New Rating System - Different ratings for different battlers

[thesovereign]

I have been noticing that several weapons that are conditionally good, have been getting downrated because they are terrible at certain levels of the BD.

The prime example of this I would say is ECA currently rated a 6, which for lower level fighting way too low and for upper level rating is way too high. As it is currently shown ECA's rating of a 6 is not very representative as its actual battle value at any level. Separating the rating would allow ECA to have a clear rating of 9-10 for lower level fighting and a 2-3 for higher HP pets.

People also seem to be basing a lot of their ratings on the likely budgets of players with pets at certain levels, however, I do not think that generalizations like this are very good assumptions to make most of the time.

This would also allow for a list of top weapons to be developed for each different class. It just seems that there is something wrong about scarab ring being above SuAP the top weapons list.

The hard part about trying to implement this kind of rating system is to define the different classes of battling. Lets start with the easiest destinction to make:

1p vs 2p

Several raters already make this distinction in their rating of a weapon by splitting the rating into 2 different ratings each worth 5 points. I think that this at the bare minimum should be done, because these two categories actually produce different ratings fairly easily. Take SoS and PCC for instance. In 2p I would consider PCC to be the superior weapon since it gets through the very common GSheild and RoTL, both of which destroy SoS. Now in 1p SoS is likely the better choice because the dark hatting weapons are less common, and if need be you can sell and reequip a different weapon that is not killed whatever the 1p opponent you are facing is using. Since icon variety matters less than quantity in 1p and SoS is ~5 mil cheaper it is the better buy.

From what I can tell ratings for weapons to not fluctuate much based on the stats of your pet, although it could be argued that variable defense weapons are only good for a very short time unless you are around or above the 550 Defense boost. So it seems that 1p probably does not need to be divided into separate catagories.

2p on the other hand, seems to have much variance in weapons usability based on your stats and who you are fighting against. The way I would rate a weapon would differ quite a bit based on whether the opponent is fighting based on boost or league, or how your pet is set up. A tank who is fighting by STR boost would value scarab ring or tower shield much more highly than a evenly trained pet fighting another evenly trained pet. Bracelet of Kings,currently rate 5.2, is another interesting weapon for tanks that is pretty bad for evenly trained pets, but can be killer in a tanks set.

The classifications that I would recommend are as follows:

1p - No subdivisions although I would be open to suggestions
2p - low level fighting - Buyable weapons only, low stat pets
2p - Low Level Fighting - low stats no limit on budget
2p - Intermediate Fighting - pets in the 80s to 250
2p - Fighting by boost (tanks)
2p - Upper Level Fighting - High stats, intermediate to unlimited budgets

If anyone has any comments on the structuring of these catagories, feel free to make any suggestions on how to subdivide the different classes of battling, I am still a bit shaky on exact definitions and unsure if more classifications would be needed or if some of these could be combined.

The main thing that I think NEEDS to happen is a separation of the 1p and 2p ratings, and would like to see some separation of the 2p ratings. Separating these ratings would lead to more uniform rating techniques and give a few borderline weapons clearer ratings that what they are currently given.

EDIT: I am not sure how difficult this would be to implement, but could anyone that would possibly have the power to do this comment on the logistics of actually programming something like this? From my very limited knowledge about this kind of thing I would guess that it would just mean adding a few more columns onto the database, and a few cosmetic things, but if it would be too difficult please comment.

[Shattered Rift]

thesovereign Wrote:EDIT: I am not sure how difficult this would be to implement, but could anyone that would possibly have the power to do this comment on the logistics of actually programming something like this? From my very limited knowledge about this kind of thing I would guess that it would just mean adding a few more columns onto the database, and a few cosmetic things, but if it would be too difficult please comment.

I wasn't going to comment, but reading this portion... once a structure was determined (both for the ratings, and for the appearance), it would be relatively simple (not easy or fast, just simple and doable).

