Warcraftlogs ilvl% over overall% which one is better?

[Deleted]

Hello there everyone,

Me and my guild have been discussing warcraftlogs a lot lately (Since we are currently reforming our mythic group).

For us being mechanically good is a big thing and it's mostly what we look for.
Havings said that dps is also quite important.

So looking at dps and ratings. Should we look at ilvl% or overall rating for dps/healing/tanks, and why?

[Deleted]

Overall.

Someone could have such a discrepancy in 2 different slots in gear but still sim as high as someone with 960+ pieces in every slot. Certain specs prefer certain relics/trinkets over ilvl not to mention having a low ilvl in a specific slot may not impact your damage output as greatly as other slots.

As well as it's too easy for anyone 970+ to cheese any lower ilvl specific percentiles.

[potis]

You look at both.

First you look at ilvl to understand how you did, there is a massive difference between current gear level and how overall changes and actual ilvl comparison.

As example on my Rogue alt, i did an overall 20%, thats because my log was at end of February, in 930 ilvl gear, when everyone was like my main at 970+, but his ilvl log was averagely 96%, which is what matters, that i did actually play very well, i just do not have the gear to be competitive.

Reading warcraftlogs is easy, but people generally tend to not understand what they read and things get complicated.

[McFuu]

ilvl matters, overall not so much.

ilvl shows the player understands the spec, the rotation, the fight (in most instances), and have geared well. Overall is just a measure of all parses and how that performance stacks up. Obviously higher ilvl will relate to overall, but ilvl means more to the ability of the player.

[Twdft]

It depends on who wanna campare yourself to. overall% means everyone who killed the boss, which isn't really useful to me as a HC only raider because overall% includes max-ilvl mythic raiders who roflcopter through the raid on their off night. So I rather look how I do against everyone in my ilvl bracket.

[Critical92]

You have to look at both. Overall means everyone and itemlvl means everyone in your bracket.
When you have only a few people in your bracket you can not really tell if that person did good or bad.

Lets take assa rogue as an example. You know you played well, did mechanics and your raid did overall good but your poison bomb damage is under 4%. That means you will be placed lower in your itemlvl bracket because of bad rng and not because you made mistakes.

You also have to look how other raids played the boss. Did they play normal or used a cheat strat to boost rankings of someone.

You can do an overall 99% but a low 33% in bracket and often you could not control that with how rng some specs are.

As a healer you will do a bad log when you do not have to heal. I had farm raids in tos where one heal solo healed with 2 other as backup to chease logs.

[Deleted]

ilvl matters, overall not so much.

ilvl shows the player understands the spec, the rotation, the fight (in most instances), and have geared well. Overall is just a measure of all parses and how that performance stacks up. Obviously higher ilvl will relate to overall, but ilvl means more to the ability of the player.

Ilvl percentiles are more inconsistent than overall especially 7 months into a tier. As at low ilvl bracket (under 960ish) you're generally only competing with alts or generally inactive players. You also have to factor that a much larger portion of the low ilvl brackets are not done in organized raiding groups and/or are done by much lesser skilled players. Overall factors in every single public parse ever made, so it takes into consideration the alts/lower skilled players AND the higher skilled players.

And the 960s range is dominated by mythic ilvl players save a few pieces dragging their ilvl down due to itemization or bad RNG but generally if you have a high percentile is the upper 960s range your overall will be decently high too.

[Nitros14]

Overall since ilvl is a smaller sample size and cuts out the best players if you're low. But you have to take into consideration a ton of factors like itemization (not all ilvls are the same), mechanics dealt with, luck, kill time (faster kills are usually much easier to parse well with, a good player with a bunch of bad dps will have a hard time getting 99+ percentile) etc if you want to form a true picture of how well someone is playing.

There's exceptions to the kill time rule for specs, a kill at 2:50 with no second set of cooldowns can be harder to parse with than a 3:30 kill where you got a full second set of cooldowns.

Most people will look at ilvl, of course, since it usually makes them look better. Who doesn't want to think they're better than they are?

[Twdft]

Overall since ilvl is a smaller sample size and cuts out the best players if you're low.

Players with higher ilvl should naturally do more damage than you, why would you want to compare yourself to them? But then again ilvl% bracket also excludes players with lower ilvl than you who still beat the same bosses. The thing is that you have no easy way to see how many of both are doing the same damage as you.

So I still think it's better to see how you compare with players who have the same ilvl as you, even if the sample size is smaller.

[clevin]

Part of the problem is that the ilelve brackets are so narrow. If someone is, say, 960, would be nice if you could see how they compared to everyone from 955-965 or even better if you could tune the bracket with. Similarly, the longest time period is 2 weeks so if you're looking at a narrow bracket over just that time, you can only get so big of a sample.


