Game Analysis: New Idea: HFNYR 2017-18 Season Player Ratings

Pavel Buchnevich

Drury and Laviolette Must Go
Dec 8, 2013
57,758
23,700
New York
Okay, so you are probably wondering what this thread is for. :laugh:

I had an idea.

What about if HFNYR kept our own ratings of our players this season? I've seen this type of thing done before, and it can be useful, and work. For someone who's not able to watch every game or if you miss a month of games or you want to use a consensus to make a point, having the forum's consensus on how each player has played for the season or even just one game and compared to the rest of the team could be useful. I also don't think its that hard to get it up and running, and working.

How does it work?

This is just my initial opinion, input on how we could make it work is welcomed. I would say that each regular season game each poster gets 10 points you can give players for playing well and 10 you can take away from players for playing poorly. And those ten points can be divided however you want. You can give one player +10, -10 to another, two players +5, one player -10, one -7, the other -3. However you want to give out your points. I'd say that its probably fair to make it so you don't have to allocate +10 and -10 for each game, we don't need to criticize players if we win 5-0. If it was a very uneven match though, you can give out +10 and -10. Please don't give a player you dislike -10 if he did nothing wrong or +10 for a player who did nothing that well. Don't use this contest for that type of thing, only participate if you believe you are giving a fair opinion of what happened. And for tracking purposes, I suggest we take the first five rankings per match. It'll be way too difficult to keep track of like 20 or 30, if that many want to participate, although if someone wants to keep track of that, maybe we can, but if I am keeping track of it, I think first five is good. No limit on how often you can participate, as much as you want, you just gotta be the first however many submissions we decide will be accepted after each game. Total +'s and -'s of the forum for each game will be calculated, and that'll be how it works. And then at the end of the season, I think we'll get the forum's consensus on how every player played. The best should have very high totals, the worst should be into the negatives.

Here's a random template for how it might look for a match.

+: Kreider +3, Zuccarello +4, Lundqvist +2, McDonagh +1
-: Staal -5, Hayes -3, Desharnais -2

Just post something simple like that, and we'll have a running total posted after the rankings for each match are done.

I wanted to start this thread a little early just so we can get this running well for game 1. Maybe we can do a test run for a preseason game. Any input is welcome, maybe its a terrible idea. And if it is, I'll probably have wasted my time suggesting this, but I thought it could be productive for the forum to have this type of analysis where we keep track of seasonal ratings/rankings of our players.
 
Last edited:

Irishguy42

Mr. Preachy
Sep 11, 2015
26,837
19,121
NJ
Would require someone to keep a record of the ratings in a spreadsheet or something. Would be interesting to see the ratings graphed out over the course of the season.

I guess a single rating for a player on a per-game basis would be based on a average of all the votes they got on here?
 

Pavel Buchnevich

Drury and Laviolette Must Go
Dec 8, 2013
57,758
23,700
New York
Would require someone to keep a record of the ratings in a spreadsheet or something. Would be interesting to see the ratings graphed out over the course of the season.

I guess a single rating for a player on a per-game basis would be based on a average of all the votes they got on here?

If we limit the submissions to like the first five to post, I can do it. But if we open it up to everyone who wants to participate every game, we'd need someone to enter all the info. Might take a bunch of work. I think first five will work, and since I suggested it, I don't want to outsource extra work to anyone else, unless they want to do so.

It would work in a way where if lets say four people decided to use their points on Hank for a game, the four submissions for Hank were +4, -2, +1 and +5, he would score 8 points for that game, as those four scores total up to +8. It could be averaged, but there's no need to, having the best player at the end of the season end up with like +200 or something like that and the worst with -200 would show the separation, as would +/- show the separation between players who did well and played who didn't. You could think of it in like a HFNYR +/- rating where we give out +/- scores for each game. If you played well, you get a high +/- score, if you played poorly, a low +/- score, but the fans would score it based on their opinion of how players played, it wouldn't be based on whether they were on the ice for goals like a normal +/- works.

Here's a random template of how the season scores could work, for example.

McDonagh: +175
Zuccarello: +160
Kreider: +140
Shattenkirk: +120
etc for players scoring in the positives.

Holden: -50
Staal: -150
etc for players scoring in the negatives.

