Antdoggydogg's 2K17 POE's Roster

[antdoggydogg]

This is my first time doing this, but I thought it would be worth it after all the positive feedback my Point of Emphasis Sliders got from the community last year. Along with providing POE's that result in real-life team statistics, I will also be uploading my roster which includes the following types of well researched edits, and for the sake of full transparency I've included all my reasoning and logic behind every change. I have so many more ways that I'd like to see the franchise evolve in the next few years and 2K does an amazing job every year trying to perfect this game.

• Head Coach & Assistant Coach Offensive, Defensive, Potential grades and Badges for current and historic teams. (Based off both career track record and recent sample size.)

• Updated Player Heights & Wingspans for current and historic teams.
  
  Spoiler

  Referred to DraftExpress's database and used "In Shoes" height to determine their height, rounded up or down based on the measurement. For historic players, thorough research was put into figuring out their correct heights.
  

• Updated Vertical for select modern and historic players.
  
  Spoiler

  Some players have generous vertical ratings, so I made my own formula on how to calculate vertical measurements:
  
  VERTICAL
  99 = 46.0 Inch Vertical
  90 = 42.8
  81 = 39.7
  72 = 36.6
  62 = 33.5
  53 = 30.3
  44 = 27.2
  35 = 24.1
  25 = 21 Inch Vertical
  
  I also discovered that experts says a player loses roughly 3 inches from his vertical every 10 years or so. With that said, I double checked all reoccurring players from historic rosters/current rosters and made sure their vertical ratings reflected that loss of athleticism. I wish the dream team was editable!! I found some strange bloopers like Shaq's vertical rating going higher when he went to LA. All this was fixed. I even did thorough research on guys like Wilt and Bill Russell to get an accurate rating.
  

• More accurate Draw Fouls ratings.
  
  Spoiler

  
  In MyLeague, the amount of free throw attempts a player draws is dependent on a multitude of factors:
  - Draw Fouls rating. (Love the addition of this as an attribute/skill and not a tendency.)
  - Shot/Touches tendency.
  - Minutes.
  
  Minutes are an obvious factor, but thanks to 2K adding Per 36 minute stats, it makes this much easier to research. I will be evaluating players free throw attempts based off of their per 36 minute simmed stats, and I will cross-check those averages with their real life per 36 minute stats via Basketball-Reference.com. So we’ve eliminated the minutes issue from this equation.
  2K’s ratings guru Mike Stauffer likes to rate things based off of a Hall-of-Fame/historical perspective. And I love the idea behind that. So I came up with my own “Hall-of-Fame” ratings scale for the Draw Fouls attribute which can also be applied to historic players. This scale is not the way I would like it to be, it was just created to adhere to the way foul stats are programmed in this game:
  
  Draw Foul Hall-of-Fame Rating Scale (Based off player's real life FTA per 36 minutes)
  12 = 99 (Shaq, Wilt, Jordan, Barkley)
  10.5 = 90
  9 = 81
  7.5 = 72
  6 = 62
  4.5 = 53
  3 = 44
  1.5 = 35
  0 = 25
  
  Sadly, despite my editing, players still won’t draw as many fouls as their real life counterparts. A player's shot/touches tendency directly affects how involved the player is in the offense, therefore affecting how many fouls he draws. You can give DeAndre Jordan a 99 Draw Fouls rating but he still won’t draw a realistic amount of fouls because he doesn't shoot enough to draw that many fouls. But in the end I’m not going to sacrifice the integrity of his shots tendencies in order to get realistic free throw stats.
  
  Yet even the scoring machines like James Harden don’t average enough free throws. No matter what I do, I haven’t been able to get James Harden to average even close to 9 or 10 free throws in a simmed season. Ultimately there seems to be a cap on this stat and the Draw Fouls rating needs to be reprogrammed to match the above scale. Actually, if we really want to be as precise as possible, creating a per-100 possessions scale would be a lot more accurate in terms of creating a global scale that applies to historic rosters as well (Because teams in the 60-80's played with a lot more possessions). In the past, other statistical categories were affected by these artificial caps, such as blocks and steals but that was fixed in 2K16. Free throws should be next.
  

• More accurate Commit Fouls tendencies.
  
  Spoiler

  
  Commit Foul Tendencies are another item on my checklist when editing rosters. Fouls affect many areas of a simulation/gameplay such as playing time due to foul trouble, defense, free throws, and ultimately the win-loss column. So it's important that all the players are up to date. I came up with my own formula that coincidentally seems to be very close to 2K’s. But every now and then some players are a little off so I go in and double check in case anyone is in need of an update.
  
  Commit Foul: (Based off player's Fouls per 36 minutes)
  95 = 6 (Jusuf Nurkic, Dewayne Dedmon)
  85 = 5.5
  75 = 4.5
  62 = 3.5
  50 = 2.5
  25 = 1
  0 = 0
  
  My formula is based off per 36 minute stats, so they can be used for historical players as well.

• Updated Overall Shot Tendency, 3-point and mid-range tendencies all-around.
  
