PDA

View Full Version : RPI Pick to Click 2014-15: Under New Management



Pages : [1] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

lugnut92
09-26-2014, 09:53 PM
First, I'd like to thank Burgie12 for running this contest for the last few years and for allowing me to take over the job.

The rules and scoring will be nearly identical to previous seasons, with only one change (we'll get there). Without further ado, here we go!

As always, this contest is open to anyone who wants to play, not just the RPI faithful. It can take as few as 15 seconds to make your picks each week so why not play!

In general, you will pick one player from each team for each RPI game. You can't pick the same player twice in a weekend, and the rules are slightly different for tournaments (like the one we have the first real week of the season, but we'll climb that mountain when we get to it).

Picks must be submitted (or finalized) by puck drop of the first game of the weekend. If a player you've chosen is injured during the course of the weekend, you may select another player if you let me know before the start of the next game. If you're not sure if you're allowed to make a change in your picks, make a new pick anyway and I'll decide whether or not you can make the change.

To keep everyone from picking the same person, the top scoring forward and defenseman (by hockey points, NOT PtC points) from each team will be ineligible for selection. This rule will not come into effect until a player has at least five points.

Point Breakdown:
Forwards
Goal - 2 pts
Assist - 1 pt

Defensemen
Goal - 3 pts
Assist - 2 pts

Goaltenders
Quality Start - 2 pts (NEW, see explanation here (http://board.uscho.com/showthread.php?110644-RPI-Pick-to-Click-2014-15-Under-New-Management&p=6005145&viewfull=1#post6005145))
Shutout - 3 pts (Only to be awarded when the goalie plays the full game)
Goal - 10 pts
Assist - 4 pts

The designation for whether a player is at forward or defenseman will be determined on a game-to-game basis according to the box score at collegehockeystats.net.

The player who earns the most points in a week will earn themselves a bonus point.

Finally, if for some reason you can’t get to the board to make your picks, feel free to contact me in another way. I’ll use the timestamp to determine if you beat puck drop. (One of) my email(s) is my USCHO handle at gmail. If you have my cell, you can text me. If you’re my friend on Facebook, you can message me there. Don’t make a habit out of submitting picks off-board, but I understand that emergencies happen.

With all of that out of the way, let’s drop the puck!

Results from the most final week and wrap-up (ECAC Quarterfinals - at SLU (http://board.uscho.com/showthread.php?110644-RPI-Pick-to-Click-2014-15-Under-New-Management&p=6125753&viewfull=1#post6125753))
Direct Link to scoring spreadsheet (https://www.dropbox.com/s/4dncxyphbwa0oec/PtC_weekly.csv?dl=0)

lugnut92
09-26-2014, 09:53 PM
QUALITY START EXPLANATION
If you don’t care about what these are, feel free to ignore this post, it’s a bit of a long read.

Anyone who has paid any attention to the new statistics movement that has taken over hockey this summer (and baseball eons ago) knows that wins are a terrible stat for goalies (and pitchers). There are way too many factors that go into whether or not a goalie wins a game that aren't the goalie's play. With this in mind, I'd like to replace the win stat for goalies with the Quality Start (QS), first developed for the NHL by Rob Vollman.

What is a quality start? Well, it's when a goalie starts a game (or comes in in relief of an injured goalie within the first 10:00 of the game) and has a save percentage greater than league average (0.9077, defined by SV% of ECAC teams over last three seasons). Looking at last season's 132 league games (or 264 starting goalies), there were 138 QS (52.3%), and goalies who earned a quality start won 98 games (.710). Conversely, goalies failing to earn a QS only won 33 games (.262). One and only one goalie earned a QS in 64 games (48.5%), and only 31 games (23.5%) had no quality starts. In one game (1/17/14, Clarkson def. Yale, 3-2), the losing goalie earned a QS while the winning goalie did not, but this is explainable due to the low volume of shots (19) faced by the winning goalie.

A perfect example of a goalie getting screwed out of a win despite performing admirably is RPI's 1-0 loss to Harvard last season. No one could say that Diebold was the reason RPI fell (30 saves on 31 shots, .968), but he was saddled with a loss because his offense didn't perform. This doesn't seem right, does it? Diebold deserves points. On the other hand, when Clarkson beat SLU 5-4, Greg Lewis was awarded a win despite giving up 4 goals on 21 shots (.8095 SV%). In a shootout like this, neither goalie deserves to be awarded for their play. (Side note: I swear I didn’t intend for Diebold to be the “worthy” goalie and for Clarkson’s Lewis to be the “unworthy goalie”. It just worked out that these were the best examples of what I was saying.)

All of this is my way of saying that I think the QS is a better way to reward goalies who “click” than the win. There is actually a slightly higher chance of picking a goalie yielding points now (assuming this year’s goalie performance matches that of the last three seasons), so I suppose it promotes picking the goalie, but who am I to oppose gaming the system. If there is massive public outcry against the change, I can revert to the old way, but I think it’s only fitting that the engineering school use a “fancystat”.

lugnut92
09-26-2014, 10:14 PM
Link back to the first post with details on scoring (http://board.uscho.com/showthread.php?110644-RPI-Pick-to-Click-2014-15-Under-New-Management&p=6005143&viewfull=1#post6005143)

RPI kicks off the season with an exhibition against the PEI Panthers (0-3 in exhibitions against other CIS teams). It's the preseason for everyone from the players to the refs to us on the USCHO board, so points accumulated this week won't count toward your total for the season.