And ideally if a team is good, they'll be more players in the positives than negatives.
 

mike14

Rampage Sherpa
Jun 22, 2006
17,921
10,937
Melbourne
We do something similar 'player of the year' on an Aussie Rules forum I post on, and it's an interesting way to see how different people view and judge a game.
That's a once a week sport which makes it a bit easy than ones with back-to-back games.

A few suggestions:
- IMO a 5, 4, 3, 2, 1 points voting system (best player gets 5, next 4 etc) is easier to manage
- don't use the negative figures
- the votes go in a seperate thread, started at the same time as the Post Game Thread
- the thread doesn't have any discussion, it's just people listing their 5 votes
- the voting thread is open for 48-72 hours post game. Enough for people that didn't watch it live to see a replay, or for those that want a second viewing to do so.

Once the thread is locked the votes are tallied. Even if the vote counter misses a week somewhere, the thread still exists so they can just go back to it and update. This also means that if they are going to be off the board for a long period, they can hand it over without any issues.

If we cap the the time the thread is open to something similar to what I've suggested above I should have enough time in schedule to tally the votes (noting there will at the minimum be a time zone delay between the thread closing and we counting the votes) and would be happy to do so.
 

TheTakedown

Puck is Life
Jul 11, 2012
13,689
1,480
if we use the table system the numbers can be easily pasted into Excel which would make for easy adding as well.
 

Pavel Buchnevich

Drury and Laviolette Must Go
Dec 8, 2013
57,758
23,700
New York
Mike, if you want to calculate it, by all means, we can use that system. I was suggesting the first 5-10 entries due to it being hard to calculate if we get like 25-30 per game. 82 games, games every few days, it could take some time to calculate that, but if someone wants to do that, they can go ahead. If I was calculating it, I'd limit it. I still think with like 5-10 per game we'd get close to the consensus.

The one reason why I think the rating all the players per game is hard is because its not so easy to remember how every player played. Can you imagine trying to remember how Fast played every game? I think it might be easier if we just ranked the notable players from every game, and by the end of the season, we'll get enough ratings on all the players.
 

egelband

Registered User
Sep 6, 2008
15,927
14,563
How about tallying the votes for three stars. And do the Stars for all games.
 

mike14

Rampage Sherpa
Jun 22, 2006
17,921
10,937
Melbourne
Mike, if you want to calculate it, by all means, we can use that system. I was suggesting the first 5-10 entries due to it being hard to calculate if we get like 25-30 per game. 82 games, games every few days, it could take some time to calculate that, but if someone wants to do that, they can go ahead. If I was calculating it, I'd limit it. I still think with like 5-10 per game we'd get close to the consensus.

The one reason why I think the rating all the players per game is hard is because its not so easy to remember how every player played. Can you imagine trying to remember how Fast played every game? I think it might be easier if we just ranked the notable players from every game, and by the end of the season, we'll get enough ratings on all the players.

You aren't ranking all the players, just your top 5.
As Egalband says, it's basically just the '3 stars' but with 2 extra spots (which aren't necessarily needed) and the #5 spot being better than the #1 spot (because the aim is to have a player with the most points be awarded 'best' at the end of the season).
 

Rangers in 7

Registered User
Dec 17, 2015
5,702
5,650
Long Island
i do like the ability to deduct points though

i worry that some will get carried away and some players on this roster no matter what they do will be seeing alot of red numbers
 

Pavel Buchnevich

Drury and Laviolette Must Go
Dec 8, 2013
57,758
23,700
New York
Sorry I haven't gotten back to anyone on this, been a little busy of late. But the preseason is coming up and maybe we could do a trial run to see how it would work. First we need to completely iron out the rules.

Of those people who would be willing to participate, can we just get some opinions on what you'd like so we can go with the consensus? The 5, 4, 3, 2, 1 system Mike suggested where you give five players points or would people here like a system where you can add and subtract points from players?
 

Pavel Buchnevich

Drury and Laviolette Must Go
Dec 8, 2013
57,758
23,700
New York
So is there still any interest in doing this? I just wanted to ask one last time before the season starts, as I didn't get an answer the first time, despite people expressing initial interest.
 

Ad

Upcoming events

Ad

Ad