|
Post by donut on Jul 29, 2019 12:39:49 GMT -5
I hope everyone is ready to spend plenty of time on Excel, lol. Not really. It's really just 1 extra step - dividing by the matches played. As an "excel whiz," I'm sure you know that you can set columns to automatically round to a certain decimal place. If any of the new players need help setting up Excel sheets, let one of the admins or vets know! But nice try at a scare tactic!
|
|
|
Post by Fight On! on Jul 29, 2019 12:41:56 GMT -5
I hope everyone is ready to spend plenty of time on Excel, lol. Not really. It's really just 1 extra step - dividing by the matches played. As an "excel whiz," I'm sure you know that you can set columns to automatically round to a certain decimal place. If any of the new players need help setting up Excel sheets, let one of the admins or vets know! But nice try at a scare tactic! Mary, people couldn’t even add properly in previous years. One extra step is a lot for some folks (Mostly J/K).
|
|
|
Post by huskerjen on Jul 29, 2019 12:42:02 GMT -5
I hope everyone is ready to spend plenty of time on Excel, lol. Data scrape -> CSV -> VBA Should take seconds.
|
|
|
Post by donut on Jul 29, 2019 12:51:24 GMT -5
Not really. It's really just 1 extra step - dividing by the matches played. As an "excel whiz," I'm sure you know that you can set columns to automatically round to a certain decimal place. If any of the new players need help setting up Excel sheets, let one of the admins or vets know! But nice try at a scare tactic! Mary, people couldn’t even add properly in previous years. One extra step is a lot for some folks (Mostly J/K). Especially when they're hungover!
|
|
|
Post by Hawk Attack on Jul 29, 2019 12:51:44 GMT -5
Excel? Dead art form.
Google sheets is where it’s at.
|
|
|
Post by Fight On! on Jul 29, 2019 12:52:53 GMT -5
Mary, people couldn’t even add properly in previous years. One extra step is a lot for some folks (Mostly J/K). Especially when they're hungover! I wasn’t hungover by the time I did my stats! Lol
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 30, 2019 0:01:28 GMT -5
I hope everyone is ready to spend plenty of time on Excel, lol. Not really. It's really just 1 extra step - dividing by the matches played. As an "excel whiz," I'm sure you know that you can set columns to automatically round to a certain decimal place. If any of the new players need help setting up Excel sheets, let one of the admins or vets know! But nice try at a scare tactic! like, kids learn =average(...) in elementary school nowadays
|
|
trojansc
Legend
All-VolleyTalk 1st Team (2022, 2021, 2020, 2019, 2018, 2017), All-VolleyTalk 2nd Team (2016), 2021, 2019 Fantasy League Champion, 2020 Fantasy League Runner Up, 2022 2nd Runner Up
Posts: 28,099
|
Post by trojansc on Jul 30, 2019 0:39:51 GMT -5
Excel? Dead art form. Google sheets is where it’s at. I use OpenOffice Calc #Free99 Google sheets is free tho too, right?
|
|
|
Post by ay2013 on Jul 30, 2019 1:50:23 GMT -5
Excel? Dead art form. Google sheets is where it’s at. I use OpenOffice Calc #Free99 Google sheets is free tho too, right? I haven't gotten into google sheets yet. The default for everything I do is a word processor, and I still find Word to be the premier application, and with Word comes all the other Microsoft stuff, so why change?
|
|
|
Post by huskerjen on Jul 30, 2019 21:23:28 GMT -5
Question: When creating a predictive model for player production, do you typically use a Gaussian Markov model like brownian motion, or do you assume more memory in the data set for upperclassmen and use a power law distribution like 1/f? Or is it better to just use prior years, segregate training sets for unsupervised learning e.g. hierarchical clustering, find the parametric bounds and derive a simplified algorithm from there?
Sorry, I'm new to this, I hope I can keep up with the vets...
|
|
|
Post by Wolfgang on Jul 30, 2019 22:04:35 GMT -5
Who are the administrators of this league? (For my records.)
I know you named them, but I forgot.
|
|
|
Post by Fight On! on Jul 30, 2019 22:07:38 GMT -5
Question: When creating a predictive model for player production, do you typically use a Gaussian Markov model like brownian motion, or do you assume more memory in the data set for upperclassmen and use a power law distribution like 1/f? Or is it better to just use prior years, segregate training sets for unsupervised learning e.g. hierarchical clustering, find the parametric bounds and derive a simplified algorithm from there? Sorry, I'm new to this, I hope I can keep up with the vets... OMG, you haven’t started yet?
|
|
|
Post by huskerjen on Jul 30, 2019 22:38:00 GMT -5
Question: When creating a predictive model for player production, do you typically use a Gaussian Markov model like brownian motion, or do you assume more memory in the data set for upperclassmen and use a power law distribution like 1/f? Or is it better to just use prior years, segregate training sets for unsupervised learning e.g. hierarchical clustering, find the parametric bounds and derive a simplified algorithm from there? Sorry, I'm new to this, I hope I can keep up with the vets... OMG, you haven’t started yet? It's fine, I have a team of quants at work that are doing it for me.
|
|
|
Post by Fight On! on Jul 30, 2019 22:41:20 GMT -5
OMG, you haven’t started yet? It's fine, I have a team of quants at work that are doing it for me. I hope your injury prediction model has a high R-squared.
|
|
|
Post by huskerjen on Jul 30, 2019 23:37:44 GMT -5
It's fine, I have a team of quants at work that are doing it for me. I hope your injury prediction model has a high R-squared. RMSE is more useful than R-squared, but in this case I care about neither. I want to know what's going to happen this year, not what has happened in the past. Probability > statistics.
|
|