2022-2023 VHL Ratings Beta Release

[Utah Hockey Club]

Well the time is upon us to start bickering like children about the nuances that is SIM ratings. The 2022-2023 package is now ready for release and questions/comments/concerns. Take everything with a grain of salt as rounding will occur when loaded into SIM so those RFA signing amounts might change slightly.

Notes from this release:

1. McDavid's insane season has produced the first ever 89OV rating.
2. No runaway goalie this year with the top 5 bunched really close together. 
2. Only one rating formula change from last year. Goalies' PH is now using dump-in retrievals as a much better proxy for puck handling as opposed. Previous years used assists.
3. Biggest change this year in how the ratings are calculated is that there is a blended split between AHL and NHL players. Previous years would simply look at each individual, determine which league they played more games in and then created ratings with the league cohort that he played the most games with. This year, every player has a rating created in any league they played at least one game in. Then based on how many games they played, weight the two ratings to get a final ratings.

Example 1: Nils Hoglander

AHL GP: 45
AHL Rating: 59OV

NHL GP: 25
NHL Rating: 67OV

Final Rating: 64OV

In previous years, since Hoglander played more games in the AHL, his final SIM ratings would've been 59, but this year, it takes into account his time in the NHL to bump him a bit.

Example 2: Nils Aman

AHL GP: 17
AHL Rating: 56OV



NHL GP: 68
NHL Rating: 66OV



Final Rating: 63OV


In previous years, since Aman played more games in the NHL, his final SIM ratings would've been 59, but this year, it takes into account his time in the AHL to scale him back a bit since he couldn't hold down a full time NHL spot. 

Granted, AHLers are pretty useless in VHL, but this feature adds to the realism of the ratings package. Credit to Duncan for recommending this last season to give me time to work on implementing it for this year.

Please let me know if you have any questions. Most of the bugs have been squashed with the private alpha release. If you find anything out of sorts or wondering why a player has a certain rating, let me know and I can explain.

Without further adieu, the public beta release of the 2022-2023 VHL ratings:
docs.google.com/spreadsheets/d/1R8C7okGDL_8LbBsqhjxjGTly0wxqhC4HBmWCXPqhfyw/edit?usp=sharing

[Edmonton Oilers]

thanks for the hard work!!

[Florida Panthers]

[Boston Bruins]

Other than you're obviously missing... if team in (‘BOS’,’ARI’): OV += 5 .... But this will work. Great job as usual!

[Chicago Blackhawks]

+10 to every player on Chicago and probably still wont make the playoffs

[Utah Hockey Club]

I forgot to mention that for those asking how this rating package compares to last year, it is fairly even. Last year there was a total of 20 skaters with an OV of 80+. This year there is an exact same amount of players with 80+. I'm unsure of the exact distribution of the bottom half of the average line-up, but if anyone is interested I can get back to them once I dig into the numbers a bit more.

[Utah Hockey Club]

All big fixes have been squashed. Thanks to those who reached out for these.

New this year is that I bought a SIM license so I could do testing on the playability settings as this past year saw more abnormal results than expected. I loaded all NHL rosters at seasons end and ran ~45 seasons for the engine VHL uses (1.5). Once these were fine tuned I ran 5 more seasons including the playoffs and the results are included in THIS GSHEET. I'm very happy with the results of the settings and hope that this will result in the most realistic VHL season yet for the 2023-2024 season.

[Calgary Flames]

Jun 5, 2023 11:02:51 GMT -5 Utah Hockey Club said:

All big fixes have been squashed. Thanks to those who reached out for these.

New this year is that I bought a SIM license so I could do testing on the playability settings as this past year saw more abnormal results than expected. I loaded all NHL rosters at seasons end and ran ~45 seasons for the engine VHL uses (1.5). Once these were fine tuned I ran 5 more seasons including the playoffs and the results are included in THIS GSHEET. I'm very happy with the results of the settings and hope that this will result in the most realistic VHL season yet for the 2023-2024 season.

This is awesome. Thank you for all the work that went into this. 

[Buffalo Sabres]

Jun 5, 2023 11:02:51 GMT -5 Utah Hockey Club said:

All big fixes have been squashed. Thanks to those who reached out for these.

New this year is that I bought a SIM license so I could do testing on the playability settings as this past year saw more abnormal results than expected. I loaded all NHL rosters at seasons end and ran ~45 seasons for the engine VHL uses (1.5). Once these were fine tuned I ran 5 more seasons including the playoffs and the results are included in THIS GSHEET. I'm very happy with the results of the settings and hope that this will result in the most realistic VHL season yet for the 2023-2024 season.

did you run with our exact sim settings?

[Utah Hockey Club]

Jun 6, 2023 11:57:30 GMT -5 Buffalo Sabres said:

did you run with our exact sim settings?

Wolf sent me a screenshot of the current settings and I tuned from there. For those interested in my methodology I did the following:

1. Run full regular season SIMs and tune the goal slider until there was an approximate amount of players with over 82 points compared to the last NHL season. I was happy when there was between 20-30. In the NHL there was 28.

2. Run full regular season SIMs and tune the shots slider until there was an approximate amount of goalies with over 0.910SV% compared to the last NHL season. I was happy when there was over 10 while over 3 goalies with 0.920SV%. This would also required further tuning of the goal slider to offset any increase or decrease in shots to ensure scoring stayed level.

3. During this tuning, also adjust the fights and penalty slider to match PIMs and fight count in the NHL.

4. Run full regular season + playoffs to see how teams perform in the post-season. I'm happy I did this because originally NYI was winning half the Cups. I attribute this to the hits slider being too high and with guys like Cizikas, Martin and Clutterbuck having too much sway on gameplay. Once I turned hits down, it became much more realistic.

5. Once all sliders were set, I ran a final 5 full seasons (which the results are in the GSheet from above) to ensure everything was looking good. The only anomaly was the Red Wings acting like the Kings of 2012 and going to the Cup final after being an unlikely 8th seed. Other than that, everything seemed to replicate the previous NHL rather accurately in a repetitive season-over-season manner.