the high assist thing is so hit or miss.  I've had teams with 125% ast not really go much higher while teams with 90% ast have an astronomical surge.

I hold to the theory that wis is monkeying around on felonius.

Posted by ashamael on 5/13/2013 8:00:00 PM (view original):
the high assist thing is so hit or miss.  I've had teams with 125% ast not really go much higher while teams with 90% ast have an astronomical surge.

I hold to the theory that wis is monkeying around on felonius.

As crazy as that sounds it just might be true.

Well if  you ran half court with a mediocre eFG squad... It's not possible to win 50 games like that IMO @52 mil. (but I'm a better armchair simmer than I am at getting results).   Looking at DH's team, he was pretty low on starting rebounding and avg assists, but had stellar efg and made it to the finals. 

efg certainly trumps most all else but the team in question fared about the same at uptempo and just had more turnovers and all the extra assists did nothing in either case - meanwhile still dont know what to make of the steals thing

The team had the following weighted averages for the actual minutes used:
D 76.58
orb% 40.21
dreb% 101.11
ast% 95.73

All those things except assists did what they were supposed to.  The assists is a big disappointment.  You should have had a better bump, but that's why I don't think they matter much between 50 and 140. 

I don't even pay attention to steals and blocks anymore and I'm starting to not care much about defense!  Efg, 3's and rebounding and keep def avg around 60 seems to be the order of the day.  Low to's and fouls if you can get em seems to make a difference too.

low TOs were also in effect but got devalued by individual penalty on attempts (seemingly) and I still am gnawing at that steals thing - team was in the bottom 3rd of OP TO - not just no effective, almost opposite effect

You have to wonder if there are just gaping flaws in this thing. Its not like they do any testing or safeguading. Sim flaws are reported by users not admin.

I've had pippen teams have super low steals before and it just seems impossible. as you say like the opposite effect.

One of the things that stands out to me is the extremely low number of FTs made vs how many the opponents made. The opponents made 4.9 more FTs per game ans the team lost by about just less than 2 points per game on average. 1187 FTs made is less than 15 per game. The team only shot 68% from the line vs the opponents 76% form the line. Another huge problem for this team that hasn't been mentioned by the math experts here is that the team had over 21000 minutes drafted. That is a huge waste. I figured that was about a 7% overage. Almost everybody almost gets stuck with over 20000 minutes in this league. But an experienced owner here has to know that drafting over 21000 minutes is a recipe for disaster. Drafting 1613s FTM made and about 2300 FTAs in RL at over 21000 minutes is a critical error IMO.

My of my most recent successes in this league, and sixernuggets who I mentored, had I believe over 3000 FTAs in RL. My current team has almost 3,300 FTA's in RL with almost 2600 FT's made. Combine that with 19,707 minutes drafted, above average boards, excellent overall EFG% and above average defense and I think that its a winner. Sixernuggets championship team from last season had less than 19600 minutes drafted.

Guys most of the time there are over 200 points scored in a game. You only have to lose by 1 point. That's a 1/2 of 1 percent margin of error. If you don't address critical areas such as foul drawing and then draft way too many minutes the team won''t be strong in the $52 million league. Sorry for using such simple numbers and all. Your advanced calculations just don't seem to be addressing the real issues that the team had.

Posted by natenoy on 5/11/2013 5:50:00 PM (view original):
I spent 7-8 hours the last few days pouring over the stats from the last dh$52 trying to see what screwed you felonius. I have concluded:
(1) Either WIS really has flagged your account or there is a lot more randomness than we would like to believe
(2) Your foul rate (RL pf/g divided by mpg, with a weight for minutes used applied) has very high compared to the other teams in the league and you draw foul rate (rl fta's, divided by minutes, dived by 1.28 with the weight applied) was very low when compared to others. The sim net was apparent as your opponents attempted 348 more FTA's, also with your FT% being lower than ave your team had 405 fewer FTM's than opponents.
(3) Your RL poss/min and fga/min rates were also very low. Will talk about this more in next post.

Vance, as you can see I agree with you on the foul rate diff possibly playing a critical role. Not sure that I 100% agree with you on the minutes thing, some guys simply have better seasons with fewer minutes, for example I could spend another $1 mill or so for a West season that has 300 more minutes with less ast% and a and a lower efg at a higher usage. Imgoing with the cheaper West and that extra mill has to go somewhere.