That said, I'm not sure anyone's ever given the number ratings much bearing. Most of the Raters never seem to give a 2 or a 3 to a weapon. It's much more the rationale behind the rating that matters, and at the risk of being insulting, I don't think anyone who takes a number at face value is going to end up very far. A battler should know why they're looking at a weapon, and the explanations behind scores should influence the final decision of the purchase. There are also certainly cases where we buy weapons even though they aren't recommended, simply because of what we need in our sets.

You'd also probably need to break One Player back into the old "low-level, mid-level, high-level" brackets, too. And even then, it doesn't really seem fair to say that a SoS scores a 1 just because it's the bare bottom of the upper-end. (Speaking of which, do we even have a mid-level, weapon-wise, to battling anymore? WoR/BoD is almost all I can think of for One Player ever since the 9-iconers came out.)

[reiksguardknight]

I know little in the realm of databases and coding. As far as your idea, I find it thought provoking and intend to keep it in mind for my future ratings when possible.

[VermontHermit]

I'm finding myself in a fair amount of agreement with Shattered on this.   I think the narrative is generally far more useful than the number ranking.   I guess the number rankings get used to create so called "best of" lists, but I'd never buy or sell a weapon based on what a number grade is.    That said, I think Sovereign raises some fair points, but I don't really see a "simple solution" in that some high level pets are owned by budget restricted owners who have simply put all of their nps into training, and have what would normally be considered vastly inferior battlesets to their stats (or in a more favorable light to those owners, vastly stronger pets than their battlesets would indicate).   I think it should be obvious to anyone that a 40 million np weapon that does four times as many icons of damage as a Scarab Ring is the "better weapon", but I think it's equally obvious that the Scarab Ring represents great value for the nps of its cost.   I think ultimately one has to respect our audience when rating, and presume them to have the intellegence to understand thier situations, and how a weapon might fit into where they are in the game.  The categories that are being discussed aren't really fixed, but overlap, have exceptions and have MANY variables.   If that weren't true, the whole game in rating would be to pick out the best 8-10 weapons in the game and tell everyone to get those.   It wouldn't be inaccurate to say that they'd make a great set, but it wouldn't be very useful to the 95%+ of the players who couldn't afford them.   My own guess is there are only about 200-300 weapons or so that one might reasonably want to use at some point during your career in the battledome.   The rest are items that are either weak, useless, or which have some value, but have alternatives that are better, or cheaper, or both, and shouldn't be used.   I think that if the raters can identify those items that really do have value, explain why they do, and offer some advice on how to unlock that value, we've been of service.   I know I've personally rated the Scarab Ring higher than the T-Tear.   I don't mean to say that the T-Tear is an inferior weapon, and if I found one on the floor, I'd take it and use it.   But I think if you can afford it, it's not a particularly great use of you nps, and that you'd be better off making other choices.   Meanwhile, the Scarab Ring is affordable to virtually everyone, and is of value to any early fighter who trains defense, and can be used situationally well into intermediate fighting for the budget constrained.   I don't necessarily disagree with the goals that have been presented to make ratings more useful, but I'm doubtful that a template approach is going to be any more successful than what we have already.   I do think ratings with good explanations and narratives will get readers on the right track, and that if they continue to be unsure, the IDB forums can get them additional information more specific to their individual needs.

[Spaz]

Suggestion for raters: If you don't already do this, use the highest of the specialized ratings for the final numerical rating that goes in the database. That seems to make more sense than some sort of average. Of course, any battler looking for weapon information should read the actual text of the ratings, not just the numbers.

[thesovereign]

Ya as I was typing it up I was starting to have some of those doubts as well, although it might be worthwhile to have separate comments on 1p vs 2p at the very least, since it seems that some people rate certain 2p harshly, although their rating makes sense from a 1p stand point. I notice a lot of people already split their posts to take this into account.

I could also see the argument saying that for most weapons its not going to make a huge difference.

