In his response he sets out a process that one must follow:
Start out with a Definition of Security, which apparently is the Foundation upon which all consensus protocols ought to be judged in our space. It defines what a secure Ledger is.
Next you need to prove that Proof of Work satisfies this so you have a benchmark.
Then you need to prove proof of stake satisfies this.
Proof of Work is not a Good Benchmark
According to Charles, the proof of work satisfies his definition of secure ledgers. In my opinion, proof of work provides terrible security based upon real world centralization in mining pools, misalignment of incentives, successful double spends, empty blocks, selfish mining, fee extortion, etc. Following the process laid out by Charles is an example of asking the wrong question: "How do we create a proof of stake algorithm with security properties of proof of work?". It will produce a proof-of-stake algorithm that fails to account for the problems present in proof of work because those problems are completely outside the scope of his definition of security.