I was asked to make a post about this now that some more misleading and poorly contextualized information about white bag drop rates is making the rounds. I lost interest in maintaining my website (pentaract dot com) that listed old drop rate info with context, so I’ll repost that info here alongside a more thorough overview of how loot works in Realm of the Mad God.
How do drop rates work?
This is a widely misunderstood subject, and I’ll probably need to make a wiki page about this later. But it’s a simple system that has changed very little since soulbound loot was introduced in 2011, so I can quote Linkshot/Chispy from this ancient thread:
To get more technical, every enemy with a drop table has its loot specified in the XML like so:
<SoulboundLoot prob=".01" threshold=".15" min="1" max="5">Item Name Here</SoulboundLoot>
- Items dropping in public bags use the
<Loot>tag instead of<SoulboundLoot>, and don’t use thresholds, but the calculation is otherwise the same.- Loot bags contain the items in the order they are listed in the drop table (a potion bag from O2 can contain att/wis/vit/def in only that order).
- You can see what Linkshot meant by “more people shooting = more loot potentially dropped” in dungeons like Magic Woods and Haunted Cemetery, whose bosses are generous with public loot. With large groups, it’s hard to see your soulbound loot amid all the junk in pink bags.
- Two bags can be generated if there are more than 8 items. This was added after some unfortunate incidents where players got full white bags that didn’t contain the UTs.
- Chests found in setpieces within the realm have loot determined in a similar way, I assume when the map is generated.
- The value specified in
probrepresents the item drop rate. An item with a probability of 1 (like a mark) is guaranteed to drop, .01 has a 1% drop rate, and so on. The value is used by a random number generator as loot is parceled out, giving you that chance of getting the item every time.- All drops are rolled independently of each other. If you qualify for an item with a 1% drop rate 100 times, you are not guaranteed the item. (Probability of getting an item with a 1% drop rate within 100 runs: 1 - .99100 ≈ .634, or 63.4%)
- Presumably, loot drop potions multiply the probability by 1.5 before the calculation takes place.
- Before the loot is rolled at all, you need to hit the
thresholdif it is specified. The damage threshold is a percentage of the enemy’s hit points. If the enemy’s health is 1000 and the threshold for an item is .1, you need to do at least 100 damage to qualify.- Different items can have different soulbound thresholds.
- Low-damage classes like priest used to be derided for their inability to consistently hit soulbound damage. This was changed when thresholds were lowered across the board (more on that later).
- There is no such thing as “overdamaging”. Once you hit the soulbound threshold for the item, your chance of getting it is the same no matter what (loot potions notwithstanding).
- With enemies that HP scale, I think the threshold is based on the enemy’s health at the moment of its death.
- The
minvalue indicates the minimum number of the specified item that will be distributed among all the players who did damage. It is possible for more players to get the item, but at least that many players (if present) are guaranteed to get it. This is usually used for stat pots.- Sometimes a specific type of drop, rather than an individual item, is guaranteed (Mighty Quest Chests guarantee a life OR a mana).
- The
maxvalue indicates the maximum number of the specified item that will drop among the players. - Most tiered items are represented in the drop tables by another parameter,
tier. This allows drop tables to include every weapon/special/armor/ring within a particular tier, without listing each item.- Loot tier potions used to work by rolling to increase the tier by 0-3, which allowed you to get gear well above the highest tier that normally dropped (like old tops from godlands). This was changed so that they now respect the maximum tier in the enemy’s loot table.
- Lost Halls tops (and presumably O3 tops) are listed individually in the loot tables, so loot tier potions cannot boost items’ tier past T12/T6/T13/T6 (making them pretty much useless).
Why are drop rates secret?
Loot tables were never officially released by either Wild Shadow, Kabam, or Deca. In the early days of the game, they could be determined by decompiling the client, so forumers would post the drop rates to the wiki. This changed in February 2012, when the Wild Shadow devs obfuscated the client to hide the drop rates and asked wiki editors to stop displaying them.
Though the game has since changed hands twice, the rationale has remained the same. Krathan summed it up thusly in the Deca AMA:
This remains the official stance as of April 2020.
Were drop rates really leaked?
Yes, eight years ago, by Skoad – one of the earliest players, who was tapped by Wild Shadow to design the Tomb and started on the LoD under Kabam. I don’t know the details of how or why the leak occurred, but a folder containing unobfuscated files used by the RotMG server (and a PDF of There Will Come Soft Rains) was released. This leak is the source of the drop rates on this page, and also served as the foundation for RotMG private servers.
