The Pittsburgh Penguins finished up the 3rd quarter of the 2013-14 NHL season just prior to the Trade Deadline, and earlier this week we took a look at the 3rd Quarter results. But those are just raw numbers, based on the standard stats available on the score sheets. If we want to look back and compare player growth throughout the season we need to look at fancystats. We did this earlier this year, comparing the 1st Quarter to the previous 3 seasons average numbers, and then later used the 2nd Quarter to compare to the performance from the start of the year. So once again we are back to look at player growth by comparing their 3rd quarter numbers to those in the first half of the season.
Like last time, we have a couple of players who were new to the lineup, so in order to assign grades to them I had to compare to their past 3 seasons of data, so their numbers will not quite mesh with those of others who have been judged based solely on this year's performance. This includes waiver pickup Pyatt and 1 game call ups Drazenovic and Letourneau-Leblond.
We also have to keep some sample size issues in mind too, so players that only played a handful of games are going to have their numbers a bit skewed. This most definitely means we need to take the results for Conner, Ebbett, and Zolnierczyk with a grain of salt. And possibly be somewhat skeptical of Despres, Sill, Kobasew, Vitale, and Megna, although those ones do have a reasonably decent sample size to work with. Perhaps Gibbons, as we are comparing him to a fairly small sample size form the beginning of the year, or even Scuderi who also missed a large chunk out of the first half of the year.
As mentioned the last time we did this, these grades do not indicate who is a good player vs who is a bad player, it is simply a measure of individual growth. A player with a B has improved on his numbers from earlier this season, whereas a player with a D has started to slump and is not playing as well as they did to start the year. They do not compare one player to another in any way, shape, or form.
The grades were calculated by taking the Percent Difference, that is to say the numbers from the 3rd Quarter minus those of the 1st half and divide that by the numbers from the 1st half. That gives us a percentage change, whether the player has improved or declined in the 3rd quarter. So in order to convert to a grade, we add 75% to that number since that is the average midpoint in the standard grading scale. Normally for us in the US a 75% is the midpoint C, but after the 1st Quarter numbers had been depressing I had decided to change to the Canadian grading scale in which a 75% was the midpoint for a B.
For skaters, we had 4 individual categories based on 5-on-5 data: Goals per 60, Points per 60, iFenwick per 60, and Fenwick Shooting Percent. Those were weighted as if they were 4-credit classes because they are strictly based on the players' individual performance. Then we had 6 on-ice categories based on 5-on-5 Close data: Goal For per 60, Goals Against per 60, Goals For Percent, Fenwick For Percent, Fenwick Shot Percent, and Corsi Save Percent. Those were weighted as if they were 3-credit classes because they are based on a group effort.
For the goaltenders we had fewer categories to work with so I included different situations. They have overall Points Percent, which is the number of points the team earned out of the total possible in the games they played, as well as Shutout Percent, which is simply what percentage of their games played ended in a shutout. Then we have Goals Against Average and Save Percent in overall, 5-on-5, and Close situations. The Save Percent was the only number that can truly be considered an individual effort by the goalie and as such was weighted as a 4-credit class whereas the rest of the categories were total team efforts and as such were weighted as 3-credit classes.
That then allows us to get an overall grade based on 12-point grading scale. I also tossed in PDO and Zone Start changes simply for reference. So for example if a player is vastly improved this Quarter and you see that he also had an increased proportion of O-Zone Starts or an unsustainable high PDO it could explain a sudden surge. Likewise, a player that is struggling could be explained partially by a decreased proportion of O-Zone Starts or a lowered PDO. I listed the actual Percent Change for these rather than converting it to a grade. As such any change within 2% is practically meaningless, and a change of up to 5% could still be considered just normal ebb and flow throughout the season. However, any time you have a greater than 5% change it certainly becomes noteworthy.