Database Reference
In-Depth Information
But Juan knows that these players are likely already on the radar screen for many other teams in
the league as well. Perhaps he should look at a few potential alternatives, that aren't quite as
obvious to everyone. Juan might be able to score a real win by thinking creatively, and his savvy
and experience has told him that sometimes the best player acquisitions aren't always the most
obvious ones. Click on confidence(Franchise_Player) twice.
Figure 11-9. The scoring data set's predicted values, with highest Franchise_Player
confidences sorted to the top.
There are 11 predicted Franchise Players in the list of 59 prospects. Perhaps Juan could suggest to
management that a solid, long-term building block player could be Fred Clarke. Clarke may be
easier to persuade to come to the team because fewer teams may already be in contact with him,
and he may be less expensive in terms of salary than most of the superstars will be. This makes
sense, but there may be an even better player to pursue. Consider Lance Goodwin on Row 20.
Goodwin is predicted to be a Franchise Player, so Juan knows he can play—consistently and at a
high level. He would be a solid and long term acquisition for any team. But add to this Goodwin's
confidence percentage in the Superstar column. Our neural network is predicting that there is
almost an 8% chance that Goodwin will rise to the level of Superstar. With 10 years of experience,
Goodwin may be poised to reach the pinnacle of his career within the next season or two.
Although he was not the first or most obvious choice in the data set, Goodwin certainly appears to