  Spoiler

  
  I know a lot of data probably gets used in order to dictate a players overall shot tendency but the science isn't perfect yet so I messed with the shot tendencies to get results that mimic real life FGA's. Hopefully the math that goes into deciding these tendencies can keep being improved upon, one day I will sit down and try to figure out a correlation between advanced stats (like USG%, etc) and 2K overall shot tendencies so that we can find a solution that is something closer to an exact science so roster editing can be less of a trial and error process with future games.
  
  A lot of players shot charts are off, there was always a combination of too many close shots/mid-range shots and too few three pointers.
  

• Updated Drive tendencies for important players.
  
  Spoiler

  
  For the past 3 seasons, SportsVU Cameras that were installed in every NBA arena have been tracking player movement and collecting all sorts of previously unavailable statistics that are neatly organized on the NBA’s official Stats Website. One of the many aspects of the game that they track, is Drives.
  
  If we could create a per 36 minute scale based off the 3 year history of available SportsVU data (going on 4 years), we could apply this to Player Tendencies. Here is the scale I came up with:
  
  Drive Tendency: (Based off Per 36 minutes)
  100 = 13
  88 = 11.5
  75 = 10
  63 = 8.2
  50 = 6.5
  38 = 4.8
  25 = 3.2
  13 = 1.6
  0 = 0
  
  There are a lot of guards in the league that are relentless in their driving efforts. You can notice one household name, Chris Paul, is not high on that list. His game is very different from Russell Westbrook’s, Isaiah Thomas’s, or Eric Bledsoe’s. You see, over the last 3 seasons, Chris Paul has averaged 6.5 drives per 36 minutes. Using the above scale, that would make his Drive tendency a 50. Yet the official 2K roster has his driving tendency set to 90, which is as high, or higher than some of the other attacking guards on that list. This results in him taking less of his patented mid-range pull-ups during gameplay, because his erroneously high Driving tendency is telling him to attack the basket instead.
  
  Even after adjusting these tendencies I still did not see Chris Paul take a lot of pull up jump shots in 48 minute games. In real-life he averages 10 pull-ups a game yet in 2K I usually see him take about 2-4 on average. It’s not the most authentic depiction of his style of play.
  
  And on the opposite side of things, some players have their Drive Tendency set too low on the Official 2K Rosters. Over the last 3 seasons, Jeff Teague has averaged around 13 Drives per 36 minutes. His Drive tendency should be set at 100 or at least in the 90s. The official 2K roster has him at 70. That’s a pretty harsh difference.
  
  Damian Lillard averages around 10 drives per 36 minutes. His Drive tendency should be around 75 but instead he is set to a generously high 85.
  
  DeMarcus Cousins has averaged around 6.6 drives per 36 minutes over the last 2 seasons. His Drive Tendency should be set to around 52 but instead it's currently at 30. In fact, last season he averaged a career high 8 Drives per 36. That would actually put him at around 60 but I’ll just stick with the 2 season sample size instead.
  

• Updated Steals and blocks tendencies for all players, resulting in accurate big men stats in these categories.

• Updated intangibles ratings to help result in more realistic contract negotiation demands. Remember to offer Bird Rights option if you want these guys getting paid realistically!

• Updated Play-Types for players.

More selective and accurate Badge distribution across the league. I start by filtering the SportsVU data, I usually use a minimum number of possessions in order to weed out any outliers and so that I don’t end up with a hundred players from one season owning a particular badge. These badges are meant to separate the truly elite of the elite, so I usually aim for filters that result in around 60 players at most. That way these specific talents are more exclusive and the league doesn’t become oversaturated with Catch-and-Shoot snipers or Putback Kings, for example.

I noticed that historic teams seem to have players with Gold badges assigned to them just based off of their notoriety in history or some other misconceived reputation. Which leads to some players being unrealistically overpowering. I know Bill Russell was a star center that played in the old school, but that doesn’t mean he was a monster in the post and a bank shot expert. I’ve watched every available Bill Russell game that exists on the internet, the guy wasn’t very efficient on the block and just looked awkward with his hook shots. I don't think he deserves some of the offensive gold badges he currently has. Same goes with 1977 Dr. J who has been the most unstoppable player in this game for me. He averaged 32 points in 4 games against my 2008 Celtics defense, didn’t matter what I did while playing on Superstar Difficulty, turnaround jumper, hand in his face, the dude was going to score. Mostly due to his plethora of Gold badges in various scoring categories.

I like there to be some rhyme or reason for why certain players have certain badges, and with most historic teams the badges seem to have been handed out way too generously. So I usually look to the advanced stats in order to come up with a scale to award these badges to deserving players. Rhyme, reason and thorough research went into consideration when creating my custom Bronze, Silver, Gold and HOF scale for the following important badges:

• Offensive General
  
  Spoiler

  
  I needed to find an advanced stat that attempts to keep track of a player's contributions to his team's offense, and ideally I also wanted this stat to go as far back in time as possible. For a while I was happy using ESPN’s Real Plus/Minus but that data only goes back about 3 seasons. So I started using Basketball-Reference’s Offensive Box Plus/Minus. I believe this data goes back to about 1974-75 so that covers the majority of historic teams in the game aside from a couple 60’s teams.
  
