Sports and Numbers

Northern Arizona at Kansas (Lawrence, KS)

PSAN-Related Player Ratings - Cumulative This Season

Note: ePSAN70 weighs recent games more, but cPSAN70 weighs all games equally

Player Analysis:
(largely based on ratings above)

Statistical Strengths and Weaknesses Analysis

Game Projections

(New Feature!)

This area features projections, not predictions.  The method assumes that the teams and players bring the same tendencies and performance into this game as they have in the games so far this season.  This method starts by computing the expected score using efficiency and tempo calculations.  It then applies every statistical category to the offenses and defenses based on a weighted average of the two teams' statistics.  For example, KU's shooting percentage is dictated by its own eFG% coming into the game as well as its opponent's defensive eFG% allowed.  The team favored to win will have the statistics weighted more heavily in its favor.  That way, the projection avoids assigning weight to the great shooting of a team that has merely played a weak schedule and shot well, for example.  Once the team statistics are determined, they are assigned to individual players based on a per-game weighted average.  So if there are 15 AST to divide up, the players with the highest AST/gm will get the most of those.  After that, the projection is rounded and then run through several iterations of adjustments to bring every statistic as close in line to the average projection as possible.  The end result is what you see below.  I will try to provide a better explanation in the future, and of course I will continue to assess the value of this feature as the season progresses.  I wanted to introduce this feature now so that most readers are familiar with it by the time the games are more meaningful and the stats are more accurate.

It would be ludicrous to assume that anyone could predict an exact boxscore that comes close to the actual result.  The best use of this is as a "baseline expectation."  That is, rather than saying that the boxscore is predicting that Jackson will score 13 points, it's best to say that Jackson performed below his average level going into this game if he scores below 13 points.  And even that would be inaccurate if the game turned out to be much slower tempo or Jackson played many more or fewer minutes than projected.

NOTE: There is very little game data so far, so these projections are much less useful than they will be several games down the road.  After several games, you won't see Arthur projected to make 1-of-1 3FG's, which is only the case because he's made one in three games and the rounding adjustments forced it on him in this projection.  That won't be the case much longer obviously.

Projected Boxscore 

Projected Four Factor Performance

• eFG% - NAU is projected to win this battle 59-52%.  Based on season-to-date performance, this should be no surprise.  If the "real KU" shows us, it's a different story.
• TURNOVERS - KU expected to dominate 32-15%
• OFFENSIVE REBOUNDING - KU projected to have 33-24% advantage.
• FREETHROW RATE (FTM/FGA) - 31-24% projected advantage for NAU.
• OVERALL - Huge advantge in TO's for KU to go with significantly more freethrow attempts overcomes a significant advantage in shooting by NAU.

Kansas Players Projection

• Highest Game Impact (cPSAN): Darnell Jackson
  
• Highest Efficiency (cPSAN70): Brandon Rush
  
• Most likely to play more efficiently than season-to-date (projected cPSAN70 - season cPSAN70): Darrell Arthur, Russell Robinson
  
• Most likely to play less efficiently than season-to-date: Jeremy Case, Rodrick Stewart
  

Sports and Numbers Projection: Kansas wins 91-73