HF GM Games Fantasy Mock Draft #3 - Discussion VI - PandaCan'tSnipeWellEnough (Voting is LIVE)

[LT]

Draft is complete - Vancouver selects Louis Domingue.

 

[LT]

Does McDavid lose out for having missed games to injury?

Yes. That's the flaw of raw PS. X TPS adjusts everything to 82 games.

 

[Exquisite]

Rank	Team	Points (Skaters)	Points (Goalies)	Points (Total)
1	Toronto Maple Leafs	93.48	17.57	111.05
2	Pittsburgh Penguins	93.72	16.82	110.54
3	Dallas Stars	91.42	16.31	107.73
4	New Jersey Devils	94.24	11.2	105.44
5	Colorado Avalanche	91.36	13.22	104.58
6	Columbus Blue Jackets	88.48	12.6	101.08
7	San Jose Sharks	85.23	14.79	100.02
8	Minnesota Wild	83.68	15.65	99.33
9	Vegas Golden Knights	82.17	17	99.17
10	Chicago Blackhawks	82.98	14.99	97.97
11	Anaheim Ducks	82.46	14.38	96.84
12	Washington Capitals	79.44	17.29	96.73
13	New York Rangers	80.75	15.25	96
14	Carolina Hurricanes	74.47	21.29	95.76
15	Philadelphia Flyers	81.09	14.38	95.47
16	Montreal Canadiens	79.9	15.22	95.12
17	Seattle Kraken	79.67	15.44	95.11
18	Arizona Coyotes	81.55	12.75	94.3
19	Calgary Flames	80.93	13.36	94.29
20	Detroit Red Wings	78.77	15.11	93.88
21	Ottawa Senators	79.69	13.42	93.11
22	Nashville Predators	81.44	11.55	92.99
23	Los Angeles Kings	75.47	17.22	92.69
24	Boston Bruins	77.02	15.63	92.65
25	Florida Panthers	77.58	14.91	92.49
26	St. Louis Blues	76.08	15.68	91.76
27	Winnipeg Jets	77.81	13.53	91.34
28	Edmonton Oilers	77.98	12.1	90.08
29	Tampa Bay Lightning	78.46	11.09	89.55
30	Vancouver Canucks	78.54	10.94	89.48
31	New York Islanders	71.71	12.39	84.1
32	Buffalo Sabres	62.14	13.93	76.07

[TBODY] [/TBODY]

Can't be right. Devils would be top 3 on both Panda and Latvian scores with a backup like Dell

imo Devils are bottom 5 lottery team... I was aiming for a top 5 pick for Lafreniere/Byfield/Stutzle because I assumed this was a keepers league

 

[Exquisite]

Hindsight team with 6th overall pick (if you drafted these players it means I like those picks)

Jeff Skinner (9) - Evgeni Malkin (1) - Bryan Rust (5)
Andrew Mangiapane (10) - Anthony Cirelli (3) - Joe Pavelski (6)
Jordan Greenway (12) - Tyler Bozak (13) - Drake Batherson (15)
Antoine Roussel (18) - Teddy Blueger (17) - Martin Frk (20)

Chychrun (4) - Theodore (2)
TJ Brodie (7) - Justin Schultz (14)
Haydn Fleury (16) - Matt Roy (11)

Mackenzie Blackwood (8)
Chris Dreidger (19)​

 

[Panda Bear]

Can't be right. Devils would be top 3 on both Panda and Latvian scores with a backup like Dell

imo Devils are bottom 5 lottery team... I was aiming for a top 5 pick for Lafreniere/Byfield/Stutzle because I assumed this was a keepers league

lel

 

[Canucks LB]

Not a fan of some charts honestly.
Not nearly enough importance on Defensive players.
Honestly is just skewers the rankings, I feel a lot of the "bottom 10 teams" have way better teams than some of the "top 10" Honestly

 

[AveryStar4Eva]

Not a fan of the charts honestly.
Not nearly enough importance on Defensive players.

I think the charts are cool and I appreciate how much time and effort people put into them, however they will make no impact on my voting

 

[Panda Bear]

Look at the real life standings and let me know what you think about them.