  Using this data I wanted to figure out what the ideal hallmark of a true Offensive General is, which is defined in the game as someone who makes their teammates better while out on the floor. Originally I thought an OBPM score of 4 would be a fair mark, but I ran into personal conflicts with guys like Kyrie Irving. Who has an career average OBMP of 4.4, I didn’t feel right giving him this badge because I just don’t think he makes his teammates better while he is on the floor. He is a great scorer, but it didn’t make sense to me. So I raised my requirement to 4.5.
  
  When evaluating players, I usually look at a 3 year sample size or the totality of a player’s career in order to decide if they deserve it. Because I like to know if the greatness is/was sustained or if it was just a career year. This year I ended up with a total of 9 players in the league that truly make a difference to their team's offense, earning themselves the very exclusive Offensive General Badge: LeBron James, Isaiah Thomas, Chris Paul, Kyle Lowry, James Harden, Russell Westbrook, Damian Lillard, Stephen Curry, and Kevin Durant.

• Defensive Anchor
  
  Spoiler

  
  Much like the Offensive General badge, I used Defensive Box Plus/Minus to determine who the true Defensive Anchors are in the league. The requirement for this badge is 3.5, this is lower than the Offensive General badge because throughout history it is much harder to find players that can single handedly provide a large defensive boost to their teams, than it is to find players who can boost a team's offense just by being able to space the floor and shoot for instance.
  
  Most stats favor big men as having the most impact on Defense, DBPM is no different. Players like LeBron and Kawhi have reached the 3.5 DBPM mark in previous seasons but haven’t sustained it long enough to earn the badge in my opinion. Again, using a ~3 year sample size, here is the list of this year's Defensive Anchors: Nerlens Noel, DeAndre Jordan, Rudy Gobert, Joakim Noah, Andrew Bogut, Tim Duncan, Cole Aldrich, Draymond Green.

• Clutch

• Championship DNA

• Putback King
  
  Spoiler

  
  Average FG% on a Putback is 52%.
  
  (Minimum 45 possessions)
  Hall-Of-Fame = FG% of 70% or higher (Aldridge, Duncan)
  Gold = FG% of 65% or higher
  Silver = FG% of 60% or higher
  Bronze = FG% of 55% or higher

• PnR Maestro
  
  Spoiler

  
  Considering that the average Points Per Possession for a Pick and Roll Ball Handler is .78, here is how I decide the different tiers of the badge:
  
  (Minimum 100 possessions)
  Hall-of-Fame = PPP of 1.00 or higher (Curry)
  Gold = PPP of .95 or higher (Harden, Chris Paul)
  Silver = PPP of .90 or higher (Durant, Lillard)
  Bronze = PPP of .85 or higher (LeBron, Kyrie)

• Pick & Roller/Pick & Popper
  
  Spoiler

  
  Considering that the average PPP for a Pick and Roll Man is .97, here is how I decide the different tiers of the badge:
  
  (Minimum 65 possessions)
  Hall-of-Fame = PPP of 1.30 or higher (DeAndre Jordan, Hassan Whiteside)
  Gold = PPP of 1.20 or higher (LeBron)
  Silver = PPP of 1.10 or higher (Horford, Steven Adams)
  Bronze = PPP of 1.00 or higher (Durant, Anthony Davis)
  

• Corner Specialist
  
  Spoiler

  
  Since the league average corner 3-point percentage is around 38%. Using a 2 year sample size, I hand out Corner Specialist badges using the following scale:
  
  (Minimum 30 Left Corner Attempts & 30 Right Corner Attempts)
  Hall-Of-Fame = 3P% of 50% or higher (Curry)
  Gold = 3P% of 47% or higher (Redick)
  Silver = 3P% of 44% or higher (Klay, Kawhi, Korver)
  Bronze = 3P% of 41% or higher (Jamal Crawford, Wesley Johnson)

• Catch and Shoot
  
  Spoiler

  
  Using the Spot-Up Playtype on SportsVU. Here is my grading scale:
  
  (Minimum 125 possessions)
  Hall-of-Fame = EFG% of 75% or higher (Curry, Redick)
  Gold = EFG% of 65% or higher (Teague, Calderon)
  Silver = EFG% of 60% or higher (Harden, Kawhi, Korver)
  Bronze = EFG% of 55% or higher (Klay, Manu, McDermott)
  

• Dimer
  
  Spoiler

  
  For this badge I go off of Assist Percentage, similar to the other badges I like to use a recent sample size of at least the last 3 seasons. Here is my scale:
  
  HOF = AST% of 45% or higher
  Gold = AST% of 40% or higher
  Silver = AST% of 35% or higher
  Bronze = AST% of 30% or higher
  

• Defensive Stopper
  
  Spoiler

  
  Average opponent FG% differential is 0%.
  (Minimum 395 field goal attempts defended)
  
  Diff%<=-7 HOF
  Diff%<=-5 Gold
  Diff%<=-3 Silver
  Diff%<=-1 Bronze

• Rim Protector
  
  Spoiler

  
  I like the fact that 2K removed the Eraser badge. Let the block attributes/tendencies handle that. Rim protector Badge is a way of representing the stuff that doesn't show up in the box score. SportsVu has been tracking the effectiveness of Rim protectors for the past few seasons now. The average defender allows a FG% of 54%. Here is my scale, once again using a 2-3 year sample size:
  
  (Minimum of at least 200 Field Goal Attempts Defended near the rim)
  Hall-of-Fame = FG% of 43% or less (Gobert)
  Gold = FG% of 45% or less
  Silver = FG% of 47% or less
  Bronze = FG% of 49% or less
  

• Charge Card
  
  Spoiler

  
  Based off Charges Drawn Per 36 minutes.