The main argument for this was to make the top weapons tables, but as I think about it the weapon spotlight list does a pretty good job of doing what I was hoping to accomplish through this idea. Although it seems like it could use a bit of updating. I am curious how sand snowball got on there in the first place.

Edit: From what I have seen looking through tons of ratings it appears the raters split it into categories, like they give x/5 for 1p and y/5 for 2p and their overall rating is (x + y)/10

[CadaverGrey]

Having such categories would be getting too much into technicalities, in which I am not fond of. I would rather see a separate write up about the difference between battling in 1-player and battling in 2-player, and break that write up down into strategies, weapon choices, training etc etc than to go through every single weapon for 1-player or 2 player or any other categories for that matter. Seriously, when a weapon sucks, it sucks. When a weapon is good, it's good.

Numerical ratings means naught. It is a personal choice of the buyer, whether or not this weapon is useful for the price you pay for it. It's the raters job to shed light on certain scenarios and influences that greatly affects that weapon, but in the end, the judgement call of value is up to you as the user

[obiwanpi]

In my opinion, the idea is sort of intruging, but not practical.

First, it would be useless for the almost 90% of BD weapons which are basically gallery items only.

Second, A bit about 1-player. Defensive weapons and dual duties would be pretty hard to gove one ranking for 1-payer for, because all the opponents have different weapons.

Thrid, when people are looking for a weapon upgrade, thier looking to fill an empty nieche in thier set. What might be a good upgrade for someone who has the exact same stats and budget as someone else, and fights the same type of battles as the other person, (1P or 2p) might not be a good upgrade for the other person depending on what's already in thier set.

Finally, getting back to the "nieche" thing. When people go to upgrade thier sets, they don't say "I need an upgrade let's go see what's out there." They say something like "I need to upgrade my offensive weapons, I want the weapons I upgrade to to do atleast x icons, but I can't/ don't really want to pay more than y nps." They enter this into the search feature, and usually, when they eliminate the weapons out of thier budget/ with ratings of 1-3 or so, they have less than 10 choices to choose from, and can fairly easily read the naratives for all of those weapons, thus having basically the same effect as having different ratings for different types of fighters, but more specific, and easier to do, I'd imagine, than having 5+ different narratives/ ratings. And as cavader said "in the end, the judgement call of value is up to you as the user."

So interesting idea, but I don't think it'd be practical.

[rabid_schnauzer]

Well as part of my preparations for the recent rater applications I worked out a 10 point rating scale, based off of sirhatter's old system. The idea was to try and fit those various nuances into the existing 10 point scale.

If I am selected as a rater (crosses fingers) I will post a detailed account of the methodology but the general idea would be to rate weapons by Power (SuAP cleans up here), Price (Scarab Ring wins this one), Tactical/Niche (ECA is very strong here) and a 1 point non-battle/galley/investment category. Thus for a weapon to get a very high rating it would have to do well in all categories. Many newer battlers DO use the "top-rated" list as a guideline (I know I did) while they are still learning about the BD, so I think that's it's important to have only good weapons with relatively broad applications on it as well as clear commentary from the raters on the individual weapons that make that list.

[theonlysaneone]

I like Spaz's idea of rating it for each aspect of the battledome, or at least discussing them, then giving the final rating as the highest of those. The real purpose of the numerical rating is to position them on the Top 50 weapons, but even those aren't good for some people. For instance, Cloudy Wand of Storms is great for 2-player, but not so much for 1-player. It still deserves a place on the Top 50.

[rylix]

In my humble oppinion i think it's the raters who have to adjust (not that you guys are doing a bad job, i find many posts make me think about aspects of a weapon i have never thought of). If a weapon would be primarily suited to 2-player than it should receive the rating it would get as if there was only 2-player battling. Just because a weapon isn't useful in both 1-player AND 2-player (x AND y) doesn't mean it deserves less of a rating (as i said, my own oppinion)