This is not a bug, and it's actually not terribly rare. What you're seeing is one of the shortcuts that we took in order to make this transfer compare scheme function on a limited amount of data. Rather than grouping players by position and whichever of their 10 skills is highest, we group them by position and whichever of 3 skills, particular to their position, is highest.
Very few centers have outside defense as their top skill, but a few do. When also restricting by age, potential, and exactly what level of OD a center has, it is very likely that there will have been no similar recent transfers at all. By restricting player grouping to three skills per position, we are able to place about 98% of players into groups where their top skill (or one of their top skills, in case there is a tie) is among the three skills.
Remember, that all skills are considered when the estimator compares a player to the recent transfers in his group, and that even in these rare cases when a player does not share his top skill with the transfers he's being compared to, that his top skill will work to elevate him within the group and increasing his price estimate. I suspect that players like this are more likely to end-up with estimates of the "Agustín Ereño is probably better than all 21 of these players..." type. Even when that's the case, we hope that it is more valuable than the "Very few players like X have been transferred recently" message we would give if we forced players into groups based on unusual top skills.
I know the next thing you're going to ask, so I will go ahead and answer it now: the eligible skills are P, H, and OD for Point Guards; JS, JR, and OD for Shooting Guards; JS, ID, and R for Small Forwards; and ID, IS, and R for both Power Forwards and Centers. I warn everyone not to overanalyze these skill sets. They were not choosed because they are the most important, or the most common skills for each position, but merely because each set of 3 together offered maximal coverage of the top skills that occur in each position. For example, outside defense is an important skill for small forwards, but inside defense may have been chosen as one of the 3 SF options instead because more SFs have one of {JS, ID, R} as one of their top skills than have one of {JS, OD, R}. I am just speculating, but this may be because there is a stonger positive correlation beween JS and OD than there is with JS and ID. In any case, it is not correct to assume based on the skill sets that ID is more important than OD, or that there are more SFs with top skill ID than there are with top skill OD, or that SFs with high ID sell for more than SFs with high OD. The sets were chosen together, as a group of 3, to leave as few players out as possible, and nothing more.
Last edited by Mike at 6/30/2008 6:27:56 AM