• Brick Wall
  
  Spoiler

  
  SportsVU Hustle Stats
  
  Screen Assists per48 minutes:
  HOF >=9
  GOLD >=8
  SILVER >=7
  BRONZE >=6

• Hustle Rebounder
  
  Spoiler

  
  (Minimum of at least 200 Rebounds collected)
  Hall-of-Fame = TRB% of % or more (Drummond, Rodman)
  Gold = TRB% of % or more
  Silver = TRB% of % or more
  Bronze = TRB% of % or more
  

• Mid-Range Deadeye
  
• Long-Range Deadeye
  
• Limitless Range

Final Roster with all these changes, POE updates, and totally accurate stats can be found on Xbox Live:

User Name: Antdoggydogg
Roster Name: Antdoggydogg Accurate Stats

Here's a link to my work for anyone who wants to make PS4 edits: https://docs.google.com/spreadsheets...it?usp=sharing

Also attached my POE's spreadsheet.

Let me know how you guys enjoy it!

[cmebfresh]

Re: Antdoggydogg's 2K17 POE's Roster

Glad your back dude, Always

[mongo13clikke]

Re: Antdoggydogg's 2K17 POE's Roster

This is great. Im currently using turty11's roster on PS4 and while im happy with user vs. cpu experience im not so happy with cpu vs. cpu sim results. For example 76ers are 11-3 after 14 games... Im not blaming the roster for this, it has been mentioned in another thread here and seems to be more of a problem of the 2k simulation engine. I made a backup save and simulated to the end of the season and almost all teams were around .50. All teams have the same POE settings so I will try these settings and simulate again...

[ibrampton]

Re: Antdoggydogg's 2K17 POE's Roster

This is my first time doing this, but I thought it would be worth it after all the positive feedback my Point of Emphasis Sliders got from the community last year. Along with providing POE's that result in real-life team statistics, I will also be uploading my roster which includes the following types of well researched edits:

- Head Coach & Assistant Coach Offensive, Defensive and Potential grades for current and historic teams. (Based off both career track record and recent sample size.)
- Correct Player Heights & Wingspans for current and historic teams.
- Updated vertical for select modern and historic players.
- More accurate Draw Fouls ratings.
- More accurate Commit Fouls tendencies.
- Updated 3-point tendencies all-around.
- Updated Play-Types for players.
- More selective and accurate Badge distribution across the league. Rhyme, reason and thorough research went into consideration when creating my custom Bronze, Silver, Gold and HOF scale for the following important badges: Offensive General, Defensive Anchor, Clutch, Championship DNA, Putback King, PnR Maestro, Pick & Roller, Pick & Popper, Corner Specialist, Catch and Shoot, Dimer, Defensive Stopper, Rim Protector, Charge Card, Brick Wall, Hustle Rebounder, Mid-Range Deadeye, Long-Range Deadeye, Limitless Range.

Temporary POE's are here, I will update in a couple weeks.

that is awesome!

if your on PS4, it really sucks for us Xbox One users. Maybe you can provide an excel document with the details and we can release your project for XBOX as well?

[Rondesuan]

Re: Antdoggydogg's 2K17 POE's Roster

Why are some of the POEs blank at the top? Thanks for this btw I used your POEs last year

[antdoggydogg]

Re: Antdoggydogg's 2K17 POE's Roster

that is awesome!

if your on PS4, it really sucks for us Xbox One users. Maybe you can provide an excel document with the details and we can release your project for XBOX as well?

I'm actually on Xbox One, so you're good. I will release my spreadsheet along with explanations of all my rating/tendency/badge reasoning soon.

Why are some of the POEs blank at the top? Thanks for this btw I used your POEs last year

I realized the offensive style doesn't really affect sim stats in any way. And since POE's don't affect gameplay either, its basically pointless to spend time on.

[knicks2k]

Re: Antdoggydogg's 2K17 POE's Roster

Could you explain what your seeing with each category i.e. Offensive tempo, defensive aggression. Like how/which stats are effected. Just curious. Great work by the way. Used these last year as well.

[the_funkmeister]

Re: Antdoggydogg's 2K17 POE's Roster

Wow!

So glad you're back man. I have used your POEs from last year (moving the coaches around to their current clubs) -- using Turty's roster set and a game speed of 47, i'm getting realistic sim stats and scores on a full NBA season so far (24 games, 32-team control).

[antdoggydogg]

Re: Antdoggydogg's 2K17 POE's Roster

Could you explain what your seeing with each category i.e. Offensive tempo, defensive aggression. Like how/which stats are effected. Just curious. Great work by the way. Used these last year as well.

