BuzzerBeater Forums

Suggestions > Player rankings: confusing when not enough data

Player rankings: confusing when not enough data

Set priority
Show messages by
This Post:
00
133014.1
Date: 3/1/2010 5:37:11 AM
Overall Posts Rated:
587587
The player rankings would be much more fun (i.e. worth checking out) if you could rely on them. The problem right now is that a player gets a ranking even when there is not enough data for him to be considered in a particular stat.

Let's say you have two players with the following 3-point stats:

Player A: 31/121, 0.256
Player B: 8/33, 0.242

Now, player A may get a league ranking of, say, 39th in 3-point %, while player B goes 51st. For player A, this is his real ranking, while for player B, this just means that he doesn't have enough attempts (or shots made) to be considered in the ranking. You have no way of knowing this without comparing a bunch of guys by yourself. But wait, isn't that the point of the player ranking system that you don't have to do the comparisons yourself?

I suggest that all the lines where the player doesn't have enough data (attempts or shots made in the example) are left blank. No information is usually better than misleading information.

This Post:
00
133014.2 in reply to 133014.1
Date: 3/1/2010 5:47:49 AM
Overall Posts Rated:
959959
i would miss it, and it gives also stats about 3pt attemps and made.

This Post:
00
133014.3 in reply to 133014.2
Date: 3/1/2010 6:58:16 AM
Overall Posts Rated:
587587
I would think there is an easier way of getting that information than trying to find out if 37th, 45th, 51st, or possibly 78th ranking means that the player is not above the threshold (which we do not know anyway).

The ranking for player B in my example is 51st. However, he is not ranked 51st; he's simply tied with all the other guys who are not above the selection criteria. The funny (?) thing, of course, is that he can be ranked lower with a better 3-point accuracy (%) than another player who has the required minimum amount of attempts (or made shots, whichever the criteria is).

Note that I'm not suggesting removing anything from the stats. I'm suggesting removing such lines from the ranking system which mean nothing. I guess another option would be showing "out of" for each ranking. So, 50 out of 51 would mean that the player is the worst ranked player. A ranking of 51 out of 51 (or even 51 out of 50) would in the above example mean that the player is not included in the comparisons.

Last edited by GM-WallyOop at 3/1/2010 7:02:01 AM

This Post:
00
133014.4 in reply to 133014.3
Date: 3/1/2010 11:52:08 AM
Overall Posts Rated:
959959
thats regular behaviour in TIE's. wouldn't expect any other and he is 51st in made 3pointer because there are just 50 better ones.

This Post:
00
133014.5 in reply to 133014.4
Date: 3/1/2010 12:22:08 PM
Overall Posts Rated:
587587
What are you talking about? This is a suggestion, not a question. :-)

This Post:
00
133014.6 in reply to 133014.5
Date: 3/1/2010 12:46:36 PM
Overall Posts Rated:
959959
What are you talking about? This is a suggestion, not a question. :-)


Imho that is removing, because all lines actually means something, so i clarifeid why they mean something:

Note that I'm not suggesting removing anything from the stats

This Post:
00
133014.7 in reply to 133014.6
Date: 3/1/2010 1:10:11 PM
Overall Posts Rated:
587587
Imho that is removing, because all lines actually means something, so i clarifeid why they mean something:

I'm not sure if you understand my suggestion correctly.

I don't want anything removed from stats, period. I want the player ranking (the Supporter feature) to be usable. This can be achieved by either adding information (how many "active spots" are there in the ranking for each stat) or by simply removing useless, misleading information. At the moment the ranking is not clear, because you need to decode it yourself whether a given ranking is an actual ranking or not.

I gave the example of 3-point shooting accuracy above. The ranking for player B is indeed useless information (while his stats obviously are useful information). The ranking suggests that he is 51st in 3-point %. He isn't. It would show 51st for him even if he had made every shot. The player has too few attempts to be considered in the 3-point % ranking. However, we don't know this off hand, we need to figure it out by comparing his stats and his ranking against the stats and rankings of other players. And we don't even see anywhere how many he needs to be considered. If we are lucky, we find the answer by looking at the other players. This situation as a whole is a problem.

The solution is to give the number of active players in the ranking -- or to remove the inactive players (the players who are below a threshold). It would also make sense to show the threshold value. In my opinion, the better solution is to remove the inactive players from the ranking. This is because there are many stats. If only the active lines show, you can quickly glance through all the players and see relevant information. If all the lines must show, there could also be a colour coding to see which lines are active rankings and which are inactive.

Last edited by GM-WallyOop at 3/1/2010 1:11:27 PM