All that said, it's still close to useless to use overall since DPS scales so much with gear. Yes, your 940 toon sucks compared to a 976 toon. But what does that really tell you? If the reply is to look at the applicant's detailed logs... well you should do that anyway. Looking at level brackets vs overall should just be a first screen, which is why I'd use level brackets. The question you're asking (or should be asking) is "Is this person even worth looking at or are they bad?" That is, you want to weed out the obviously bad applicants so you spend your time wisely. It's like any competitive application process... if you're a good college, you don't bother investigating the D students much. You want to spend your limited time on the B+ and A students and may the B students who show promise. Same here...

[Deleted]

Ilvl really does not matter very much unless your overall percentage is also very high. When you have an undergeared char carried in a heroic raid, then you may get good ilvl logs. That however is not really representative of your performance as a player. There are probably very few good players in that ilvl range, so if you get orange ilvl logs, congratulations but it does not mean very much.

Overall only matters when you have mythic gear as reasonably you will never achieve high overall numbers without mythic gear. However, overall is a better performance indicator when you do have access to pretty much all of the gear - because now your job as a mythic dps is to be the best at whatever class you are playing and to bring out the highest dps of your class/spec, period, not the highest in your ilvl bracket.

[McFuu]

Ilvl percentiles are more inconsistent than overall especially 7 months into a tier. As at low ilvl bracket (under 960ish) you're generally only competing with alts or generally inactive players. You also have to factor that a much larger portion of the low ilvl brackets are not done in organized raiding groups and/or are done by much lesser skilled players. Overall factors in every single public parse ever made, so it takes into consideration the alts/lower skilled players AND the higher skilled players.

And the 960s range is dominated by mythic ilvl players save a few pieces dragging their ilvl down due to itemization or bad RNG but generally if you have a high percentile is the upper 960s range your overall will be decently high too.

Overall is a combination of all the parses, so your reasoning for not wanting to use ilvl applies to Overall just the same.

The only aspect that hurts ilvl is mythic+. Mythic+ players geared in non tier gear drag down the parses within their ilvl range because they can't compete with bonuses.

But this would apply to overall as well.

[Low Hanging Fruit]

I look at item level tier to determine the individual ability of the player. I look at the over all percentage to determine if they fit into our tier. Then I weight these two things together and discuss it with others to come to the conclusion if their output is worth our time. After that of course you make sure that these numbers are not achieved by cheeses and if mechanics are being followed. Do they use defenses and do they ever do more than just their base role.

[Deleted]

Overall is a combination of all the parses, so your reasoning for not wanting to use ilvl applies to Overall just the same.

The only aspect that hurts ilvl is mythic+. Mythic+ players geared in non tier gear drag down the parses within their ilvl range because they can't compete with bonuses.

But this would apply to overall as well.

Low ilvl brackets means little the further your boss progression goes. Such as the ilvl bracket for a 940 toon on heroic argus will be distorted with carry/ boost runs and actually trying would guarantee you an okay ilvl parse even if you didn't do very good performance wise.

Another example is that 950 is more than enough to kill mythic garothi however the amount of 950 players currently killing mythic garothi is almost completely consisted of boosting runs or parses from 6+ months ago so again, ilvl percentile doesn't mean much here as it doesn't really tell you anything.

And like a previous poster said, there are few skilled players playing in low ilvl brackets as well. Ilvl percentile just loses it's value the longer the tier goes on for.

[Deleted]

Just wear a 1ilvl transmog piece on one of your slots and guaranteed 100% ilvl parses yahoo

[Atherions]

The best players also generally have the best gear so overall is the only competitive metric. As others have stated the ilvl brackets are heavily biased, esp if you start going really low.

[Deleted]

ilvl matters, overall not so much.

ilvl shows the player understands the spec, the rotation, the fight (in most instances), and have geared well. Overall is just a measure of all parses and how that performance stacks up. Obviously higher ilvl will relate to overall, but ilvl means more to the ability of the player.

Not true. Having an arcano and proper traits may lead to e.g. 5 ilvls less for the same dps. Just like preferringvthe 985 ring with wrong stats over the 950 with bis stats bit only 2k dps difference.

Additionally on high ilvls, skill matters a lot. There are e.g. nearly no bad players in the 980+ bracket but a lot in 950 bracjet.

So on average i might be a 99% player in the 950 bracket, 92% in 960 bracket, 81% in the 975 bracket and 60% in the 980 bracket. Because the 980 bracket is filled with top players while the 950 bracket is full of casuals and badly or medium played alts....

[Sencha]

A mix of both.

If you only look at ilvl, a titanforged gem slot would be too desirable. And people will only want the perfect item per ilvl.

[Deleted]

I look at ilvl on my toons, as comparing to overall is not really good for a char with 940ilvl. As for main I look mainlu for overall, as ilvl includes some cheesing, sockets and bis stats lower ilvl items, with insane tf ilvl stuff with shit stats. Ilvl could be ok, if main stat was most powerful of all, but there very little cases like it in legion.

[amaze123]

Ofc you have too look at both. But watching overall% you have to consider the difference in ilvl and the amount of cheese used. If you look at ilvl% you have to consider that you only compete with toons, heroic raiders and overall much worse players. It is much easier to parse orange in the 960 bracket than in the 970.