The quickest explanation would be:
Offensive tempo - affects pace heavily
Offensive rebounding/limit transition - affects defense slightly
Defensive Focus - Affects three pointers allowed/point paints allowed
Defensive Aggression - Affects foul stats and defensive efficiency rating
Defensive Rebounding - Affects pace

[2kCharles]

Re: Antdoggydogg's 2K17 POE's Roster

Quick question, how did you determine if you would usee limit transition or not? I looked at opponent fast breaks for the most part, to determine what the team got. For example, Philly on avg allows 18 ppg through the fast break...yet you have them on limit transition which wouldn't even get close to that many fast break points allowed per game. My assumption is you might have looked at how many Offensive boards a team gets per game, with Philly at 27th overall in offensive boards per game, but as long as their ratings are correct (i'd look at sonicmages rebound ratings to be specific, since 2k's rating system for rebounds is once again complete crap right now) they will still be right at the bottom of the league in offensive boards.

Just seeing a couple of those that are head scratches. Nets, LA, Phoenix all should be focusing on getting offensive boards.

Know you said you'd edit them, just wondering what had you pick those specific settings.

side note, never knew about the foul tendencies being affect by defensive aggressiveness. Going to do some testing for that.
Need to edit foul tendencies for players, but right now i'd say never go below neutral...lowest fouls per game you should see is 16, and thats almost a foul less then charlotte (the lowest in the league) and 1.5 lower then the next team (atlanta). Foul tendencies are all likely wrong and i need to edit them, but its likely it should never go below neutral. Yet I have multiple below 16, all of them with conservative.

[antdoggydogg]

Re: Antdoggydogg's 2K17 POE's Roster

Quick question, how did you determine if you would usee limit transition or not? I looked at opponent fast breaks for the most part, to determine what the team got. For example, Philly on avg allows 18 ppg through the fast break...yet you have them on limit transition which wouldn't even get close to that many fast break points allowed per game. My assumption is you might have looked at how many Offensive boards a team gets per game, with Philly at 27th overall in offensive boards per game, but as long as their ratings are correct (i'd look at sonicmages rebound ratings to be specific, since 2k's rating system for rebounds is once again complete crap right now) they will still be right at the bottom of the league in offensive boards.

Just seeing a couple of those that are head scratches. Nets, LA, Phoenix all should be focusing on getting offensive boards.

Know you said you'd edit them, just wondering what had you pick those specific settings.

side note, never knew about the foul tendencies being affect by defensive aggressiveness. Going to do some testing for that.
Need to edit foul tendencies for players, but right now i'd say never go below neutral...lowest fouls per game you should see is 16, and thats almost a foul less then charlotte (the lowest in the league) and 1.5 lower then the next team (atlanta). Foul tendencies are all likely wrong and i need to edit them, but its likely it should never go below neutral. Yet I have multiple below 16, all of them with conservative.

So let me explain my way of looking at this, because I think we're approaching it in different ways. For me, the first and most important stat I aim to fix is Pace. Because it predicates every stat, it either inflates them or deflates them based off how fast or slow the league/teams are. So I alter Offensive Tempo POE's based off that. But Defensive Rebounding also plays a part in affecting Pace.

Now the problem is, unless you are playing in an inaccurately slow paced league, 2K's simmed rebounding stats are always going to be too high when it comes to the defensive boards. So I choose to ignore per game numbers, because they aren't going to be realistic unless you completely butcher player's rebounding ratings across the league. And I actually think 2K's rebounding rating scale for individual players is spot on when you approach it by the "hall-of-fame" rating scale that they introduced last year. I decided to create my own scale one day and found that it's pretty much identical to 2K's so I'm on board with that. They go off of advanced stats not per game numbers, Dennis Rodman sets the benchmark at 99 because he has recorded the highest defensive rebound percentages in history. (Lets not turn this into a wilt chamberlain, bill russell debate, just using an example)

So with that said, my goal is to get teams defensive rebounding percentages to be accurate, not their per game averages. That goes for offensive rebounding too, which I think also has a small effect on the teams defensive efficiency stats. (I'm not sure if you're referring to in-game fast break points or simmed stats but POE's do not affect the actual game, just simmed stats.) You might not get the right per game numbers, but the Pace of play in the league will be right and thats the most important stat because it effects every other number.

The Defensive Focus POE's are decided by looking at how many three pointers a team allows per game, If you set them to Limit Threes, you will see a difference in their simmed stats. The goal with deciding how aggressive a team should be is completely based off their real life defensive efficiency, not their fouls per game. I want to strive to get their efficiency numbers correct because ultimately that leads to more realistic W/L simulations. I have edited all commit fouls tendencies for individual players, which will be a part of my roster that I release soon.

Hope this explains my thought process, I know people really care about per game rebounding numbers for teams, but its not my main area of concern as long as the individual players are averaging the right stats and the teams defensive rebounding %'s are on-point. My way might not be for everyone, in brief here is the order in which I approach POE's:

1) Pace
2) Defensive Rebounding Percentage
3) Offensive Rebounding Percentage
4) Three Pointers Allowed Per Game
5) Defensive Efficiency

[Barncore]

Re: Antdoggydogg's 2K17 POE's Roster

I hope other roster makers incorporate this into their rosters

[antdoggydogg]

Re: Antdoggydogg's 2K17 POE's Roster

I have a huge update that I think a lot of you would enjoy. If anyone sees this, please spread the word, I put months of work into this. I know its February but its hard to juggle my actual job and this extreme hobby of mine that I treat like a job.