But you are right on the minutes theory as are actually drafting the team, it's always best to track salary and minutes as the team is put together as to minimize waste at the end, but sometimeswasted minutes can become unavoidable once you are locked into your seven guys.

the minutes thing is almost meaningless by itself but indicative of other causal factors - that other guys in the league are getting more bang for their buck for instance - but are/were they? you'd have to go back and look at how all the other teams fared in that regard

and while the FTA/FTM thing probably was a factor it doesnt address either the 'steals opposite effect' or 'assists meant nothing' results

I think the simplest answer is best here, felonious - that steals and assists are reactive stats in the sim more than impactful ones, and as such often aren't worth the money. I think the sim just flat doesn't work how its supposed to.

Posted by felonius on 5/18/2013 3:19:00 PM (view original):
the minutes thing is almost meaningless by itself but indicative of other causal factors - that other guys in the league are getting more bang for their buck for instance - but are/were they? you'd have to go back and look at how all the other teams fared in that regard

and while the FTA/FTM thing probably was a factor it doesnt address either the 'steals opposite effect' or 'assists meant nothing' results

Yea it's a bang for the buck thing, look how much Wilt costs and his Cost per minute is high, but he's worth every penny and no one else can get that kind of production from a guy no matter the cost.

I will say this the data backs up vance, but we have to keep in mind that the better players cost a lot more per minute, so by default the best teams get more expensive minutes and have less left over at the end to purchase excess minutes.

Here is what I found by looking that the last $52 league:

Ave min for the middle ten playoff teams (drop the high and low): 20,374
Ave min for middle ten non-playoff teams: 20,878

Correlation between minutes and wins: -.8124

Playoff team ave minutes: 20.358
Non-playoff ave minutes: 20,946

Only two teams with fewer than 20,300 minutes missed the playoffs
No team with more than 20,899 made the playoffs 
Of the 14 teams that drafted more than 20,600 minutes only 4 made the playoffs
Of the 10 teams that drafted 20.355 minutes or fewer only 2 missed the playoffs

So, Vance you are right for the most part, but I think it all comes back to paying as much as you can on a per minute basis, and the fact that as a general theory the more you have to pay per minute the more production you will get per minute.

My two cents:

You drafted a malone team!

Everything in order EXCEPT:

Turnovers and fouls.

Boerwinkle and Webber are great, but for limited minute players they are foul monsters.

Let's get deeper into the numbers though.  Look at the value of the possessions.  You made 1.044 points per possession (not counting FT's), but your opposition is making 1.031.  You are ahead by 1.2 points per possession.  Now adding 185 TO's more than your opposition means you are giving 2.25 possessions up per game.  This is an extra 2.32 ppg which puts you down 1.1 ppg, and again this doesn't even touch the free throw problem.

Hindsight being 20/20 what can we say from the draft to see where this was going...Too many drafted fouls.  Williams, Tinsley, Ray, Rondo, Boerwinkle, and Tarpley were going to be a goon squad, and they were.  The fouls per 36 minutes were going to be maximized by the high volume free throw shooters other teams had.  Any player averaging more than eight FT attempts per game were going to kill you, and that list reads like a who's who of this league.  The quick and dirty list is:

Wilt
Kareem
Jordan
Moses
Big O
Robinson
Duncan
McAdoo
Shaq
Barkley
Mailman
Durant
LeBron
Kobe
Bellamy
Howard
Gilmore
Dream
Hawkins
Nowitzki
TMac
Dantley
Pierce
Wade
Zo


Important things to notice here as I based this on guys with more than 650 ftas. These are just 25 guys.  This isn't the entire list, and I did omit guys like Iverson, Tiny Archibald, West, Hondo, Baylor, Hayes, and others that are rarely used.  You likely faced one or sometimes two of these guys every game, and they literally KILLED you, and not to be to brash, but you got raped by the bigs on this list since your bigs where just clubbing people right and left.  The other thing is that none of those 25 are on your team.  You lost 4.94 points at the line.  Now you are down 6 ppg.  Your advantage in rebounding, which could easily be predicted, really trimmed this a lot, but losing the possession battle on TO's and Fouls was painful.

I know this was more than my two cents, but while I was gone i did two things...rediscovered my love of the NBA, and created a new stat.  The new stat allows me to see the game from a different perspective, and understand exactly what one point in a game means and the variables that go into making that occur.

Malone, care to share your stat?

The stat is Points Created Per Minute.

It takes into account the effect that any quantifiable stat has and boils it down to what a given player contributes to his team.  It is somewhat long, but it looks at what happens when a shot is made, a rebound is made, an assist is credited, a free throw is attempted, or a steal or block occurs.  Then it looks at the effect of missing a shot, a turnover, or a foul.  It takes those numbers in relation to team points per shots made to weigh the good things done verses the opponent points per shot made to weigh against negative things done.

To give you an idea at the All-Star break LeBron was the only player to average more than one point created per minute played at 1.026.  His next closest challenger was Durant at .82.