1. Introduction

It takes me about 5 solid months from start to finish to get through an NBA Sim Tournament (and, my god, I hope that length starts to shorten). I'd estimate that the first 40% of the time is spent doing research: figuring out how I will be rating players, building rosters, simulating attributes, etc. The next 40% of the time is devoted to actually simulating the games on Playstation 3. This takes way too long, as the only way of entering in any kind of value (whether it be for a team slider, a player attribute, even a player's jersey number) is by pressing left or right on the control pad to increase or descrease the option by 1. Very painstaking, but I'm looking into ways of speeding that process up as well. The final 20% of the project is spent recording all of the data, entering it into my database, and presenting the results on this web site.

The following is just a brief overview of all the changes that were made between the 2007 NBA Sim Tournament and the 2008 NBA Sim Tournament. For nauseating, in-depth explination as to the whys and hows behind these changes please check out the other 2008 FAQs:

• How are Team Rosters determined?

• How are Player Attributes and Tendencies determined?

• How are Team Sliders determined?

 

2. PANDEX In, TENDEX Out

Halfway through my 2007 NBA Sim Tournament I realized a huge bias in the way that I rated players. For all player comparisons, roster determinations and attribute balancing, I used a "holy grail" formula called TENDEX. Essentially, a "holy grail" formula is any one that attempts to consolidate everything a player does on the basketball court into one number. This number can then be used to say (for example) "well, Wilt Chamberlain is better than Bill Russell because he is valued at 25 widgets, while Russell is only 21 widgets." For a brief description as to why I chose TENDEX over other "holy grail" formulas, check out this really old post of mine.

TENDEX may covert all things a player does on a court into one number but one thing it does not record is the setting the game was played in. The 50's were slow, the 80's were fast, the 90's were physical battles of attrition - and every team plays at slightly different paces each year. Therefore, the TENDEX of players from different eras and different teams can be inflated or deflated just based on the rules from that time or style from that team! PANDEX flattens out these fluxuations and removes historical/syle bias by standardizing the pace of the game at 100 possessions. PANDEX is used anywhere TENDEX was last year - selecting which Legends will appear on a team, determining MVP of each game, etc.

 

3. Team Roster Selection Changes

Besides using PANDEX instead of TENDEX as my ultimate comparison device, I've made some other tweaks to the Team Roster Selection process.

 

Twelve Players > Ten Players

I've increased the Team Rosters from ten Legends to twelve. In both NBA 2K7 and NBA 2K8, each team must have at least 12 players on it. For the 2007 NBA Sim Tournament, I added my 10 Legends and then kept the 2 shortest players left from that team's original roster. I then depowered those two individuals so that they would have the lowest value for each Attribute (50). I foolishly thought that having two under 6-foot, terrible players on each team along with historical Legends from the past would mean that they would never see the light of day on the court. Unfortunately, this wasn't so. These "leftover" players often saw quite a bit of time depending on injuries and foul trouble and I didn't want them to be any sort of deciding factor in games. The increased roster number for the 2008 NBA Sim Tournament eliminates the "leftover" affect.

 

Natural Positions Needed

In the 2007 Tournament, Wes Unseld (who played center in the NBA) was the starting shooting guard for the Washington Wizards. Ugh. This was due to a self-imposed rule, limiting a team's starting five players to also be their historically best five players. I originally felt that teams like Houston - who have a long history of outstanding centers - would be penalized if some of their better players were backups and playing shorter minutes. However, the plan backfired and I realized there needed to be a change. So, for the 2008 Tournament, Legends can only play a position they actually played in real life (thanks to position data from basketball-reference.com).

 

4. Player Attribute/Tendency Changes

 

New Attributes

NBA 2K8 introduces 9 new Attributes to each player. I go into further detail about how they affect a Legend's performance on the court and how they are determined in the "How are Player Attributes and Tendencies determined?" 2008 FAQ. But, just for reference, these 9 additional Attributes are: Strength, Vertical, Shoot Off Dribble, Shoot In Traffic, Quickness, Hustle, Hands, On Ball Defense, and Standing Dunk.

 

Retooled Counterparting

Counterparting is the first step I take when simulating a Legend. Basically, I find where a player is ranked in certain statistical categories during his Historical Season and then grant him the corresponding Attributes and Tendencies of these categories based on who was ranked similarly during the latest NBA season. It's actually a little tickier than that, but for this FAQ, this description is adequate. Anyway, whereas I used seven different Counterparting categories to simulate players in 2007, I've upped that to 10 categories for 2008. Here they are with the Attribute and Tendencies they affect:

 

Position Counterpart = Position Counterpart + Field Goal Attempts Counterpart + Fouling Counterpart

The meat-and-potatoes Counterpart, Position Counterpart, still handles a huge chunk of Attribute and Tendency simulation. However, I felt I could more accurately simulate players by removing the Shot Tendency and Commits Foul Tendency from the Position Counterpart equation and, instead, make two new Counterparts. Again, this is all explained with a lot more detail in the "How are Player Attributes and Tendencies determined?" 2008 FAQ.

Position Counterpart
Year	Attributes/Tendencies Affected
2007	Attributes: Layup, Dunk, Low Post Offense, Speed Tendencies: Shot Tendency, Close Shot Tendency, Mid Shot Tendency, Drive The Lane, Attempts Dunk, Back To Basket, Triple Threat, Shoots Fadeaways, Shoots Hookshots, Attempts Putback, Foul
2008	Attributes: Layup, Dunk, Low Post Offense, Speed Tendencies: Close Shot Tendency, Mid Shot Tendency, Drive The Lane, Attempts Dunk, Back To Basket, Triple Threat, Shoots Fadeaways, Shoots Hookshots, Attempts Putback

 

Field Goal Attempts Counterpart
Year	Attributes/Tendencies Affected
2007	Didn't Exist
2008	Attributes: N/A Tendencies: Shot Tendency

 

Fouling Counterpart
Year	Attributes/Tendencies Affected
2007	Didn't Exist
2008	Attributes: N/A Tendencies: Foul

 

Scoring Counterpart

No change.

 

Rebounding Counterpart

No change.

 

Assisting Counterpart

No change.

 

3-Point Counterpart = 3-Point Counterpart + 3-Point Attempts Counterpart

For the 2007 NBA Sim Tournament, the 3-Point Counterpart determined both how well a player shot from beyond the arc as well as how often he did so. Unfortunately, this made quite a mess with a few players such as Ben Wallace. From his Historical Season, Wallace shot 1 for 6

3-Point Counterpart
Year	Attributes/Tendencies Affected
2007	Attributes: Shot (3-Point) Tendencies: 3-Point Shot Tendency
2008	Attributes: Shot (3-Point) Tendencies: N/A

 

3-Point Attempts Counterpart
Year	Attributes/Tendencies Affected
2007	Didn't Exist
2008	Attributes: N/A Tendencies: 3-Point Shot Tendency

 

Blocking Counterpart

No change.

 

Stealing Counterpart

No change.