This is my first time doing this, but I thought it would be worth it after all the positive feedback my Point of Emphasis Sliders got from the community last year. Along with providing POE's that result in real-life team statistics, I will also be uploading my roster which includes the following types of well researched edits, and for the sake of full transparency I've included all my reasoning and logic behind every change. I have so many more ways that I'd like to see the franchise evolve in the next few years and 2K does an amazing job every year trying to perfect this game.

• Head Coach & Assistant Coach Offensive, Defensive and Potential grades for current and historic teams. (Based off both career track record and recent sample size.)

• Correct Player Heights & Wingspans for current and historic teams.

• Updated vertical for select modern and historic players.

• More accurate Draw Fouls ratings.
  
  Spoiler

  
  In MyLeague, the amount of free throw attempts a player draws is dependent on a multitude of factors:
  - Draw Fouls rating. (Love the addition of this as an attribute/skill and not a tendency.)
  - Shot/Touches tendency.
  - Minutes.
  
  Minutes are an obvious factor, but thanks to 2K adding Per 36 minute stats, it makes this much easier to research. I will be evaluating players free throw attempts based off of their per 36 minute simmed stats, and I will cross-check those averages with their real life per 36 minute stats via Basketball-Reference.com. So we’ve eliminated the minutes issue from this equation.
  2K’s ratings guru Mike Stauffer likes to rate things based off of a Hall-of-Fame/historical perspective. And I love the idea behind that. So I came up with my own “Hall-of-Fame” ratings scale for the Draw Fouls attribute which can also be applied to historic players. This scale is not the way I would like it to be, it was just created to adhere to the way foul stats are programmed in this game:
  
  Draw Foul Hall-of-Fame Rating Scale (Based off player's real life FTA per 36 minutes)
  12 = 99 (Shaq, Wilt, Jordan, Barkley)
  10.5 = 90
  9 = 81
  7.5 = 72
  6 = 62
  4.5 = 53
  3 = 44
  1.5 = 35
  0 = 25
  
  Sadly, despite my editing, players still won’t draw as many fouls as their real life counterparts. A player's shot/touches tendency directly affects how involved the player is in the offense, therefore affecting how many fouls he draws. You can give DeAndre Jordan a 99 Draw Fouls rating but he still won’t draw a realistic amount of fouls because he doesn't shoot enough to draw that many fouls. But in the end I’m not going to sacrifice the integrity of his shots tendencies in order to get realistic free throw stats.
  
  Yet even the scoring machines like James Harden don’t average enough free throws. No matter what I do, I haven’t been able to get James Harden to average even close to 9 or 10 free throws in a simmed season. Ultimately there seems to be a cap on this stat and the Draw Fouls rating needs to be reprogrammed to match the above scale. In the past, other statistical categories were affected by these artificial caps, such as blocks and steals but that was fixed in 2K16. Free throws should be next.
  

• More accurate Commit Fouls tendencies.
  
  Spoiler

  
  Commit Foul Tendencies are another item on my checklist when editing rosters. Fouls affect many areas of a simulation/gameplay such as playing time due to foul trouble, defense, free throws, and ultimately the win-loss column. So it's important that all the players are up to date. I came up with my own formula that coincidentally seems to be very close to 2K’s. But every now and then some players are a little off so I go in and double check in case anyone is in need of an update.
  
  Commit Foul: (Based off player's Fouls per 36 minutes)
  95 = 6 (Jusuf Nurkic, Dewayne Dedmon)
  85 = 5.5
  75 = 4.5
  62 = 3.5
  50 = 2.5
  25 = 1
  0 = 0
  
  My formula is based off per 36 minute stats, so they can be used for historical players as well.

• Updated 3-point and mid-range tendencies all-around.

• Updated Drive tendencies for important players.
  
  Spoiler

  
  For the past 3 seasons, SportsVU Cameras that were installed in every NBA arena have been tracking player movement and collecting all sorts of previously unavailable statistics that are neatly organized on the NBA’s official Stats Website. One of the many aspects of the game that they track, is Drives.
  
  If we could create a per 36 minute scale based off the 3 year history of available SportsVU data (going on 4 years), we could apply this to Player Tendencies. Here is the scale I came up with:
  
  Drive Tendency: (Based off Per 36 minutes)
  100 = 13
  88 = 11.5
  75 = 10
  63 = 8.2
  50 = 6.5
  38 = 4.8
  25 = 3.2
  13 = 1.6
  0 = 0
  
  There are a lot of guards in the league that are relentless in their driving efforts. You can notice one household name, Chris Paul, is not high on that list. His game is very different from Russell Westbrook’s, Isaiah Thomas’s, or Eric Bledsoe’s. You see, over the last 3 seasons, Chris Paul has averaged 6.5 drives per 36 minutes. Using the above scale, that would make his Drive tendency a 50. Yet the official 2K roster has his driving tendency set to 90, which is as high, or higher than some of the other attacking guards on that list. This results in him taking less of his patented mid-range pull-ups during gameplay, because his erroneously high Driving tendency is telling him to attack the basket instead.
  