 

[Panda Bear]

1 BOS
2 TBL
3 PIT
4 COL
5 VEG
6 EDM
7 STL
8 PHI
9 CAR
10 WPG
11 TOR
12 VAN
13 NYR
14 FLA
15 ARI
16 DAL
17 NYI
18 NSH
19 MTL
20 CBJ
21 WSH
22 CHI
23 CGY
24 MIN
25 BUF
26 SJS
27 NJD
28 ANA
29 DET
30 OTT
31 LAK


^^
Real

 

[Canucks LB]

1 BOS
2 TBL
3 PIT
4 COL
5 VEG
6 EDM
7 STL
8 PHI
9 CAR
10 WPG
11 TOR
12 VAN
13 NYR
14 FLA
15 ARI
16 DAL
17 NYI
18 NSH
19 MTL
20 CBJ
21 WSH
22 CHI
23 CGY
24 MIN
25 BUF
26 SJS
27 NJD
28 ANA
29 DET
30 OTT
31 LAK


^^
Real

I like this much more!
Don't get me wrong panda, I think your fancy stats things is amazing, I just feel it's not a great indicator on how well teams draft.
You are still great

 

[LT]

1 BOS
2 TBL
3 PIT
4 COL
5 VEG
6 EDM
7 STL
8 PHI
9 CAR
10 WPG
11 TOR
12 VAN
13 NYR
14 FLA
15 ARI
16 DAL
17 NYI
18 NSH
19 MTL
20 CBJ
21 WSH
22 CHI
23 CGY
24 MIN
25 BUF
26 SJS
27 NJD
28 ANA
29 DET
30 OTT
31 LAK


^^
Real

Awful

Everyone knows LAK > OTT, DET

 

[Panda Bear]

I like this much more!
Don't get me wrong panda, I think your fancy stats things is amazing, I just feel it's not a great indicator on how well teams draft.
You are still great

and you're aware that this is actual NHL teams where Pettersson is a Canuck, Crosby a Penguin, and so on?

 

[Panda Bear]

Awful

Everyone knows LAK > OTT, DET

1) this is strength of squads after trade deadline
2) it does not account for injuries as all stats are prorated
3) algorithm is strength of squad's top 12 forwards and top 6 defencemen so depth is not a thing

NHL Injury Viz: 2019/20 team injury breakdowns (at suspension)
Detroit actually had really bad injury problems this year but no one talks about it

 

[Exquisite]

Look at the real life standings and let me know what you think about them.

top 5 ok, 6-31 is interesting. Clearly defense personnel is not as heavily weighted as some others

 

[Panda Bear]

top 5 ok, 6-31 is interesting. Clearly defense personnel is not as heavily weighted as some others

I didn't apply positional scarcity to things like DFF and xGoals for defencemen, but could do so.

I think it gets the tiers right.

 

[Canucks LB]

and you're aware that this is actual NHL teams where Pettersson is a Canuck, Crosby a Penguin, and so on?

Thats the joke

 

[Panda Bear]

Thats the joke

okay

but to clarify

you are okay with the real life rankings but hate the fantasy rankings even though they use the exact same ranking algorithm?

 

[Canucks LB]

okay

but to clarify

you are okay with the real life rankings but hate the fantasy rankings even though they use the exact same ranking algorithm?

 

[Joey Moss]

The issue is Pettersson isn't on Vancouver he's on Vegas now

 

[Panda Bear]

you're really in no place to use that emoticon

 

[Panda Bear]

The issue is Pettersson isn't on Vancouver he's on Vegas now

the suggestion is that if you find the real life rankings largely agreeable, then you should find the fantasy hockey rankings largely agreeable as they operate off the same algorithm

yes, I know, teammates chemistry usage all the classics etc etc etc

 

[LT]

1) this is strength of squads after trade deadline
2) it does not account for injuries as all stats are prorated
3) algorithm is strength of squad's top 12 forwards and top 6 defencemen so depth is not a thing

NHL Injury Viz: 2019/20 team injury breakdowns (at suspension)
Detroit actually had really bad injury problems this year but no one talks about it

was a joke

they're all bad

 

[Panda Bear]

was a joke

they're all bad

oh, cool. good

 

[Llamamoto]

Major update. Complete rosters.

I did a series of correlation studies to see how the different PandaRankings categories correlated to win percentage and goal differential. I did this using the real standings with real teams with real rosters, i.e. what you actually see when you look up the league standings.

PandaScores, which grade players, have a ~0.78 r-value for win percentage and ~0.8 r-value for goal differential.

PandaRankings, after a series of tweaks for just how to weight the different rankings, achieved a 0.854 r-value for win percentage and 0.887 r-value for goal differential. Given that these values were derived using my battery of statistics, this is phenomenal. That's right: PandaRankings have a 0.854 r-value for real-life win percentage.

My correlation started off at 0.74, and through a series of different techniques for weighting, I was able to get it this high. What's interesting, though, is how far some teams fell. As always, it's not perfect...but it's pretty f***ing good. Los Angeles is a major victim of the changes, but that's life.

So you like my team more than St. Louis? Awesome.

I have to say though, the bracket looks pretty good.

 

[Joey Moss]

the suggestion is that if you find the real life rankings largely agreeable, then you should find the fantasy hockey rankings largely agreeable as they operate off the same algorithm

yes, I know, teammates chemistry usage all the classics etc etc etc

Your charts are probably about as close as you can get to judging teams based solely off numbers. But it's still a secondary tool to me.