BuzzerBeater Forums

So, your concern is noted. And while I appreciate that you have a good deal of knowledge with numbers, what you could help more with is a solution to the possible issue of sub-levels on the cap. For example: how to prove they exist (or not) and if they do, how to model appropriately.

The trick in figuring out potential potential sublevels is making sure you don't confuse them with skill sublevels. My best guess would be using the "current" salary level (or DMI in proficient game shape) as a regressor to control for unobserved sublevels in skills. If the per-position regressions produce significant differences between predicted and observed values of potential (or high error terms), this indicates differences within potential levels.

The problem with this approach is a certain collinearity on the right-hand side, since salary is also a product of skill levels. This may be addressed by running a regressin using just the salary levels (this will test against the assumption that the potential "value" and salary are linearly proportional).

This is mostly brainstorming, and I am sure there are at least a handful of users here who know how to work data, and can toss in their own ideas (Coco?).

All in all, it's not that I have concerns, per se. I do think that it is great that someone is willing to do this type of "reverse-engineering", since it's long overdue. I don't mean to come through as condescending, just trying to provide some pointers about data work, since I do have some experience there.

Actually, thanks to your message I was thinking about it today and realized that you are indeed right. You are right that one variable by position is not the right approach. I actually need 5 indicator variables (one for each position). Or I could just do one model per position, but that leaves no way to test for the significance of position.

As for regressing using current salary as a regressor, this again leaves potential as a y-variable. And in that case, if there are sub-levels on potential (ie: error), it still leads to a biased model.

Actually, thanks to your message I was thinking about it today and realized that you are indeed right. You are right that one variable by position is not the right approach. I actually need 5 indicator variables (one for each position). Or I could just do one model per position, but that leaves no way to test for the significance of position.

The standard practice for cathegorical variables is to include (n-1) dummies. In this case, we' need 4 variables (think of them as describing the "offset" from a base category).

This is only appropriate when you believe the position difference is a fixed effect, and the coefficients of other variables are identical. If you have reasons to believe that your categorical variable also affects the coefficients of other variables, you need to add interactions f all your LHS variables with all categorical variables.

You will still need a lot of observation, because the model now has a lot of variables (12 skill + 36 interactions + 4 dummies). At least 100 datapoints will be advisable.

As for regressing using current salary as a regressor, this again leaves potential as a y-variable. And in that case, if there are sub-levels on potential (ie: error), it still leads to a biased model.

You can try to flip the model, and regress salary on potential. If there is unexplained variance, this will indicate a sublevel in the potential -- though to me the results are pretty much a foregone conclusion, given that one can pretty much see that the same potential may lock at different salariy without running any parametric tests.

I don't think the game letting people know when a player reaches potential cap makes the game easy mode, in any aspect of the game, again, other than saving people some time for the weeks of training loss for when soft cap is unknowingly reached.

The problem is that potential is a soft cap and thus isn't reached per se, but training simply slows down as a player approaches his cap. Training never stops completely, either.

Plus, saving yourself the training loss is part of the skills necessary to play the game well.

Last edited by GM-kozlodoev at 9/6/2010 2:52:49 AM

Great suggestions. Can we plan to chat some more once I have more data? I can tell you some more of my observations and you can give more suggestions. Also, if you think of anything else do not hesitate.

I just want to also make the comment (if I have not already, I can't remember) that any data sent to me will be kept strictly confidential. Even if I get help from others on this study, I will only discuss data at the macro level and I will not release skills from individual players. And hopefully it also goes without saying that I will not discuss individual players in any public forum.

I thought about something.

The GM dameisel proposes to gather data thanks to his firefox add-on FoxyBB, and now also on the website (http://foxybb.tk/) in order to improve some formulas for this add-on.

Perhaps a partnership might be possible with people that would like to lead studies about skills like yours

Last edited by Dunker Joe at 9/7/2010 5:01:07 AM

I thought about something.

The GM dameisel proposes to gather data thanks to his firefox add-on FoxyBB, and now also on the website (http://foxybb.tk/) in order to improve some formulas for this add-on.

Perhaps a partnership might be possible with people that would like to lead studies about skills like yours

If he wants to share some information about players who are capped (and provided he knows a bit of their training history), I am more than happy to take it. But because I have guaranteed confidentiality, I will not be sharing the other way. Sorry.

An interesting suggestion. However, this data would not be useable in the model since it requires the precise moment of capping in order to be useful. At least that is how I see it.

I would be open to such a thing if someone could explain to me how to model effectively with uncapped players.

I could see some small use for it (ie: if I could get enough players, maybe by observation I could determine if sub-levels on potential exist). But just the fact that sub-levels exist does not get me any further along in the model (although, determining that they don't exist might).