Dynasty Hockey League

Season 21 Ratings Analysis Predictions

So i started writing up a season 21 predictions article and kept looking at teams and I’ll be honest, I lost interest after doing 5 teams. See, I’m a data analyst, so that’s what really makes me purr.

I tried to find an image of a cat doing some data analysis but could only find this image, so ill go with it.

First i grabbed all the current rosters and compiled them by team. Then I isolated each teams best players using top 8 forwards, top 5 D and starting goalie. I sorted by OV rating which on it’s own is not a great rating, in fact when Wynn created FHL sim he tried to avoid an OV rating but eventually added it in an effort to define superstar players (80+OV).

This is Minny, Minny is good, Minny is trending up

But back to buisness, i didn’t use OV for anything other than isolating each teams projected starters. So then i wanted to analyse each team on offense, defense and intangibles.

So offense i figured would be easy enough, SC obviously but PA seems to go hand in hand so i took each rosters F and D average PA and SC to generate an Offensive Prowess for each team.

By seasons end I want to hold this chart up against GF to see if there is any correlation between offensive prowess and GF, really this analysis has a ton of unknowns but if we can measure prediction based on ratings, it would be awesome.

Next i decided to use Puck Control (many consider PC an offensive rating but it can also be considered a defensive rating since possession of the puck is important to keeping other teams from scoring) and DF to isolate each teams average, this is called Defensive Prowess.

The Bruins are a team that historically has had very poor DP iirc, that didn’t hurt them when they won a cup and is it a major factor, let’s see how it plays this season.

The next ratings I wanted to analyze was Intensity and Strength a category i coined as Intimidation Factor.

The Penguins, ranked last stood out as a team that could find it hard to win with such a low IF. The Avs, Sharks and Oilers also sit below the water mark, it will be interesting to see what, if anything, that means this season.

Finally i looked at starters, nothing too intense here, i just looked at each teams starting goal using the important IT/SP/SK/PC:

It will be interesting to see how goaltending equals success, we all know its important but how important, can a team like the Rangers who rank in the middle in every category exceed those expectations on the strength of top 5 goaltending?

So that one’s pretty straight forward. The next question was how can i determine the best system to rank teams to try and predict the standings. Should i just do an overall average and rank teams 1-28? Here is what that looks like:

It would also make sense to rank based on each teams rank with the top rank going to the best in show:

You can notice how high a team like Ottawa sits in this analysis and notice how a better goalie would put them in the into #1 overall.

I also wanted to look within each conference, since that’s how the league is split to see if the results, when measured vs. teams you compete against for playoff berths, mean anything. First let’s look at the Eastern Conference:

According to this the Canes still rule the East. The Bruins will just beat out Winnipeg for 8th. Of course, in season trading will skew this but I really can’t argue with the Jets, Islanders, Leafs, Bolts, Habs and Sabres missing the playoffs in the East.

Here is the West results:

In the West the Oilers at the top isn’t a surprise, neither is the Avs. The Predators ranked higher than I would have originally thought and I would be lying if I said I thought the Stars would miss the playoffs. Something else i noticed is the Blues OP which you’d think would be better based on all that SC but a lack of PA brings down their average, we’ll see how that plays out this season.

As far as predictions this season, it’s pretty simple to project the beasts in each conference. What I think this really will tell us is how possible is it to project based on ratings using the categories I’ve used. Maybe this will just show the sim has it’s own non predictable factors which make this analysis pointless. If that’s true, I won’t bother doing this in future seasons but I predict there is some conclusions we will be able to build as we compare this to the results.

One Comment

  1. Chris-Wild

    June 16, 2016 at 2:21 PM

    This was an absolute fail…none of these stats match ratings analysis…back to the drawing board.

You must be logged in to post a comment Login

Leave a Reply