  Even after adjusting these tendencies I still did not see Chris Paul take a lot of pull up jump shots in 48 minute games. In real-life he averages 10 pull-ups a game yet in 2K I usually see him take about 2-4 on average. It’s not the most authentic depiction of his style of play.
  
  And on the opposite side of things, some players have their Drive Tendency set too low on the Official 2K Rosters. Over the last 3 seasons, Jeff Teague has averaged around 13 Drives per 36 minutes. His Drive tendency should be set at 100 or at least in the 90s. The official 2K roster has him at 70. That’s a pretty harsh difference.
  
  Damian Lillard averages around 10 drives per 36 minutes. His Drive tendency should be around 75 but instead he is set to a generously high 85.
  
  DeMarcus Cousins has averaged around 6.6 drives per 36 minutes over the last 2 seasons. His Drive Tendency should be set to around 52 but instead it's currently at 30. In fact, last season he averaged a career high 8 Drives per 36. That would actually put him at around 60 but I’ll just stick with the 2 season sample size instead.
  

• Updated Steals and blocks tendencies for all players, resulting in accurate big men stats in these categories.

• Updated Play-Types for players.

More selective and accurate Badge distribution across the league. I start by filtering the SportsVU data, I usually use a minimum number of possessions in order to weed out any outliers and so that I don’t end up with a hundred players from one season owning a particular badge. These badges are meant to separate the truly elite of the elite, so I usually aim for filters that result in around 60 players at most. That way these specific talents are more exclusive and the league doesn’t become oversaturated with Catch-and-Shoot snipers or Putback Kings, for example.

I noticed that historic teams seem to have players with Gold badges assigned to them just based off of their notoriety in history or some other misconceived reputation. Which leads to some players being unrealistically overpowering. I know Bill Russell was a star center that played in the old school, but that doesn’t mean he was a monster in the post and a bank shot expert. I’ve watched every available Bill Russell game that exists on the internet, the guy wasn’t very efficient on the block and just looked awkward with his hook shots. I don't think he deserves some of the offensive gold badges he currently has. Same goes with 1977 Dr. J who has been the most unstoppable player in this game for me. He averaged 32 points in 4 games against my 2008 Celtics defense, didn’t matter what I did while playing on Superstar Difficulty, turnaround jumper, hand in his face, the dude was going to score. Mostly due to his plethora of Gold badges in various scoring categories.

I like there to be some rhyme or reason for why certain players have certain badges, and with most historic teams the badges seem to have been handed out way too generously. So I usually look to the advanced stats in order to come up with a scale to award these badges to deserving players. Rhyme, reason and thorough research went into consideration when creating my custom Bronze, Silver, Gold and HOF scale for the following important badges:

• Offensive General
  
  Spoiler

  
  I needed to find an advanced stat that attempts to keep track of a player's contributions to his team's offense, and ideally I also wanted this stat to go as far back in time as possible. For a while I was happy using ESPN’s Real Plus/Minus but that data only goes back about 3 seasons. So I started using Basketball-Reference’s Offensive Box Plus/Minus. I believe this data goes back to about 1974-75 so that covers the majority of historic teams in the game aside from a couple 60’s teams.
  
  Using this data I wanted to figure out what the ideal hallmark of a true Offensive General is, which is defined in the game as someone who makes their teammates better while out on the floor. Originally I thought an OBPM score of 4 would be a fair mark, but I ran into personal conflicts with guys like Kyrie Irving. Who has an career average OBMP of 4.4, I didn’t feel right giving him this badge because I just don’t think he makes his teammates better while he is on the floor. He is a great scorer, but it didn’t make sense to me. So I raised my requirement to 4.5.
  
  When evaluating players, I usually look at a 3 year sample size or the totality of a player’s career in order to decide if they deserve it. Because I like to know if the greatness is/was sustained or if it was just a career year. This year I ended up with a total of 9 players in the league that truly make a difference to their team's offense, earning themselves the very exclusive Offensive General Badge: LeBron James, Isaiah Thomas, Chris Paul, Kyle Lowry, James Harden, Russell Westbrook, Damian Lillard, Stephen Curry, and Kevin Durant.

• Defensive Anchor
  
  Spoiler

  
  Much like the Offensive General badge, I used Defensive Box Plus/Minus to determine who the true Defensive Anchors are in the league. The requirement for this badge is 3.5, this is lower than the Offensive General badge because throughout history it is much harder to find players that can single handedly provide a large defensive boost to their teams, than it is to find players who can boost a team's offense just by being able to space the floor and shoot for instance.
  
  Most stats favor big men as having the most impact on Defense, DBPM is no different. Players like LeBron and Kawhi have reached the 3.5 DBPM mark in previous seasons but haven’t sustained it long enough to earn the badge in my opinion. Again, using a ~3 year sample size, here is the list of this year's Defensive Anchors: Nerlens Noel, DeAndre Jordan, Rudy Gobert, Joakim Noah, Andrew Bogut, Tim Duncan, Cole Aldrich, Draymond Green.

