Valve publishes loot box odds for Dota 2

Chances of receiving a rare item from a bundle now disclosed in game, with escalating odds calculated

As a part of a Dota 2 update adding various new bundles to the game, Valve has also updated the game's loot box reward system of treasures to show odds of acquiring items of different rarities ahead of purchase.

"We've also taken this opportunity to simplify and rework the way we calculate escalating odds for this treasure and going forward," reads the official blog post. "You can now click on the escalating odds arrow next to each of the rare, very rare, or extremely rare drops to see the exact odds of receiving them based on how many you've already opened."

Included in the update is a change and simplification to how these loot box odds work, though the post does not specify the exact nature of the change. Dota 2's boxes increase odds of receiving a rare item based on number of bundles opened, offering a sort of "bad luck protection" to those who purchase frequently and consistently do not get rarer rewards.

Disclosing loot box reward odds has often been suggested as a possible solution in conversations with a number of governments seeking to regulate loot boxes as gambling, and several Asian countries already require the disclosure. Some companies have already opted to disclose odds internationally - Psyonix disclosed odds on Rocket League loot boxes earlier this year, and Apple has required games in its App Store with such transactions to disclose odds since the end of last year.

Earlier this year, Valve gave Dutch Dota 2 players the ability to see the contents of a loot box ahead of purchase in order to comply with Netherlands gambling laws.

Related stories

Indies on Steam are betting on discoverability

After a year of missteps by Valve, indie devs are cautious about Steam Labs and their future visibility on Steam

By Rebekah Valentine

Steam launches Steam Labs to house feature experiments

First three test projects all involve new ways to discover games, including a machine learning recommendation feature

By Rebekah Valentine

Latest comments

Sign in to contribute

Need an account? Register now.