Having FG%+ without having FG%# is like having a condom without a hard dick...its worthless...what the hell's the point?

This post could not be converted. To view the original post's thread, click here.

It's possible I'm not explaining the normalization process very well or that I misunderstand something myself since I'm not the resident statistical analyst.

Quote: Originally Posted By seble on 8/08/2008(1) Not questioning whether people can catch onto it, just that it would require a lot of explanation and not necessarily any more value than FG%+. (2) The # numbers indicate expected % against an opponent from a season where the league average was the all-time historical average. (3) So for that to actually be the adjusted number in the engine, all 5 defenders would have to be from a historically average season. I think most people would see that number and expect the player to shoot that over the course of a season, when really there's a lot more that plays into it.

(4) When I mentioned factoring in the defenders' league average, I mean that we essentially take the defensive ratings and block% and translate that into a FG% allowed for each defender. That number is then compared to the league average for that season and incorporated in the normalization. The baseball equivalent is comparing the pitcher's average allowed against the league average as well as the hitters average against his league average

(1) Saying that the + number has as much value as the # number is incorrect. When I do a mlb search for hitters, I don't even look at the + number, I only look at the # number (because that is what the sim uses).

(2) Yes

(3) If anyone is expecting that the player shoot exactly that percentage in every league, then they don't understand how a stat vs stat game works. That isn't a good reason to not show the # number. Also, what FG% are you using as a "historical league average"? IMO, if you aren't going to show the # number, then you shouldn't have normalized the FG% at all. You're making us do, imo, unnecessary work to compare players. I want to search by the FG%# number so I can compare players, I'm not expecting them to shoot that exact percentage.

(4) Ok

My point was that the plus number gives you enough to compare players, which is basically what you need to do. From that you can tell that player A is X amount better or worse than player B. It also tells you if that player was better or worse than average and whether they can expect to shoot better or worse than their actual number.

A # number would pretty much give you the same comparison, just with a different number. Since you point out that a # number isn't really what you can expect in the sim, then I'm not sure why it's more useful.

Quote: Originally Posted By seble on 8/08/2008(1) My point was that the plus number gives you enough to compare players, which is basically what you need to do. From that you can tell that player A is X amount better or worse than player B. It also tells you if that player was better or worse than average and whether they can expect to shoot better or worse than their actual number.

(2) A # number would pretty much give you the same comparison, just with a different number. Since you point out that a # number isn't really what you can expect in the sim, then I'm not sure why it's more useful.

(1) Unless you want to go through the log5 calculation (and know what number you use as the historical average FG%) you can't make an exact comparison unless the two players played in seasons where the league average FG% was the same. For example, let's say player A has a FG% of 48% and a FG+ of 105 while player B has a FG% of 46% and a FG+ of 110. A lot of people wouldn't be able to just look at those numbers and know who has the higher FG%#.

(2) It's more useful because it saves time. There is a reason that the mlb sim allows a search of the # stats.

I understand. We'll think about it some more.

Quote: Originally Posted By seble on 8/08/2008I understand. We'll think about it some more

Thanks. Can you tell me what you use as the historical league average FG%?

I'd have to check from the office on Monday.

Quote: Originally Posted By seble on 8/08/2008
I'd have to check from the office on Monday.

let me get this straight - the guy who does the programming doesnt even know the season/figure he based his modelling on?

You're right, I should have every line of code and every formula memorized.

so now you are confusing the underpinning average fg% with a line of code? is there any chance that WiS is going to hire a real programmer any time soon?

I doubt it, if I'm not mistaken, they still don't have full time programmers for CRD or HBD...and its been what, almost 2 years?

I think that falls into the category of formula.

'league average FG%' is not a formula it is a number