• Clutch

• Championship DNA

• Putback King
  
  Spoiler

  
  Average FG% on a Putback is 52%.
  
  (Minimum 45 possessions)
  Hall-Of-Fame = FG% of 70% or higher (Aldridge, Duncan)
  Gold = FG% of 65% or higher
  Silver = FG% of 60% or higher
  Bronze = FG% of 55% or higher

• PnR Maestro
  
  Spoiler

  
  Considering that the average Points Per Possession for a Pick and Roll Ball Handler is .78, here is how I decide the different tiers of the badge:
  
  (Minimum 100 possessions)
  Hall-of-Fame = PPP of 1.00 or higher (Curry)
  Gold = PPP of .95 or higher (Harden, Chris Paul)
  Silver = PPP of .90 or higher (Durant, Lillard)
  Bronze = PPP of .85 or higher (LeBron, Kyrie)

• Pick & Roller/Pick & Popper
  
  Spoiler

  
  Considering that the average PPP for a Pick and Roll Man is .97, here is how I decide the different tiers of the badge:
  
  (Minimum 65 possessions)
  Hall-of-Fame = PPP of 1.30 or higher (DeAndre Jordan, Hassan Whiteside)
  Gold = PPP of 1.20 or higher (LeBron)
  Silver = PPP of 1.10 or higher (Horford, Steven Adams)
  Bronze = PPP of 1.00 or higher (Durant, Anthony Davis)
  

• Corner Specialist
  
  Spoiler

  
  Since the league average corner 3-point percentage is around 38%. Using a 2 year sample size, I hand out Corner Specialist badges using the following scale:
  
  (Minimum 30 Left Corner Attempts & 30 Right Corner Attempts)
  Hall-Of-Fame = 3P% of 50% or higher (Curry)
  Gold = 3P% of 47% or higher (Redick)
  Silver = 3P% of 44% or higher (Klay, Kawhi, Korver)
  Bronze = 3P% of 41% or higher (Jamal Crawford, Wesley Johnson)

• Catch and Shoot
  
  Spoiler

  
  Using the Spot-Up Playtype on SportsVU. Here is my grading scale:
  
  (Minimum 125 possessions)
  Hall-of-Fame = EFG% of 75% or higher (Curry, Redick)
  Gold = EFG% of 65% or higher (Teague, Calderon)
  Silver = EFG% of 60% or higher (Harden, Kawhi, Korver)
  Bronze = EFG% of 55% or higher (Klay, Manu, McDermott)
  

• Dimer
  
  Spoiler

  
  For this badge I go off of Assist Percentage, similar to the other badges I like to use a recent sample size of at least the last 3 seasons. Here is my scale:
  
  HOF = AST% of 45% or higher
  Gold = AST% of 40% or higher
  Silver = AST% of 35% or higher
  Bronze = AST% of 30% or higher
  

• Defensive Stopper
  
  Spoiler

  
  Average opponent FG% differential is 0%.
  (Minimum 395 field goal attempts defended)
  
  Diff%<=-7 HOF
  Diff%<=-5 Gold
  Diff%<=-3 Silver
  Diff%<=-1 Bronze

• Rim Protector
  
  Spoiler

  
  I like the fact that 2K removed the Eraser badge. Let the block attributes/tendencies handle that. Rim protector Badge is a way of representing the stuff that doesn't show up in the box score. SportsVu has been tracking the effectiveness of Rim protectors for the past few seasons now. The average defender allows a FG% of 54%. Here is my scale, once again using a 2-3 year sample size:
  
  (Minimum of at least 200 Field Goal Attempts Defended near the rim)
  Hall-of-Fame = FG% of 43% or less (Gobert)
  Gold = FG% of 45% or less
  Silver = FG% of 47% or less
  Bronze = FG% of 49% or less
  

• Charge Card
  
  Spoiler

  
  Based off Charges Drawn Per 36 minutes.

• Brick Wall
  
  Spoiler

  
  SportsVU Hustle Stats
  
  Screen Assists per48 minutes:
  HOF >=9
  GOLD >=8
  SILVER >=7
  BRONZE >=6

• Hustle Rebounder
  
  Spoiler

  
  (Minimum of at least 200 Rebounds collected)
  Hall-of-Fame = TRB% of % or more (Drummond, Rodman)
  Gold = TRB% of % or more
  Silver = TRB% of % or more
  Bronze = TRB% of % or more
  

• Mid-Range Deadeye
  
• Long-Range Deadeye
  
• Limitless Range

Final Roster with all these changes, POE updates, and totally accurate stats can be found on Xbox Live:

User Name: Antdoggydogg
Roster Name: Antdoggydogg Accurate Stats

Let me know how you guys enjoy it! I updated the original post as well.

[DC]

Re: Antdoggydogg's 2K17 POE's Roster

Question for POE for HOU and BOS

Both launch many 3's but both are lead by one dude that takes the majority of shots.

How do you approach this?

Play Through Star or Look for Shooters?

[Dudabskt]

Re: Antdoggydogg's 2K17 POE's Roster

Hey man!!! Congratulations on your work!! Will this be avaliable on ps4???