A quick and dirty look at the overhead of calling a double-dot method.

doubleDotPerf.js

const testCount = 1e7;
const keyCount = 10;

const times = count => Array.from(Array(count))
const message = (test, runtime) => `${test} took ${((runtime[0] * 1e9) + runtime[1]) / 1e6}ms`;

const obj = times(keyCount)
  .map(Math.random)
  .reduce((o, p) => Object.assign({}, o, { [p.toString()]: Math.random() }), {});
const hasKey = Object.keys(obj)[0];
const hasNotKey = 'missing';

const cached = process.hrtime();
const has = Object.prototype.hasOwnProperty;
times(testCount).map(() => {
  const one = has.call(obj, hasKey) === true;
  const two = has.call(obj, hasNotKey) === true;
  return one && two;  
});
console.log(message('Cached', process.hrtime(cached)));

const nested = process.hrtime();
times(testCount).map(() => {
  const one = Object.prototype.hasOwnProperty.call(obj, hasKey) === true;
  const two = Object.prototype.hasOwnProperty.call(obj, hasNotKey) === true;
  return one && two;
});
console.log(message('Nested', process.hrtime(nested)));