Every year at this time, I am always asked the same question - “Do you guys have anything that will calculate inflation at my auction?”
And every year, I regretfully respond – “Sorry, no, not this year.”
Here’s the deal. I know there is a conventional treatment to account for draft inflation that is….um....
Actually, I should probably first start with a general explanation of inflation, for those new to the hobby or more familiar with drafts than auction, as inflation is a concept intrinsic to the auction format, specifically keeper auctions.
A rotisserie auction is a zero-sum game. There is a fixed amount of money available to purchase a fixed amount of talent. The price of the talent is tailored to perfectly match the available money by assigning a value to each player based on their projected performance. The number of players assigned value is exactly equal to the number of players needed to legally populate everyone’ roster.
In keeper leagues, owners are allowed to own the same players the following year. The players they most often choose to keep have salaries less than what they are projected to earn. A typical keeper list may consist of eight or ten players, projected to earn $100 but only costing that owner $50. Keeping in mind that every league owner is going to be keeping players less than projected value, what results is an economy where there is more money available to be spent than there is available talent, according to their projected value.
This scenario is what leads to owners buying players for inflated prices, hence the term draft inflation. It is a cardinal sin to leave money on the table at an auction, so in order to leave spending your entire available budget, you need to pay more than projected value in keeper league auctions. So what the original question I posed was really asking is “do you guys have a means of resetting the projected value to account for league inflation?”
Which brings me back to…"Here's the deal."
I know there is a conventional treatment to deal with draft inflation that is algebraically correct and even makes some logical sense, but quite frankly, it is impractical and not representative of what really occurs in keeper auctions. As such, I feel it does more harm than good and perhaps selfishly, have opted not to incorporate it into any of our tools available in our Platinum subscription package.
The concept is as follows. A league inflation factor can be calculated based on the total money available to be spent and the total of the projected value of the remaining draft-worthy players (the best players left to fill each roster legally). The inflation is the former divided by the latter. By means of example, let us say that we are playing with a 12-team league with a $260 budget and each team has a keeper list described above (protecting $100 worth of talent for $50). The total money in the pool is 12 x $260 = $3120. The money available is $3120 - $600 = $2520, where the $600 is derived from 12 teams each freezing $50 in salary. Another way of looking at it is each team has $210 left to spend so there is 12 x $210 = $2520 leftover. The projected value of the remaining players is $3120 - $1200 = $1920, where $1200 is generated from 12 teams each freezing $100 worth of talent. The inflation factor is the former divided by the latter or $2520/$1920 = 1.3125. This example league is said to have 31.25% inflation. That is, if every available player is purchased for 31.25% more than their projected value, every owner would spend exactly all of their money. Therefore, the conventional means of accounting for inflation is to multiply each projected value by the inflation factor, a treatment I call linear since every value gets adjusted the same percentage.
As suggested, the chief beef I have with this treatment is this is never the way real keeper auctions precede. For reasons that are beyond the scope of this essay (but will be discussed in the future) the tendency is for the best available players to be purchased at prices beyond this linearly adjusted value. My concern is if we provided this value, you would never spend all your money in a keeper auction because this adjusted linear value would always suggest that you stop bidding at a price where you will never acquire quality talent. Granted, to be completely truthful, I have the same concerns about our standard values being a stop/go mechanism and not a guideline, but there is no way we could get away with not publishing dollar values, so we do it and offer companion strategy essays covering this subject.
In the back of my mind, I have always had an idea I wanted to try to better account for draft inflation and it turns out it “works,” or at minimum, renders inflated bid guidelines I am more comfortable providing. Amid the problems with the conventional treatment is the fact that there is no real theoretical or even scientific reason for doing it beyond “the math works out.” My idea incorporates some basic valuation principles. Plus, the math works out.
One of the principles of a sound valuation system is the last player drafted at each position should be assigned a value of $1 (or whatever the minimum bid is necessary to purchase the player). The linear application of inflation makes a $1 player to be $1.3125 in our example. While I understand when you round off, the value returns to $1, but that is not the point. The non-rounded off, lowest value should be the minimum required bid at all times. To carry this further, the order of the players put up for bid in an auction does not have to go from top to bottom. The lowest valued player can be put up for auction at any time. Once that player is off the table, the next lowest ranked player at that position should now assume the minimum bid. That is, if the second to bottom player is valued at $3, once the bottom player is purchased, that $3 drops to $1. While I realize you can set up the conventional method to recalculate inflation each time, this adjustment is not talked about enough.
So we have established that at all times, the lowest ranked player at each position should always be assigned a value of $1. What happens next is the available money is then distributed to the available players in proportion to their percentage of statistical contribution to the total statistical contribution. This is the key, as it makes the adjustment in proportion to the player’s contribution, not linearly.
The way my treatment of inflation works is just as described. The lowest ranked player is set at $1 and everyone else is scaled up proportionately. This calculation is repeated after every player purchase. The result is far more money is allocated to the top players as is done via the conventional treatment.
The reason I am now comfortable providing this functionality in a couple of our Platinum tools is now, these values are on a par with the normal dollar values we publish. By on par, I mean in both instances, the lowest ranked player is worth $1 and everyone else is scaled up. I have the same concerns I had before, but at least now I feel I can better address them in an essay. There is no algorithm that can accurately predict how others will think. There is no set of equations that can determine that someone is looking to rebuild and is willing to pay whatever it takes to purchase Albert Pujols so they have a trading chip. There is no macro that can be written to get inside someone’s head that believes they have a great, cheap offense and are thus willing to pay through the nose for Roy Halladay to anchor their pitching staff.
With that as a backdrop, at the bottom of this column, I have linked a sample version of an Excel tool that will soon be available for our Platinum subscribers. The MastersDraft, our own drafting software already has this functionality built in. The Excel tool will be offered as a standalone product which you can use to track inflation (or deflation in redraft auctions) to HELP/GUIDE/ASSIST you with your bid price. The sample version is loaded with hitters and their year ending value from 15 team mixed leagues last season. The version made available for Platinum subscribers will allow you to use any of our 2012 values for common leagues or to import values customized to your own leagues.
Here is what you need to do. The only column you need worry about is D, labeled REAL. Here is where you type in the salary of the protected players as well as the amount they are purchased for in your auction. The column labeled PROJECTED is the unadjusted projected value, it will never change. The column labeled OLD computes the inflated value using the conventional method. The one called NEW is the value adjusted in the manner that allows me to sleep at night.
To see how things work, type in some typical keeper salaries in the REAL column. Once you reach a point where values are affected, you will see the numbers change in OLD and NEW. You will also see how the two columns differ with the NEW column containing values higher than the OLD column. As a matter of fact, you will see that before any player is even entered to the REAL column, which is what usually happens even in redraft leagues – the top players go for more than the projected value.
You will note as each player is purchased, their price disappears and the available players are re-ranked. If you opt to use this to track inflation in a real auction, you can sort on the rank column by placing the cursor in cell AI (where it says RANK) and sort ascending. All the available players are brought to the top to make it easier to see who is left.
One final note: inflation should be calculated separately for hitting and pitching. All I have here is hitting and it assumes the standard 69% budget allotted to hitting. MastersDraft and the standalone tool allow for customization in this regard. You can say to lump hitters and pitchers into one pool and adjust globally, or set them separately and assign a hitting to pitching split.
CLICK HERE to download sample inflation tool