTION to game by game outcomes - adopt a long term 10+ game by 10+ game approach (the only reason to check in is to manage fatigue issues) - assume wide scale statistical variance as a given DO NOT assume your tiny little coaching options have anything to do with a single game outcome THEY DO NOT. DO NOT ASSUME any 10 or 20 game pattern means anymore than 20 FLIPS OF A NICKLE (you can sometimes get 15 heads in a row, it's not magic, it's just random)

by the end of the season if you are lucky in: a) divisional draw b) conference draw and c) just plain lucky - you may have home court advantage and a bye

DO NOT ASSUME this means anything - the wreckage from the 60 win teams that '2nd round and out' are all around you, lift up your heads - if seble allows that 10 games are not enough to normalize statistical variation then YOU BETTER BE HOLDING THE F-ING RABBIT'S FOOT TIGHT because that series you are in is only 7 games long

take a moment with that, 7 GAMES is less than the 10+ games seble says are required for statistical normalization - CONSIDER that before you start patting yourself on the back for that 2nd round masterpiece - KNOW THIS: once you get to the play offs prayer is more likely to have a bearing on the outcome than your ****** little coaching options (about the only meaningful choices you can make are to max out minutes and pick tempo - the rest is COMPLETE BS)

SO

you know the rest - I guess when I say 'good luck' what we should all take away from this is simple : you can work the numbers to a certain extent and that will improve your chances BUT

that is all this all is really about: CHANCE - you might as well be playing black jack, drafting high efg and low TOs etc is an approach that has a whole lot in common with hitting on 16 and standing on 17 you're working the angles and narrowing the possible outcomes

you know just like in basketba- -er no that has nothing at all to do with this game, does it?

GOOD LUCK TO YOU ALL!!!
12/6/2013 1:11 AM
drops mike, out
12/6/2013 2:15 AM
nah... that can't be right, not all the time.
12/6/2013 2:39 AM
yep, sorry - the successful owners are the ones who understand how to shift the odds into their favor and so 'on average' they tend to win
12/6/2013 3:33 AM
Posted by felonius on 12/6/2013 2:39:00 AM (view original):
yep, sorry - the successful owners are the ones who understand how to shift the odds into their favor and so 'on average' they tend to win
While my opinion doesn't have quite as much... flare... as what the felon posted above, this is, at its core, what I try to do & what I've tried to help others realize.  If you stack your team to be favored in several key areas, it's like you're playing with a pair of 8 sided dice versus a pair of 6 sided dice.  Yeah, I've got a 7 & an 8 to play with... but I can still roll snake-eyes.  The odds are a little bit less that I will roll snake eyes than you, but I still will.  When that 55 win team loses to that 20 win team... that's exactly what's happening.  When that 65 win team goes out in round 2 to that 43 win team, that's generally what's happening.  When that 70 win team runs the table, hey, he was playing with 10-sided dice & managed to not roll snake-eyes 4 games in 7.

It sounds extreme. 
It sounds ugly.
But at its core, it is the truth.  That's why you see so much more parity in our draft leagues as opposed to OLs & progressives:  You have a bunch of really crafty vets that are all trying to stack the deck their way... in mostly the same fashion.  That means that most of us are truly playing with a pair of 6-sided dice.  There's a couple of guys with 4s, and maybe one or two with an 8.

How many people are addicted to dice & gambling?

"Well, the next throw has GOTTA be good."

12/6/2013 3:55 AM
Posted by felonius on 12/6/2013 2:39:00 AM (view original):
yep, sorry - the successful owners are the ones who understand how to shift the odds into their favor and so 'on average' they tend to win
that's not what i'm referring to, monk... what "can't be right, not all the time" refers to the bleekness of which you paint. to me, wis is not really dark art. or sinister, or cold hard facts and sobering equations, or even shifting the odds and likelys to one's favor. to me, it's like the morning newspaper. it comes everyday. it's there when i want it. it's laughable and dramatic. it's my hobby.  
of 13
All Forums > SimLeague Basketball > Theme Leagues > ODL - Team Rosters & Commentary

Terms of Use Customer Support Privacy Statement

Popular on WhatIfSports site: Baseball Simulation | College Basketball Game | College Football Game | Online Baseball Game | Hockey Simulation | NFL Picks | College Football Picks | Sports Games

© 1999-2014 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.