Apparently although the statistical averages are pretty useless I'm supposed to leave them alone. I don't mind that too much as they are small and actually look nice. However, the additional type chart is going a little too far. The information was already there, in a format that looked better, conveyed the information better, didn't take up unnecessary space, and didn't stretch the screen in a resolution smaller than 1024x768. Is there any problem with the format evkl established?
Another thing, we need to decide on either a numerical or alphabetical listing of the Pokemon. I'm partial to numerical personally - among other things, it groups most evolution chains nicely.
I was going to add the average stats to all the other type pages, but my calculations don't turn up the same for Grass as what's reported on this article. I get these:
HP: 63 ATK: 68 DEF: 64 SPD: 56 S.ATK: 74 S.DEF: 68
Were the originals calculated manually? I made a simple routine to do it automatically that's sound but I want to make sure with the original poster. Sheep 01:13, 17 Feb 2005 (GMT)
The dude who posted the stat totals (whose name I don't remember well enough to spell) hadn't included dual-types into the calculations yet. Feel free to include a more accurate version, but keep the current appearance. IMO it'd be a good idea to discount pre-evos as well.
But with respect to pre-evos, for instance, I don't think we should consider Scyther a pre-evo of Scizor for the Bug-types, or Onix and Steelix Rock-types.
Although that sets up an interesting "when do we draw the line" question, but I think we can figure that out for ourselves.
It seems to me like the statistical average should include all those pokemon listed on the page, otherwise it would be misleading without a note under each statistical average report that lists all the pokemon counted. People still use pre-evolutions anyway, if they're playing the game and they havn't evolved yet. Sheep 16:08, 19 Feb 2005 (UTC)
So have a note.
What value does including pre-evolutions have to someone just playing the game? I never said they weren't used, but they aren't very relevant to the stat averages. Not that the stat averages are relevant in the first place.
It doesn't have any value at all to anyone playing the game, you're right. It's simply more encyclopedic. I think we need to be careful about turning an encyclopedia into a more game-guide oriented thing. Not that this is that big of an issue, but accuracy and noteworthiness need to be taken together. The problem with having a note is that it may pad or confuse the point of stat averages - in an article about a type, isn't every pokemon of that type noteworthy? Sheep 23:07, 19 Feb 2005 (UTC)
Why exactly is it any good to be "encylopedic" if the information isn't very useful? Not including pre-evolutions would simply make the stat averages more relevant. Adding a few extra words is a pretty tiny price to pay. Jshadias 07:56, 25 Feb 2005 (UTC)
I did the averages so that it only considers pure type Pokemon for each respective type. Of course you get some strange sets for the Rock and Ice type because they are under-represented. Throwing out pre-evolutions and keeping last-stage evolutions doesn't matter - the difference between the stats is what appears in the averages so Squirtle and Blastoise still have the same statistical percentages from the normal but Squirtles is a percentage of the pre-evolution set and Blastoise of the final-stage set of regular stats. You can include all pokemon listed on the page in your calculation but you have to add some weighting - I've written the program to approximate a Pokemon's stats without a certain type influence so making Bulbasaur a pure Grass-type rather than dual Grass- and Poison- type, but the process of making real averages would take many repeats. It can be done and I can put up the source code for anyone who wants to improve it. - Ainohuyuah
What the hell's going on with the duplicate "Super/not very" effective thing here? No reason to have doubles of it. Evkl 02:54, 25 Feb 2005 (UTC)