Wild West: 2018 NHL Playoff prediction strategies revisited

by chriskane on June 11, 2018
  • The Wild West
  • Wild West: 2018 NHL Playoff prediction strategies revisited

Another season in the books, and congratulations to the Washington Capitals. Let's take this opportunity to follow up on the picks I put forward at the beginning of the playoffs and check in on whether any of our prediction strategies were useful this year.

 

To start, we'll touch on the various prediction models. For a full reminder on what each of the models was check out the intro article. In short, we had one model that used a team’s goal differential, a scoring system that took into account goal and chance differentials, my personal expert opinion, and the consensus picks of the Dobber Panel.

 

Full disclosure: none of the models predicted either team in the Cup Final. So right off the bat we see that no model is perfect. However, if given two Western Conference teams facing off, some of the models proved quite effective at picking a winner.

 

Western

Round 1

Round 2

Round 3

Total

 

 

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

% Correct

Panel

3

1

1

1

0

1

4

3

57.14%

Goal Differential

3

1

1

1

0

1

4

3

57.14%

Matchup Chance Diff

4

0

2

0

0

1

6

1

85.71%

Expert

4

0

1

1

0

1

5

2

71.43%

 

Here we can see that Travis Yost’s model of using a combination of goal and chance differential seems to have been the best predictor- and by a fair margin. I had a decent picking rate here in the west (though as you may notice in a minute, my Eastern Conference picks were not fantastic.) and the Panel’s picks and the Goal Differential strategy were slightly better than 50/50, which is not ideal.

 

In an effort to get a more well-rounded picture, I ran the numbers for the entire playoffs and the results are below.

 

 

All Playoffs

Total

 

Correct

Incorrect

% Correct

Panel

8

7

53.33%

Goal Differential

9

6

60.00%

Matchup Chance Diff

12

2

85.71%

Expert

9

6

60.00%

 

With all of the available data we can see again that Yost had the most success this playoff season and it wasn’t particularly close. A rough outing with some Eastern predictions dropped me down to 60%, tied with the Goal Differential strategy, but at least those two were still reasonably well above 50/50. The panel again did not fair all that well, but since it is an undisputed fact that all of our advice always pans out, we will assume that is due to random variation.

 

*****

I also want to briefly return to the Sneaky Value adds that were depth or under-the-radar considerations for your playoff pools. Some of these I reviewed already after round one (when their teams were eliminated) so I included those reviews at the end in italics just for completion’s sake.

 

Vegas Shea Theodore

Theodore got to play 20 games this postseason. He ended the playoffs with three goals, seven assists (10 points) 39 shots and five power play points. He was the highest scoring defensemen on Vegas and sixth highest scoring defensemen overall. Incidentally he was also sixth on the Golden Knights  in points. He could have shot a bit more, but the point total was reasonable and if you got him this year, it was likely well after the Victor Hedman’s of the world (who finished with 11 points though granted in fewer games) 

 

San JoseEvander Kane

Kane started reasonably well, but went a bit quiet in later series. He suited up for nine of the Sharks games and put together a stat line of four goals, one assist (five points), two power play points, and 35 shots. Shooting almost four shots a game was certainly helpful, as were the four goals, but I was hoping for more from Kane.

 

Winnipeg Kyle Connor

Connor played 17 games for Winnipeg this postseason. He ended up being their sixth point producer with ten over the 17 games. He scored three goals, ten assists, one power play point and 36 shots. Clearly he managed to put some points on the board, but again I was hoping for a bit more than a 50 point pace, particularly on the power play.

 

Nashville Craig Smith

Smith had a of a dud of a postseason. He played in 13 games for Nashville but managed to only put up four points. They came in the form of two goals and two assists, one power play point and 32 shots. Almost three shots a game is decent, but the point production just was not there.

 

Minnesota Matt Dumba

Minnesota got five games in this series before losing to Winnipeg, so hopefully your playoff pool didn’t contain too many Wild players. Matt Dumba managed a goal, an assist and 10 shots on goal in the series. None of the points came on a power play, but his goal was a game winner.

 

For a little context, the Wild managed only eight goals over the course of those five games. They had no point-per-game players, with their top point producer Mikko Koivu, rounding out the series with four assists. Dumba then was tied for third overall in point production, and for shot production. While none of the totals were mind blowing, given the general team performance I think I have to give Dumba the win here as well.

 

LAK Jeff Carter/Dustin Brown

So LA lost in four games, and I had thrown out Jeff Carter and Dustin Brown as potential options. Carter managed 11 shots over that time span, while Brown accumulated one assist and 15 shots.

 

Hopefully you did not invest in LA Kings this playoff season. They played four games and managed just three goals, with Anze Kopitar being the only player to scrape more than one point (he had two). Given that we can say Dustin Brown getting an assist and managing the second highest shot total on the team might be a win, and Jeff Carter having the third highest shot total for the team is decent, but overall the King’s performance for your playoff pool was abysmal.

 

Anaheim Adam Henrique

Anaheim also managed to lose in four games, which certainly reduces the fantasy potential on the team. I wasn’t all that excited about Anaheim players in the preview, but I threw out Adam Henrique as an option. He played in all four games and totalled nine shots.

 

As with LA, Anaheim did not give much to work with. Over the course of the four games they scored four goals. Unlike with LA though seven players managed to get points, with the top four point producers scoring two. Henrique was not one of the seven though, and he also does not compare well in the shots department as he was tied with three other players for sixth most for the team. This was a short series, and hopefully you did not add Henrique or really any Anaheim players to your team.

 

Colorado Gabriel Landeskog

Gabriel Landeskog and the Avs got to play six games and Landeskog certainly made the most of his time. He put together four goals and three assists for seven points during his postseason. He also added 17 shots and two power play points.

 

His seven points ties him for the lead in the Western Conference with points with Austin Watson and Colton Sissons. I am not sure that anyone would have predicted those as the top three scorers in the west after round one (In another surprising side note seven of the top eight scorers from round one are from either the Predators or the Avalanche). The Avs did not advance so that certainly hurts his playoff pool value, but Landeskog was a great own for round one.   

 

Thanks for reading

 

 

Comments are closed.