The Best Sets article has been a long standing favorite among my visitors. While it was rather straightforward to come up with the lists when we had full sets of gear, the introduction of all the new unique Legendary gear I items made it more and more difficult, as the possible combinations of gear and their buffs and debuffs grew more and more as well…
To overcome this problem, I have taken it upon me to write a program which does the work for me. Using a database of all the different gear items in-game, the program goes through all possible combinations of gear and produces lists of them with all the necessary information included. With this, I can order those lists to either defense, offense, or total buff+debuff percentages to make it easier to figure out the Best Sets. This also meant I had to change the Best Sets article:
- The top 5 sets have been removed and been replaced with individual gear slot recommendations.
- Older sets from before Replenish and Necromantic have been removed from the lists.
- A print out for the top 30 offensive and defensive sets has been included
- The random 4th attribute of Legendary 1 gear has been omitted. Meaning the given Sets are (still) ‘best’ for most players, but you can do even better with a good random rolled Legendary 1.
- Likewise, other gear with possible good random rolls have been omitted as well, such as specific Void gear, Scorched, etc..
- All combinations have either level VII Artamirs in Ring and Shield for defensive sets, or in Weapon and Amulet for offensive sets.
- Im thinking about a way to make this program available to you all, but not sure yet if and how to do this….
I hope this clears up the changes to the article. Any comments or suggestions can be posted here or at the Best Sets article.
the last buffer disrupts the entire ranking. Some pieces considered attack end up being excellent both in the CHS and in defense…..
what do you mean with the “last buffer” ?
The casual buff
It is clearly explained in the article that only 3 first attributes are considered due to the variability of the fourth one. Of course it will vary the classification but it is your work to check the pieces. It is impossible to take into consideration all the different combinations including fourth attribute
No variation of the fourth attribute. Consider only the two useful attributes:
Defense 36% life or 36% debuff attack.
Attack 36% attack or 36% debuff life.
Due to diminishing returns … looking at the attack table, for instance, couldn’t the line #6 or line #7 set perform better than say line #3 or line #4?
no because the attack power on which the list is sorted takes that into account. It is an objective score on how well the sets perform, so the 3 and 4 set simply perform better than the 6 or 7….
How is the attack/defense power calculated ?
Wow, this is awesome! Good job!
One thing I see missing here, and something that confuses people a lot, is that some of this gear is MUCH better when on the right hero. e.g. Orcrist 1 + Oakenshield 1 + Thorin’s Fur 1 on Thorin is something like +18% better defense than on anyone else. Right now, attacking stuff doesn’t have as much overlap on heroes… we have some good bilbos, bards, tauriel, and elrond stuff… but that’s about it.
Just a thought… it’s almost like a set bonus in other games.
And thank you for including the “attack power” and “defense power” — that’s ultimately the stat I’m looking at maximizing, of course.
Congratulations for a great job
thank you so much! 😉
This is a screenshot of a small part of my Excel file.
Actually I replied to your email including a screenshot that is not appearing 🙂
If you make your algorithm available, don’t do it for free.
It’s a lot of work. I did pretty much the same, using an exel file, compiling all the stats for every piece of equipments but Legendary I.
It calculates all the levels of every pieces of Equipements, summarising the attack/defense %.
I had the same idea as yours in order to select the best sets but could not figure it out an easy way to do it.
Could you just slightly detail how you create your algorithm ?
I would understand that you don’t communicate on it 🙂
Brilliant job as usual!