This question is based on a quick experiment that I ran when questioning the effectiveness of my RFID blocking wallet (this one). In my experiment, I walked out of my office, placed my RFID id card in the wallet, and attempted to use the wallet containing the card to open the office door. To my surprise, the door opened with no issue!
My question is this: does the opening of the door indicate that this RFID blocking wallet is not actually doing it's job? I physically tap the wallet against the reader in order to open the door, so does the wallet just reduce the effective range at which the card can be read? Without the wallet, the card also needs close proximity (within a few cm) to the reader to open the door.
Edit: reading this question's accepted answer, it is mentioned that RFID frequency can vary- would it be reasonable to assume that the difference in operating frequency between a credit card and ID badge would be large enough to cause a credit card but not ID badge to be obfuscated by the wallet?
Edit #2: I went back to try some more controlled experiments based on the answer by @Herringbone Cat and comments from @AdamShostack below. Varying the orientation of the card does not appear to make any change. HOWEVER, I tested for distance at which the card could open the door, and found it effectively halved by the use of the wallet.
In terms of frequency vs. power, I would not expect power to be the issue here, as this is a passive card (should have specified earlier). I would expect a malicious RFID reader to be of arbitrary power, and therefor comparable to the door reader at a minimum. Investigating further, I suspect that my ID card is this brand, which has an operating frequency of 13.56MHz. This is the operating frequency for credit cards as well.
That said, halving the range a reader must be to the card may be an effective deterrent for RFID card theft.