04-22-2020, 05:15 AM
(This post was last modified: 04-22-2020, 05:17 AM by 37thchamber.)
Elo in and of itself is an interesting rating solution, so thumbs up from me.
Though it arguably gets less effective for more complex scoring games like football. For chess it's perfect because there are only 3 possible outcomes, and while you can technically quantify the margin of victory, it is less meaningful than margin of victory in football (for example). So there's a case to be made for adjusting the calculation with margin of victory (as is done in the soccer equivalent)
Not sure whether I would adjust the k factor for playoffs or not (again, this is done in the soccer equivalent), and it's probably best to experiment with that. Same goes for adjusting k factor based on rating (as is done by the USCF, so lower rated players don't get the same rating boost by beating a peer as a grandmaster might).
You've already addressed the home advantage issue (any tweaks to k factor may require re-evaluation here, but at least there is a good base in place), so thats pretty kickass.
Would probably be interesting to track this all the way back to season one, with expansion teams being added with a base rating of 1500 as and when they were created.
Anyway this is all a longwinded way of saying that Elo here would be particularly useful for predicting outcomes (and by extension, running a casino of some sort). I might throw something together with all results in it at some point, and see what I can cook up. Been meaning to do something like this for a while. If you wanna look at that as a larger project, hit me up on the dev/wiki discord server and we can plot something out.
I feel like it would be a nice bit of added value for the league as a whole, personally.
Though it arguably gets less effective for more complex scoring games like football. For chess it's perfect because there are only 3 possible outcomes, and while you can technically quantify the margin of victory, it is less meaningful than margin of victory in football (for example). So there's a case to be made for adjusting the calculation with margin of victory (as is done in the soccer equivalent)
Not sure whether I would adjust the k factor for playoffs or not (again, this is done in the soccer equivalent), and it's probably best to experiment with that. Same goes for adjusting k factor based on rating (as is done by the USCF, so lower rated players don't get the same rating boost by beating a peer as a grandmaster might).
You've already addressed the home advantage issue (any tweaks to k factor may require re-evaluation here, but at least there is a good base in place), so thats pretty kickass.
Would probably be interesting to track this all the way back to season one, with expansion teams being added with a base rating of 1500 as and when they were created.
Anyway this is all a longwinded way of saying that Elo here would be particularly useful for predicting outcomes (and by extension, running a casino of some sort). I might throw something together with all results in it at some point, and see what I can cook up. Been meaning to do something like this for a while. If you wanna look at that as a larger project, hit me up on the dev/wiki discord server and we can plot something out.
I feel like it would be a nice bit of added value for the league as a whole, personally.
I impersonate a programmer for a living
Father of the League Wiki • Friendly Neighbourhood Angry Black Guy™ • NOT British
Originator of the Sim League Cinematic Universe (SLCU)
Super capitalists are parasites. Fite me.
Alternatively, if you agree, you can support a grassroots movement dedicated to educating and organising the working class by buying a digital newspaper subscription. Your support would be greatly appreciated.
