I've worked in the space, and I don't know of any robust method in use that is using Schelling points to solve the Oracle Problem (without the help of any "trusted dispute resolution").
It's done to some degree with prediction markets, but the main problems those generally have are that they predict binary events, which just like binary options are prone to fraud. Even when used for continous ranges (like your temperature example), they generally suffer from semantic "fuzziness" if they are expressed in natural language.
Prediction markets also exist for non-binary events, whether a finite set of outcomes, countably infinite, continuous, or beyond.
Instead, the problem is they still require an oracle for the actual event outcome. That is, everyone is betting on some event, but once it happens, an oracle is needed to tell the market what the actual outcome was. So they don't solve the oracle problem. They're another application that can make use of oracles.
It's done to some degree with prediction markets, but the main problems those generally have are that they predict binary events, which just like binary options are prone to fraud. Even when used for continous ranges (like your temperature example), they generally suffer from semantic "fuzziness" if they are expressed in natural language.