Today's Fantasy Rx.

While many people still vociferously defend Standings Gain Points, Category Dollars offer two unique advantages in these 5 or 7 statistical categories.

The first, and more obvious of these advantages, is that CD reduces the degree of risk in your calculations instead of relying upon guesstimated denominators from previous seasons. The only inherent risk rests with the actual projections, and removing the unnecessary addition of denominators increases the internal validity of your calculations.

Second, and far more intriguing to most owners, is the ability to incorporate category inflation into your projections. We'll look at a common example from last season to provide a basic inflation calculation.

Phil Nevin's 41 HR out of the total 2480 HR produced by the top 168 position players earns Nevin a .0165 share of HR$. Multiplied by the $507 dollars assigned to HR, he earned $8.382 HR$.

Albert Pujols hit 37 homers in 2001 but only went for a buck or two on draft day in many leagues, especially those drafting prior to the finalization of Opening Day rosters. Pulling Pujols' 37 HR from the total 2480 raises Nevin's share from 41/2480 to 41/2443, or about .0168. Every other players' HR$ will increase by an amount directly proportional to their home run total.

The second step of the inflation calculation is to recalculate the dollars assigned to each category. With Pujols normally earning about $2, that $2 is subtracted from the total available $3120 ($260 * 12 teams). $3118 * .65(offense allocation) = $2026.7 / 4 (offensive categories) = $506.675 available HR$.

41/2443 (Nevin's share) * $506.675 HR$ = $8.503 HR$, and once rounded, Nevin now earns an extra HR$ in his value calculation.

Aside from the obvious benefits of category inflation like values adjusted for scarcity, you must avoid certain pitfalls, particularly with closers. If more than half the closers are kept, saves start appearing overvalued. The key is that even though the remaining closers and middle relievers with a few projected saves have zoomed to the top of the pitching charts, because of the growing scarcity of saves, they're worth these values.

Now, just as most of us wouldn't pay the going rate for injury-prone and older players, you can actually make a calculation adjustment to remove this increased saves' bias. Especially when we've updated our spreadsheet during drafts, we often remove saves from consideration after we've acquired one or two good relievers, instead distributing the money from those four categories over the remaining three. This move allows us to focus on relievers likely to contribute regardless of their role and has also netted us some fantastic starting pitching bargains like a $1 Kelvim Escobar and a couple of $1 Joe Mays from 2001 alone.

While you theoretically could run into similar problems with stolen bases late in the draft, we've never really seen this potential effect materialize in the field. We suspect that this omission is due to two primary reasons. Off-season player movement keeps steals somewhat fluid and the natural scarcity of steals forces owners to draft with an eye on steals early in the draft, thereby removing practically every projected double-digit SB player from the draft board by shortly after the half-way point of the draft.

Hopefully you now understand category inflation more thoroughly after these last two days of articles. Now, I'm going to provide another brief summary of how we arrive at our actual values once we've determined each player's Category Dollars in each category.

The first adjustment to convert Total Category Dollars into Draft Value is to modify all dollar values with respect to the worst player likely to be drafted. In a 12-team league with 14 position players per team, we use the value of the 168th player.

If the 168th players' TCD is over $1, we subtract the different between his TCD and $1. If he's somehow exactly at $1, we leave everyone's value alone, and if he's under $1 in value, we bump his value up to $1 while simultaneously providing a corresponding value increase to every other player. We call this new value Category Dollar Value for the sake of simplicity, but the name really has no intrinsic meaning

The second and final adjustment converts CDV into Draft Value by reallocating all the offensive dollars to account for keeper salaries. We accomplish this process in a couple of steps.

First, we take the sum of the top CDV values for the top 168 players to arrive at Total CDV. We then divide each player's CDV by the TCDV to determine their share of available dollars.

Second and last, we subtract the total amount of all keeper salaries by the allocated dollars for offense, and then multiply those available dollars by each player's CDV share to determine their Draft Value.

As alluded to in the above section on category inflation, any kept players are removed from the equation from the beginning. We keep them in their Draft Value position by zeroing all their statistics in the table and then instead of a calculated DV, we simply label them with their keeper salary, forcing the computer to keep them in the top 168 players when we sort by Draft Value.

The spreadsheet totals their salaries in a separate column to allow for the easy subtraction of total keeper salary from the allocated offensive dollars.

After we enter all our formula and data, we sort the spreadsheet using two criteria: Draft Value in descending order and Category Dollar Value in descending order. While we usually look at only the integer values for DV, we allow the CDV calculations to remain unrounded so players with identical DV can be placed in order for the draft. Usually between three to five sorts are necessary before the player order stops changing with every sort.

We know that many people prefer to use value ranges when drafting; if you want to use player groupings in dollar value ranges, simply manually transfer groups of players into those ranges. Keeping the players in their CDV order still allows you to know how players should be ranked according to your projections.

Due to the mostly random statistics inherent in our beginning alphabetized projection list, we occasionally wind up with the worst pitchers on top of the sorted list. If you run into that "problem", simply resort once in ascending order, and then return to descending sorts; this move has always fixed the glitch for us.

I hope I've provided a fairly thorough explanation of our value calculations, equations, and rationale between this column's first four and past three articles, but please feel free to e-mail with any additional questions. We're always looking to refine our valuation systems with new ideas, although we remain quite confident in the usefulness of this current model.

Tomorrow I expect to finally move on to a look back at all the player reviews as a whole to examine patterns and trends, as well as discussing potential changes for next year. I already have several modifications in mind due to what I learned in writing all these paragraphs and predictions over the last ten weeks.

Today's Fantasy Rx: What would you like to see changed in next year's after-season reviews? Please send in your comments and suggestions while the reviews are still fresh in your mind, and I'll discuss potential changes within the next couple of days.