binding this vs creating an anonymous function

functionCallMicrobenchmark.js

var Benchmark = require('benchmark'),
  suite = new Benchmark.Suite;

var obj = {
  name: 'myobj',
  meth: function() {
    return this.name;
  }
};

var anon = function() {
  obj.meth();
}

var bound = obj.meth.bind(obj);

suite.add('anon-function-in-iter', function() {
  (function() {
    obj.meth();
  })();
})
.add('bind-function-in-iter', function() {
  obj.meth.bind(obj)();
})
suite.add('anon-function-once', function() {
  anon();
})
.add('bind-function-once', function() {
  bound();
})
.on('cycle', function(event) {
  console.log(String(event.target));
})
.on('complete', function() {
  console.log('Fastest is ' + this.filter('fastest').pluck('name'));
})
.run();

package.json

{
  "name": "node-function-speed-tests",
  "version": "1.0.0",
  "description": "",
  "main": "index.js",
  "dependencies": {
    "benchmark": "^1.0.0"
  },
  "devDependencies": {},
  "scripts": {
    "start": "node index.js",
    "test": "echo \"Error: no test specified\" && exit 1"
  },
  "author": "",
  "license": "MIT"
}

Clone with: git clone git@gist.github.com:35e440d2a2e81ca08464.git node-function-speed-tests

My results:

anon-function-in-iter x 28,976,729 ops/sec ±0.68% (100 runs sampled)
bind-function-in-iter x 1,213,125 ops/sec ±0.71% (96 runs sampled)
anon-function-once x 59,548,241 ops/sec ±1.14% (94 runs sampled)
bind-function-once x 6,256,005 ops/sec ±2.37% (87 runs sampled)
Fastest is anon-function-once