PAVT Prorated Average Value Theory Close
PAVT is an enhanced version of AVT that uses average rankings to amass only those points that a ranked player deserves from the statistical averages.  Just like AVT, key statistics for the past 3 to 5 years are combined with your league scoring rules to determine the average number of fantasy points scored.  The number of fantasy points scored by the baseline player at each position is then subtracted from each player.  PAVT then deviates from AVT by using the average rank of players and not the sequential ranking to determine which player fits into each slot.  For example, only a player with a ranking of 1.0 deserves all of the points available in slot number one.  The prorated points for each slot can be awarded to none or multiple ranked players.
 
The Math
Steps 1 through 4 are identical to AVT.
  1. Determine the total number of fantasy points scored each season by the top players over the past 3-5 years.
    (Example of top three RBs for each of the past three years)
    Year Player Calculation Points
    2001 Faulk, Marshall 21TDs * 6pt + 3Fum * -2pts + 2147Yds * .10pts 335
    2001 Holmes, Priest 10TDs * 6pt + 3Fum * -2pts + 2169Yds * .10pts 271
    2001 Green, Ahman 11TDs * 6pts + 4Fum * -2pts + 1981Yds * .10pts 256
    2002 Holmes, Priest 24TDs * 6pts + 1Fum * -2pts + 2287Yds * .10pts 371
    2002 William, Ricky 17TDs * 6pts + 5Fum * -2pts + 2216Yds * .10pts 314
    2002 Tomlinson, Ladainian 15TDs * 6pt + 1Fum * -2pts + 2172Yds * .10pts 305
    2003 Holmes, Priest 27TDs * 6pts + 1Fum * -2pts + 2110Yds * .10pts 371
    2003 Tomlinson, Ladainian 17TDs * 6pt + 0Fum * -2pts + 2370Yds * .10pts 339
    2003 Green, Ahman 20TDs * 6pts + 5Fum * -2pts + 2259Yds * .10pts 336
     
  2. Determine the average number of fantasy points scored by the top players over the past 3-5 years.
    Slot Calculation Points
    RB1 (335Pts + 371Pts + 371Pts) / 3 359
    RB2 (271Pts + 314Pts + 339Pts) / 3 308
    RB3 (256Pts + 305Pts + 336Pts) / 3 299

  3. Determine the baseline player for each position.  This needs to be done very carefully because it is what allows one position to be compared to another.  In this example, the last starter in a league of 12 teams is used as the baseline. learn more about baselines...
    Position Worst Starter Baseline Points
    RB 24th RB 177

  4. Subtract the baseline points determined in step 3 from each player and sort by the resulting value.
    Slot Calculation Value
    RB1 359 - 177 182
    RB2 308 - 177 131
    RB3 299 - 177 122

  5. Prorate the value available at each slot based on average rankings.  Each player receives 100% of the value available at the slot for which they fully qualify and the percentage of the next higher slot that their ranking indicates they deserve.  In this example, the 4th and 5th RB AVT values are 112 and 96.
    Rank Player Average Rank Calculation Value
    1 Holmes, Priest 1.53 131 + (2 - 1.53) * (182 - 131) 155
    2 Tomlinson, Ladainian 1.69 131 + (2 - 1.69) * (182 - 131) 147
    3 Green, Ahman 4.53 96 + (5 - 4.53) * (112 - 96) 104
 
Positives
  • Allows players at different positions to be compared to one another
  • Requires less maintenance than RLV (rankings vs. projections)
  • Avoids over-allocation of statistics that often occurs when making RLV projections
  • Reduces AVT exaggeration of the difference between the strongest and weakest players
  • Merges statistics and current-year rankings without demanding that past-year performances repeat themselves.  The following table shows the differing results for AVT and PAVT using the numbers above:

     

    		 AVT

    PAVT
    Player Value Player Value
    1 Holmes, Priest 182    
    2 Tomlinson, Ladainian 131 Holmes, Priest
    Tomlinson, Ladainian

    155
    147
    3 Green, Ahman 122    
    4     Green, Ahman

    104
Negatives
  • Arithmetic is more complex than AVT or RLV
  • Very dependent upon the selection of baseline players for each position.
During the draft
Like most player value lists, PAVT does not necessarily specify the order in which players should be drafted.  It can, however,  be used during a draft to determine if a player justifies his draft pick.  The goal is to draft a player whose PAVT value is in-line with the other players drafted near the pick.  Draft a player with a lower PAVT value than those around him, and you will have overpaid.  A player picked with a higher PAVT value than those around him is considered a bargain. The most effective use of PAVT is to combine it with drafting tendencies and a real-time analysis of what is going on in the draft.  Average Draft Position (ADP) is helpful for determining where players are actually being drafted, but determining how much more or less a player is worth than their original PAVT value is dependent upon how many things can be considered in the time it takes to make a draft pick.  A great drafter will be able to consider factors such as everyone else's roster status, their estimation of player values, remaining strength at each position, bye week implications, back-up needs, etc.  Computer programs such as Draft Predictor are often employed to analyze this data and combine it with a list of player values to suggest and even predict draft selections.
 
Acknowledgements
PAVT was created by Robert Cobb in 2004.