The leaked files came from Release 7.0.0. I won’t be linking to the actual leak, but the UT drop rates from that build are as follows:
| Item | Drop Rate | SB Threshold |
|---|---|---|
| Dirk of Cronus | 0.1% | 7.5% |
| Helm of the Juggernaut | 0.2% (Sphinx) 0.026% (each of 8 Hermit drop locations) |
7.5% (Sphinx) |
| Orb of Conflict | 0.1% | 7.5% |
| Seal of Blasphemous Prayer | 0.1% | 7.5% |
| Shield of Ogmur | 0.2% | 7.5% |
| Crystal Sword | 4% | 3% |
| Crystal Wand | 1% | 3% |
| Demon Blade | 0.15% (Abyss Idol) 0.1% (Malphas) |
15% |
| Candy-Coated Armor | 0.5% (MegaRototo) 0.1% (other bosses) |
20% |
| Ring Pop (now Candy Ring) | 2% (MegaRototo) 1.5% (other bosses) 0.15% (Gumball Machine) |
18% |
| Staff of the Crystal Serpent | 8% | 17% |
| Cracked Crystal Skull | 8% | 14% |
| Robe of the Tlatoani | 8% | 11% |
| Crystal Bone Ring | 8% | 10% |
| Spirit Dagger | 0.25% | None |
| Ghostly Prism | 0.08% | None |
| Spectral Cloth Armor | 4.5% | None |
| Captain’s Ring | 4% | None |
| Amulet of Resurrection | 2% | 8% |
| Plague Poison | 0.1% | 16% |
| Resurrected Warrior’s Armor | 0.25% | 18% |
| Conducting Wand | 0.3% | 15% |
| Scepter of Fulmination | 0.1% | 14% |
| Robe of the Mad Scientist | 0.2% | 11% |
| Experimental Ring | 0.5% | 15% |
| St. Abraham’s Wand | 5% | 15% |
| Tome of Purification | 0.1% | 15% |
| Chasuble of Holy Light | 8% | 11% |
| Ring of Divine Faith | 10% | 10% |
| Coral Bow | 10% (Coral Gift) 0.45% (Thessal) |
45% (Coral Gift) 15% (Thessal) |
| Coral Venom Trap | 5% (Coral Gift) 4% (Thessal) |
45% (Coral Gift) 10% (Thessal) |
| Coral Silk Armor | 5% (Coral Gift) 4% (Thessal) |
45% (Coral Gift) 8% (Thessal) |
| Coral Ring | 5% (Coral Gift) 4.1% (Thessal) |
45% (Coral Gift) 4.5% (Thessal) |
| Ancient Stone Sword | 0.3% | 15% |
| Anatis Staff | 5% | 1% |
| Chicken Leg of Doom | 5% | 1% |
| Wand of the Bulwark | 0.1% | 15% |
| Snake Skin Shield | 10% | 10% |
| Snake Skin Armor | 10% (Stheno) 2% (Stheno Pet) |
1% (Stheno) None (Stheno Pet) |
| Snake Eye Ring | 10% | 10% |
| Sprite Wand | 2% (Limon) 1.5% (Native Nature Sprite) |
18% (Limon) None (Native Nature Sprite) |
| Cloak of the Planewalker | 0.1% | 15% |
| Staff of Extreme Prejudice | 0.1% | 15% |
| Ring of the Nile | 0.7% | 4% |
| Ring of the Pyramid | 0.7% | 4% |
| Ring of the Sphinx | 0.7% | 4% |
| Tome of Holy Protection | 0.3% | 3% |
| Doom Bow | 0.1% | 15% |
| Wine Cellar Incantation | 1.5% (Thessal) 1% (Dr. Terrible)* 0.7% (tomb bosses) 0.1% (everywhere else) |
18% or none |
*My pet theory is that someone made a typo in Dr. Terrible’s drop table, which led to incantations becoming much more common
“But wait,” you might be saying, “I got a bunch of X item last event – how can the drop rate be so low?” Well…
Are these rates still accurate?
No, not even close.
In late 2012, UT items were tradeable and duping was rampant, so items were widely disseminated even with the sometimes-astronomically low drop rates you see above.
The dungeon drop rates listed became outdated weeks after the leak, when they were buffed. Subsequent updates in the years since UTs became soulbound have changed drop rates across the board. Damage thresholds, too – some whites, like A.S.S. and Davy’s loot, were difficult to obtain because the thresholds were too high to reach in crowded fights. Now it seems that just a few shots can qualify you for a white or orange bag pretty much anywhere.
While certain individual drop tables may have gone unchanged in the eight years since the leak (for instance, low-tiered gear from realm and dungeon enemies probably drops at the same or similar rates), the UT rates are likely to differ widely.
What do we actually know?
Since Deca seems unlikely to release drop tables anytime soon, we are left with guesswork, data collection, and the snippets of information released by the developers. (Apparently, saving a dragon for later in LoD gives it a better drop rate. Oryx 3 will be generous with new tops, guaranteeing a weapon, armor, ability, and T6 ring to one person each run – solo and you get all four.)
The overall trend seems to be toward making most drops more common. Older ST sets drop more commonly after new ones are introduced. Earlier this year, LH tops and dozens of UTs had their drop rates increased. This seems to be part of a larger design by Deca in which dying becomes more common (O3, IC/OOC) but also more forgiving (higher drop rates, UT exchange).
Several players have embarked on large-scale drop recording projects, and while those are at risk of being marred by poor data collection practices or invalidated by further drop rate changes, they can be useful as long as you don’t take them as gospel.
Another option is to just guess like everyone else. Remember… anytime someone tells you what a drop rate is, unless they work for Deca, take it with a grain of salt. And if they do work for Deca, they’re messing with you.
