Each point on the graph represents one player. Price is plotted along the x-axis, and SWP's per game is plotted along the y-axis. For this graph, I like to use the average SWP's per game for the last 10 games. The red line represents the best fit line. (For those with a desire to know, I fitted a cubic equation, and minimized the squares of the price errors, where price was a function of SWP/game.)

In general, players represented by points above and to the left of the line are better value than those below and to the right. Of course, there are some exceptions - in particular when injuries are a factor.