You may NOT choose the following players for this game:
Everyone is eligible!

Statistics and Rosters
RPI: 13-14 Statistics (http://collegehockeystats.net/1314/teamstats/renm) | 14-15 Roster (http://collegehockeystats.net/1415/rosters/renm)
PEI: 13-14 Statistics (http://english.cis-sic.ca/sports/mice/2013-14/teams/UPEI?view=lineup) | 14-15 Roster (http://www.gopanthersgo.ca/roster/mens-hockey/2014-15) (Yes I know it's empty, I'm looking I swear)

MavHockey14
09-27-2014, 04:56 PM
Thanks for taking over the reigns!

vizoroo
09-28-2014, 09:20 PM
Glad you've taken up the mantle of PtC. :)

Exhibition Game on the 4th--what time?
QS: .9077 is the standard for the full season including non-conference?

Ralph Baer
09-28-2014, 10:34 PM
glad you've taken up the mantle of ptc. :)

exhibition game on the 4th--what time?
Qs: .9077 is the standard for the full season including non-conference?

7 pm EDT

Ralph Baer
09-28-2014, 10:37 PM
Thanks for taking over the reigns!
"The king is dead, long live the king!" ;)

Sorry, I had to say it. It seemed so appropriate.

lugnut92
09-28-2014, 11:21 PM
QS: .9077 is the standard for the full season including non-conference?

Yes. Over the last three seasons, ECAC teams have faced 38,029 shots and made 34,520 saves. This includes all official NCAA games (i.e., conference and non-conference, but not exhibition).

FlagDUDE08
09-29-2014, 06:53 AM
Quality starts?! Great, my strategy's screwed...

RPI - Jacob Laliberté
UPEI - Mavric Parks

lugnut92
09-29-2014, 11:10 AM
Quality starts?! Great, my strategy's screwed...

Does it? Fewer than half of games have a goalie win, and more than half have a quality start.

FlagDUDE08
09-29-2014, 09:47 PM
Does it? Fewer than half of games have a goalie win, and more than half have a quality start.

Yes, but how many shots does RPI actually take in a game? Not many.

1988
09-30-2014, 04:01 PM
Well done Lugnut92!! Quality start data has been out there for years and is starting to gain momentum now that hockey is beginning to see the benefits of data to review performance. Save percent is part of the quality start equation and helps give a true picture of performance. I will take a .905% 55% quality start tender over a .912% 50% quality start tender ( in most instances).

lugnut92
09-30-2014, 07:01 PM
Yes, but how many shots does RPI actually take in a game? Not many.

Well, as Vollman defines it, the QS has a caveat for low shot volume (fewer than 20 shots faced) where a goalie only has to have a "replacement-level" save percentage (.885 for the NHL), but I really didn't feel like separating backups from starters and figuring out the proper line for shot count in NCAA hockey.

EDIT: Just checked, opposing goalies earned quality starts in 10 of RPI's 22 ECAC contests last season. RPI's goalies (well, goalie) earned quality starts in 11 of 22 games. This is compared to 17 games that had a winner.

lugnut92
10-01-2014, 10:45 AM
Time to make my own picks!

Saturday
RPI ENGINEERS: LUKE CURADI
PEI: Jordan Mayer

Ralph Baer
10-01-2014, 11:01 AM
lugnut,

A couple of questions.

1. You state "The player who earns the most points in a week will earn themselves a bonus point." Do you mean hockey player or the player in this contest. The latter being what was done in the past.

2. If the goalie is switched roughly in the middle of the game, as might be done this weekend, can the starting goalie still be awarded a quality start?

lugnut92
10-01-2014, 11:12 AM
lugnut,

A couple of questions.

1. You state "The player who earns the most points in a week will earn themselves a bonus point." Do you mean hockey player or the player in this contest. The latter being what was done in the past.

2. If the goalie is switched roughly in the middle of the game, as might be done this weekend, can the starting goalie still be awarded a quality start?

1. The player in this contest. Same as before. Sorry if that was unclear.

2. Yes, the starting goalie will be awarded the quality start regardless of time played if he earns it based on his save percentage. The only time the starting goalie isn't considered for a quality start is if he leaves in the first ten minutes before giving up a goal (this provision is to account for goalies leaving early in the game due to injury). If this happens, the goalie who comes in in relief of the injured player will then be considered the "starting goalie." In any case, to answer your question, if Kasdorf has a save percentage greater than .9077 before he gets pulled at the 20/30 minute mark, he will have earned a quality start.

engineerhockeyfan
10-01-2014, 12:09 PM
RPI-Schroeder

PEI-McIver

Ralph Baer
10-01-2014, 12:15 PM
I might as well make picks now and change them when the UPEI lineup shows up.

RPI: Bubela
UPEI: Desousa

hawthr
10-01-2014, 07:05 PM
RPI - Bubela
UPEI - Brown

vizoroo
10-01-2014, 07:51 PM
Engineers--Matt Neal
Panthers--Tyler Brown