Some thoughts:

I think the net result would be slow decay of confidence in all feeds unless some compensating control is considered.

Compensation control can come from potential loss of job - similar to block production in DPoS Witnesses.

Others thoughts for thought:

Oracles data may need to overlap at some high threshhold, such as 75%, for the data to become consensus over allowing rewards distributions.

Potential Oracles need to compete with each other for jobs and exposure of poor performance is in the challengers' interests.

Consensus over permitting rewards distributions may come from anywhere between one and thirty elected Oracles. The number of Oracles is defined by the SMT creator.

Oracles as they affect SMTs Rewards Pools are only here to answer a simple question: Did the entity behind this account perform action X, or X+Y, or X+Y+N? If yes, publish the account on the Oracle's feed.

Oracles as they affect SMTs Rewards Pools are only here to answer a simple question: Did the entity behind this account perform action X, or X+Y, or X+Y+N? If yes, publish the account on the Oracle's feed.

Some Oracle systems will fail, some will succeed. 90 to 10.