Phantom inflation is one of our favorite draft day sightings and is a bigger
problem for people who bring computers into the draft.
The concept behind phantom inflation is relatively simple. When you set up a
computer program you tell the system how you want your league's salary cap
allocated. Assume that you use salary split of 65/35 between offense and
pitching, respectively. However, when all is said and done you find your league
only spent 32.5% on pitching (a fairly common number, especially in the
You may already start to guess the problem. In almost all valuation systems, the top 1/2 of the pitchers will generate about 75% to 80% of the value allocated to pitching. What happens when you are half way through the draft and you are using a computer to constantly update the values you should expect to pay for players? Obviously the system starts calculating a draft inflation or deflation based on the variance of what is expected to be paid and what is actually being paid.
To illustrate this very simply, we look to a standard
Obviously this has an impact right from the beginning as most pitchers are going for about 92% of what you thought. However, as the draft moves on the computer starts to give you some really bad results. The reason for this is that as the draft goes on the computer is still "off" the same set dollar amount but is factoring that into a smaller and smaller remaining value pool.
To continue the illustration, after 54 pitchers are taken (50%) about 75% of the value allocated to pitchers will have been bought. However, the computer program expected those 54 pitchers to go for $819 (75% of the $1,092 above). In actuality only $756 (75% of the $1,008 above) was spent. When you walked into the draft you expected the bottom 1/2 of the pitchers to go for $273 and they were valued accordingly. In actuality they will go for $252. However, the computer expects the league to spend a total of $1,092 on pitching or $336 on the remaining 54 pitchers. To compensate the system will start inflating the value of these pitchers by 23% to account for this "phantom inflation". Thus, the remaining pitchers (which you valued at $273) will actually sell for a combined $252. However, because the computer is calculating the phantom inflation it will have you value those pitchers at a 33% premium (336/252) over what you can actually purchase them at. Thus, a pitcher you originally valued at $6 you will bid to $7 or $8 and think you are getting a bargain.
Of course just the opposite is happing on the hitting side. Instead of seeing bargains everywhere the computer tells you that the prices will fall off the table and to wait. Of course they never do and you are stuck bottom feeding at the end with a large pile of unused salary cap, a major tactical mistake.
Of course this gets worse as the draft goes on. By the time you get to the final 20 pitchers, they will go for about $30. Thus, the league has spent $978 of the $1008 that it will actually spend. What does the computer tell you? It still expects the league to spend $1,092 in total so it is valuing those last 20 pitchers at $114 ($1,092 - $978). Thus a $2 pitcher will be valued at $7 or $8. When one of these pitchers goes to $4 you will happily bid $5 and get him not realizing that you just overpaid by $3 or more than double what you originally thought he was worth.
This "phantom inflation" has one further insidious effect. You went in trying to hold the line on pitching at $91. However, there were so many "bargains" in the pitchers and such good buys to come on the hitters that you ended up spending $100 on pitching. This is over 38% of the salary cap!! Of course you kept waiting for the bargains on the hitters that never materialized.
Remember, you told the computer only 65% would be spent on hitting, not the 67.5% that actually was. Thus, the computer kept telling you that all these hitters were left and that no money would be spent on them, it couldn't be. But it was. Thus, at the end of the draft you would have $10 or $20 that you were holding expecting to have to outbid others and the need never arose. Thus, it is very possible that you end up spending $100 on pitching, $140 on hitting and $20 to light your cigars.
Unfortunately this happens to many people every year and they come out
wondering what happened, wait a year and then purchase another software program
to be used in the draft again the next year because they think it will help
them. In the infamous words of P.T. Barnum, "There's a sucker born every
minute". Luckily for our customers, he was right.
Return to Strategic Essay List
Return to PPI Home Page
Email computer in draft horror stories
to PPI or email support@stallvalue.com
Copyright (c) 2003, Player Projections, Inc. All rights reserved.